From Hicham.ElHarti at dekabank.de Mon Oct 1 16:05:30 2007 From: Hicham.ElHarti at dekabank.de (Hicham.ElHarti at dekabank.de) Date: Mon, 1 Oct 2007 16:05:30 +0200 Subject: [rrd-users] Import old data to rrdtool Message-ID: <8D78FA5058F34E419A26073041547D8306650468@DKDEM103.dekager.dekabank.intern> Hello, I have some text files with old monitoring data, which should be insered to rrdtool DB and showed with graphics. The files have this form: (timestamp art_avg art_min art_max) =============================== 1189585215 0 0 2 1189585515 0 0 3 1189585815 0 0 3 1189586115 5 0 6 1189586415 0 0 2 1189586715 1 1 3 1189587015 7 1 14 1189587315 0 0 2 1189587615 0 0 3 1189587915 0 0 2 1189588215 1 1 4 1189588515 0 0 3 1189588815 0 0 2 1189589115 0 0 1 1189589415 0 0 21 ... ================================= I created a DB for these data using: rrdtool create test.rrd --start 1188597600\ -s 300 DS:art_avg:GAUGE:180:U:U \ DS:art_min:GAUGE:180:U:U \ DS:art_max:GAUGE:180:U:U \ RRA:AVERAGE:0.5:1:8640 \ RRA:AVERAGE:0.5:1:8640 \ RRA:AVERAGE:0.5:1:8640 ======================================== I then inserted the data in form: Timestamp:art_avg:art_min:art_max with the command rrdtool update I tried to get the graph with: (for month september 2007) /usr/bin/rrdtool graph test.png \ -s "00:00 09/01/07" -e "23:59 09/30/07" \ -v "ZGS Average Response Time" -w 1000 -h 450 -t " ZGS AJA " \ DEF:art_avg=test.rrd:art_avg:AVERAGE \ DEF:art_min=test.rrd:art_min:AVERAGE \ DEF:art_max=test.rrd:art_max:AVERAGE \ COMMENT:" " \ LINE1:art_avg#00FFFF:"Average Response Time" \ LINE1:art_min#FF00FF:"Min Response Time" \ LINE1:art_min#00FF00:"Max Response Time" \ COMMENT:" \j" But the graph is empty!!!! When I run rrdtool fetch test.rrd AVERAGE I get several line like: ================================= 1191242100: nan nan nan 1191242400: nan nan nan 1191242700: nan nan nan 1191243000: nan nan nan 1191243300: nan nan nan 1191243600: nan nan nan 1191243900: nan nan nan 1191244200: nan nan nan 1191244500: nan nan nan 1191244800: nan nan nan 1191245100: nan nan nan 1191245400: nan nan nan 1191245700: nan nan nan 1191246000: nan nan nan ================================== Can you tell me please why the graphs are empty Thank you in advance! Cheers, Hicham -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071001/d9ffb2e6/attachment.html From emiller at imvu.com Mon Oct 1 20:25:40 2007 From: emiller at imvu.com (Evan Miller) Date: Mon, 01 Oct 2007 11:25:40 -0700 Subject: [rrd-users] does rrd aberrant behaviuor detection really work ? In-Reply-To: <47920.10.0.1.56.1191188045.squirrel@laposte> References: <50107.10.0.1.56.1191072554.squirrel@laposte> <47920.10.0.1.56.1191188045.squirrel@laposte> Message-ID: The beta version of RRDtool does fix at least one known bug in the update logic, although I'm not sure it's the one that's affecting you. See this thread for more info: https://lists.oetiker.ch/pipermail/rrd-developers/2007-August/002055.html It's possible this bug was introduced between 1.2.15 and 1.2.19, but I'm not sure. It's also possible the bug you're seeing was fixed as part of more general code cleanup in the update logic. Anyway, glad it works in the beta version, at least. Evan sylvain at meduse.homelinux.net wrote: > Hello ! > > I just tried the last beta1 version (compiled whith ./configure && make) > and aberrant behviour detection works. The 5 aberranr behaviour detection > RRAs are update . > With the previous version of rrdtool (1.2.23), compiled the same way, it > doesn't work. > > Could it be considered as a bug, or did i missed something ? > > regards, > > sylvain > >> Hello ! >> >> I'm trying to use abd functionality with my rrd files, but without success >> for now. >> I create a rrdfile like this : >> rrdtool create rrd/nfs3-ip-in-00h51-dup.rrd >> -b 1184885390 -s 300 >> DS:ipInReceives:COUNTER:600:0.0000000000e+00:9.2233720369e+17 >> RRA:AVERAGE:5.0000000000e-01:6:700 RRA:AVERAGE:5.0000000000e-01:1:8640 >> RRA:AVERAGE:5.0000000000e-01:24:775 RRA:AVERAGE:5.0000000000e-01:288:797 >> RRA:MIN:5.0000000000e-01:6:700 RRA:MIN:5.0000000000e-01:1:8640 >> RRA:MIN:5.0000000000e-01:24:775 RRA:MIN:5.0000000000e-01:288:797 >> RRA:MAX:5.0000000000e-01:6:700 RRA:MAX:5.0000000000e-01:1:8640 >> RRA:MAX:5.0000000000e-01:24:775 RRA:MAX:5.0000000000e-01:288:797 >> RRA:LAST:5.0000000000e-01:6:700 RRA:LAST:5.0000000000e-01:1:8640 >> RRA:LAST:5.0000000000e-01:24:775 RRA:LAST:5.0000000000e-01:288:797 >> RRA:HWPREDICT:2016:0.1:0.0035:288 >> >> i have had a look at the output of : >> * "rrdtool dump rrd/nfs3-ip-in-00h51-dup.rrd" and >> * "rrdtool info rrd/nfs3-ip-in-00h51-dup.rrd" >> and everything seems to be ok >> >> i then use "rrdtool update ..." to fill in the rrd file with 2 weeks of >> data , but the 5 abd related RRA stay leer (only NaN). >> I tried with the last version of rrdtool downloaded from www.rrdtool.org >> I also tried with the standard kubuntu package (RRDtool 1.2.19) and i got >> the same result. >> The only way i got it working, is by using the standard (stable) debian >> rrdtool package (RRDtool 1.2.15) >> >> Can someone help me with this ? >> >> regards, >> >> sylvain >> >> _______________________________________________ >> rrd-users mailing list >> rrd-users at lists.oetiker.ch >> https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users >> From jjneely at ncsu.edu Tue Oct 2 18:48:36 2007 From: jjneely at ncsu.edu (Jack Neely) Date: Tue, 2 Oct 2007 12:48:36 -0400 Subject: [rrd-users] RRDs in AFS Message-ID: <20071002164836.GD30917@anduril.unity.ncsu.edu> Folks, Does anyone have experience using/updating RRDs that live in the OpenAFS filesystem? Normally, AFS provides our shared storage for redundant and highly available servers. I have several RRDTool scripts/applications setup in AFS and I ensure that only one process writes to a given RRD. However, while some work well others suffer from silent data corruption when updating the RRDs. A small handful of rows are written properly (according to rrdtool dump) but the next update removes all the PDPs except for the newly inserted row. Any clues? I'm using rrdtool 1.2.23 and 1.2.18. Jack Neely -- Jack Neely NCSU Campus Linux Services Lead Office of Information Technology, NC State University GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 From tobi at oetiker.ch Wed Oct 3 00:07:10 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Wed, 3 Oct 2007 00:07:10 +0200 (CEST) Subject: [rrd-users] does rrd aberrant behaviuor detection really work ? In-Reply-To: References: <50107.10.0.1.56.1191072554.squirrel@laposte> <47920.10.0.1.56.1191188045.squirrel@laposte> Message-ID: Sylvain, in the 1.3er dev cycle quite a lot happened to rrd_update ... first the whole write logic was redone for better performance and mmap ability ... then evan refactored the whole update code and fixed the bug he mentioned ... so in the whole process it could well be that other things were fixed ... there was for example a bug in ABD triggered when running rrdtool update with multiple updates in a single call on linux (it worked fine on solaris) which was gone after the dma update to rrd_update. cheers tbi -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From tobi at oetiker.ch Wed Oct 3 00:08:23 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Wed, 3 Oct 2007 00:08:23 +0200 (CEST) Subject: [rrd-users] RRDs in AFS In-Reply-To: <20071002164836.GD30917@anduril.unity.ncsu.edu> References: <20071002164836.GD30917@anduril.unity.ncsu.edu> Message-ID: Hi Jack, you may want to compile rrdtool without mmap support cheers tobi -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From sylvain at meduse.homelinux.net Wed Oct 3 01:09:17 2007 From: sylvain at meduse.homelinux.net (sylvain at meduse.homelinux.net) Date: Wed, 3 Oct 2007 01:09:17 +0200 (CEST) Subject: [rrd-users] does rrd aberrant behaviuor detection really work ? In-Reply-To: References: <50107.10.0.1.56.1191072554.squirrel@laposte> <47920.10.0.1.56.1191188045.squirrel@laposte> Message-ID: <49294.10.0.1.56.1191366557.squirrel@laposte> Hello , thanks a lot for your answers ! I can confirm 2 things : * the discussed bug is fixed in the latest 1.3-beta1 released (and also in the latest svn snapshot). * what i'm doing i a mass update in one time (about 2 weeks of update, with a 300 seconds step, by using "./script.pl | rrdtool -"). Some people (me, and others, i hope...) want their old rrd without abd transformed in rrd with abd, without losing history. I've written a quick perl script that parse the output of a "rrdtool info foo.rrd" and produce a bar.rrd with the same structure + abd structure, but with no data. I've written a very dirty other perl script that parse the output of "rrdtool dump foo.rrd" and fill in bar.rrd with the data (for now, in only takes care of RRA having a pdp_per_row=1). I still have a little bug when datas contain NaN, but it will be hopefully solved tomorow, or at least this week. I use rrdtool at work (cacti + some homemade scripts), and i have some constraints : what i do shouldn't be 'intrusive'. More precisely, rrd perl package is not installed, and i should not install it (don't try to understand, i have renounced my self :-/ ). Thas whyl i'm parsing xml, instead of using perl bindaries. I can't install some libs (pango, cairo, ...), so i produced at home a static binary of rrdtool. That's not that easy, since the --enable-static-programs dont realy work for now, and compiling some parts by hand (without the makefile and libtool) is required. If you (or someone else ?) think my quick and dirty perl scripts could help, tell me, and i'll translate my french comments inside. (sorry for my poor english) regards, sylvain > in the 1.3er dev cycle quite a lot happened to rrd_update ... first > the whole write logic was redone for better performance and mmap > ability ... > > then evan refactored the whole update code and fixed the bug he > mentioned ... > > so in the whole process it could well be that other things were > fixed ... > > there was for example a bug in ABD triggered when running rrdtool > update with multiple updates in a single call on linux (it worked > fine on solaris) which was gone after the dma update to rrd_update. From pfelt at feltonline.com Wed Oct 3 04:41:18 2007 From: pfelt at feltonline.com (Patrick Felt) Date: Tue, 02 Oct 2007 20:41:18 -0600 Subject: [rrd-users] does rrd aberrant behaviuor detection really work ? In-Reply-To: <49294.10.0.1.56.1191366557.squirrel@laposte> References: <50107.10.0.1.56.1191072554.squirrel@laposte> <47920.10.0.1.56.1191188045.squirrel@laposte> <49294.10.0.1.56.1191366557.squirrel@laposte> Message-ID: <4703014E.5000401@feltonline.com> Sylvain, I'd love them! I've been toying with the idea of whipping up a similar thing to extend cacti to use the holt-winters stuff, but haven't had time. pat sylvain at meduse.homelinux.net wrote: > Hello , > > thanks a lot for your answers ! > > I can confirm 2 things : > > * the discussed bug is fixed in the latest 1.3-beta1 released (and also in > the latest svn snapshot). > * what i'm doing i a mass update in one time (about 2 weeks of update, > with a 300 seconds step, by using "./script.pl | rrdtool -"). > > Some people (me, and others, i hope...) want their old rrd without abd > transformed in rrd with abd, without losing history. > > I've written a quick perl script that parse the output of a "rrdtool info > foo.rrd" and produce a bar.rrd with the same structure + abd structure, > but with no data. > > I've written a very dirty other perl script that parse the output of > "rrdtool dump foo.rrd" and fill in bar.rrd with the data (for now, in only > takes care of RRA having a pdp_per_row=1). > > I still have a little bug when datas contain NaN, but it will be hopefully > solved tomorow, or at least this week. > > I use rrdtool at work (cacti + some homemade scripts), and i have some > constraints : what i do shouldn't be 'intrusive'. > More precisely, rrd perl package is not installed, and i should not > install it (don't try to understand, i have renounced my self :-/ ). Thas > whyl i'm parsing xml, instead of using perl bindaries. > > I can't install some libs (pango, cairo, ...), so i produced at home a > static binary of rrdtool. That's not that easy, since the > --enable-static-programs dont realy work for now, and compiling some parts > by hand (without the makefile and libtool) is required. > > If you (or someone else ?) think my quick and dirty perl scripts could > help, tell me, and i'll translate my french comments inside. > > (sorry for my poor english) > > regards, > > sylvain > > >> in the 1.3er dev cycle quite a lot happened to rrd_update ... first >> the whole write logic was redone for better performance and mmap >> ability ... >> >> then evan refactored the whole update code and fixed the bug he >> mentioned ... >> >> so in the whole process it could well be that other things were >> fixed ... >> >> there was for example a bug in ABD triggered when running rrdtool >> update with multiple updates in a single call on linux (it worked >> fine on solaris) which was gone after the dma update to rrd_update. >> > > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > From tobi at oetiker.ch Wed Oct 3 09:04:43 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Wed, 3 Oct 2007 09:04:43 +0200 (CEST) Subject: [rrd-users] does rrd aberrant behaviuor detection really work ? In-Reply-To: <49294.10.0.1.56.1191366557.squirrel@laposte> References: <50107.10.0.1.56.1191072554.squirrel@laposte> <47920.10.0.1.56.1191188045.squirrel@laposte> <49294.10.0.1.56.1191366557.squirrel@laposte> Message-ID: Hi Sylvain, if you make that perlscript so that it is halvway presentable and documented, I guess others might be interested too ... I will also be glad to host it in the contrib download area ... > I use rrdtool at work (cacti + some homemade scripts), and i have some > constraints : what i do shouldn't be 'intrusive'. > More precisely, rrd perl package is not installed, and i should not > install it (don't try to understand, i have renounced my self :-/ ). Thas > whyl i'm parsing xml, instead of using perl bindaries. > > I can't install some libs (pango, cairo, ...), so i produced at home a > static binary of rrdtool. That's not that easy, since the > --enable-static-programs dont realy work for now, and compiling some parts > by hand (without the makefile and libtool) is required. > > If you (or someone else ?) think my quick and dirty perl scripts could > help, tell me, and i'll translate my french comments inside. wouldn't it be alright if you installed everything under a common subdirectory ? because this is what the whole rrdtool install system is made for ... you can even install cairo and friends in that tree cheers tobi -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From peter.o.mueller at gmx.de Wed Oct 3 09:20:42 2007 From: peter.o.mueller at gmx.de (Peter Mueller) Date: Wed, 3 Oct 2007 09:20:42 +0200 Subject: [rrd-users] rrd with clientside scripting In-Reply-To: <20070930171803.GB20252@ergens.op.het.net> References: <20070930144614.GA20252@ergens.op.het.net> <20070930171803.GB20252@ergens.op.het.net> Message-ID: <1CD9EB25-13D7-4566-80BC-CF5824E0B80F@gmx.de> Hi Alex, Am 30.09.2007 um 19:18 schrieb Alex van den Bogaerdt: > On Sun, Sep 30, 2007 at 06:45:43PM +0200, Peter Mueller wrote: > >>>> E.g. I want the user to define the time span and sensor signals of >>>> interest and out of this the graph should be generated. >>> >>> This is a private system, right? Why not write some basic html page >>> yourself, combined with rrdcgi ? >> >> Right, do you know of any examples in the docs that show how to >> transfer user input from a html page back to the server and then into >> rrdcgi? > > Example 2 in the manual page for rrdcgi should give a start. Yes this is exactly what I looked for. No big sw installations just simple CGI. Regards, Peter -- Peter Mueller peter.o.mueller at gmx.de -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071003/0fdbaa31/attachment.html From sl at jm-data.at Wed Oct 3 11:24:55 2007 From: sl at jm-data.at (Stefan Lochbihler) Date: Wed, 3 Oct 2007 09:24:55 +0000 Subject: [rrd-users] Edit rrd files Message-ID: <0000CA02.47037C04@217.16.112.23> undefined -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071003/6280b63b/attachment-0001.html From guthrie at mum.edu Wed Oct 3 14:21:47 2007 From: guthrie at mum.edu (Gregory Guthrie) Date: Wed, 03 Oct 2007 07:21:47 -0500 Subject: [rrd-users] rrd graph - multiple graphs on one plot? Message-ID: <6.2.5.6.2.20071003072004.02564998@mum.edu> I wanted to have multiple data on one graph, and also add an additional set of y-axis labels. I'd like to have wind speed and direction on one graph, and have the velocity labels as standard on the left Y-axis, but also add a direction indicator (N S E W) scale label over the scaled (0-360 => 20-30) wind direction on the right hand side of the graph. A site with a similar thing is at: http://www.iwindsurf.com/windandwhere.iws?regionID=100&siteID=1399 but they use icons for direction, I wanted to just plot it as a graph. Thanks, Greg ----------------------------------------------- Gregory Guthrie MUM Faculty Mail - FM 1068 Fairfield, IA 52557 http://www.mum.edu/~guthrie (641)472-7773 ------------------------------------------------ From guthrie at mum.edu Wed Oct 3 14:18:32 2007 From: guthrie at mum.edu (Gregory Guthrie) Date: Wed, 03 Oct 2007 07:18:32 -0500 Subject: [rrd-users] rrd error: attempt to put segment in horiz list twice Message-ID: <6.2.5.6.2.20071003071256.02580938@mum.edu> I have a simple application that plots weather data from an XML feed. (I started with rrdWeather). It was beautiful and working fine, but just yesterday it started giving intermittent errors: "attempt to put segment in horiz list twice" and then the resulting rrdTool graph was bad. A google search seemed to just show past messages saying this was a bug in old libraries. I am using rrdTool 1.2.15 from perl libraries (librrdp-perl 1.2.15-0.3, librrds-perl 1.2.15-0.3) which are the current latest stable builds on Debian (Linux). Help please? ----------------------------------------------- Gregory Guthrie MUM Faculty Mail - FM 1068 Fairfield, IA 52557 http://www.mum.edu/~guthrie (641)472-7773 ------------------------------------------------ From linux at thehobsons.co.uk Wed Oct 3 14:14:13 2007 From: linux at thehobsons.co.uk (Simon Hobson) Date: Wed, 3 Oct 2007 13:14:13 +0100 Subject: [rrd-users] Edit rrd files In-Reply-To: <0000CA02.47037C04@217.16.112.23> References: <0000CA02.47037C04@217.16.112.23> Message-ID: Stefan Lochbihler wrote: >I created a rrd file with the following settings. > >--step 240 >DS:temperature:GAUGE:11160:U:U >RRA:AVERAGE:0.5:1:11160 >RRA:AVERAGE:0.5:15:4392 >RRA:AVERAGE:0.5:4392 >RRA:AVERAGE:0.5:15:4392 >RRA:AVERAGE:0.5:360:366 > >The file has already a vast amount of data entries. > >Now I want to put this data entries in a new rrd file >which i want to create with the following command. > >--step 300 >DS:temperature:GAUGE:900:U:U >RRA:MAX:0.5:1:8928 >RRA:MAX:0.5:11:4464 >RRA:MAX:0.5:288:366 > >You can see that the values of step, RRA, min,max,... >would be changed. > >Do you know if is there a way do to this without corrupt my data entries. Not really, you would have to dump what you have and feed it in to a new database. But apart from re-consolidating the data (which will obviously have an effect on the accuracy), there is no way you can ever get a meaningful MAX value from an average. Consider this sequence : 0 0 0 0 0 0 0 0 0 100 Average is 10, max is 100 ! From linux at thehobsons.co.uk Wed Oct 3 20:08:30 2007 From: linux at thehobsons.co.uk (Simon Hobson) Date: Wed, 3 Oct 2007 19:08:30 +0100 Subject: [rrd-users] rrd graph - multiple graphs on one plot? In-Reply-To: <6.2.5.6.2.20071003072004.02564998@mum.edu> References: <6.2.5.6.2.20071003072004.02564998@mum.edu> Message-ID: Gregory Guthrie wrote: >I wanted to have multiple data on one graph, and also add an >additional set of y-axis labels. > >I'd like to have wind speed and direction on one graph, and have the >velocity labels as standard on the left Y-axis, but also add a >direction indicator (N S E W) scale label over the scaled >(0-360 => 20-30) wind direction on the right hand side of the graph. It's a FAQ, and I think it's already been asked once this week (have you read the archives ?). The answer is no you can't. I suspect that there are people looking at how to code it since it is such a popular request. On the other hand, when you stop and think about it, it stops being such a trivial exercise to scale both scales sensibly while sharing a common grid. From guthrie at mum.edu Wed Oct 3 22:08:19 2007 From: guthrie at mum.edu (Gregory Guthrie) Date: Wed, 03 Oct 2007 15:08:19 -0500 Subject: [rrd-users] rrd error: attempt to put segment in horiz list twice Message-ID: <6.2.5.6.2.20071003150812.02001158@mum.edu> I have a simple application that plots weather data from an XML feed. (I started with rrdWeather). It was beautiful and working fine, but just yesterday it started giving intermittent errors: "attempt to put segment in horiz list twice" and then the resulting rrdTool graph was bad. A google search seemed to just show past messages saying this was a bug in old libraries. I am using rrdTool 1.2.15 from perl libraries (librrdp-perl 1.2.15-0.3, librrds-perl 1.2.15-0.3) which are the current latest stable builds on Debian (Linux). Help please? ----------------------------------------------- Gregory Guthrie MUM Faculty Mail - FM 1068 Fairfield, IA 52557 (641)472-7773 ------------------------------------------------ -------------------------------------------------------- Dr. Gregory Guthrie guthrie at mum.edu (641)472-1125 Fax: -1103 Professor and Dean Computer Science Department College of Computer Science and Mathematics Maharishi University of Management http://www.mum.edu/cs_dept -------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071003/9e75e3b3/attachment.html From guthrie at mum.edu Wed Oct 3 22:08:08 2007 From: guthrie at mum.edu (Gregory Guthrie) Date: Wed, 03 Oct 2007 15:08:08 -0500 Subject: [rrd-users] rrd graph - multiple graphs on one plot? Message-ID: <6.2.5.6.2.20071003150752.01efd6c0@mum.edu> I wanted to have multiple data on one graph, and also add an additional set of y-axis labels. I'd like to have wind speed and direction on one graph, and have the velocity labels as standard on the left Y-axis, but also add a direction indicator (N S E W) scale label over the scaled (0-360 => 20-30) wind direction on the right hand side of the graph. A site with a similar thing is at: http:/ /www.iwindsurf.com/windandwhere.iws?regionID=100&siteID=1399 but they use icons for direction, I wanted to just plot it as a graph. Thanks, Greg ----------------------------------------------- Gregory Guthrie MUM Faculty Mail - FM 1068 Fairfield, IA 52557 (641)472-7773 ------------------------------------------------ -------------------------------------------------------- Dr. Gregory Guthrie guthrie at mum.edu (641)472-1125 Fax: -1103 Professor and Dean Computer Science Department College of Computer Science and Mathematics Maharishi University of Management http://www.mum.edu/cs_dept -------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071003/d0eea7fd/attachment.html From rjconway at bigpond.com Wed Oct 3 23:32:16 2007 From: rjconway at bigpond.com (Rob Conway) Date: Thu, 4 Oct 2007 07:32:16 +1000 Subject: [rrd-users] rrd graph - multiple graphs on one plot? References: <6.2.5.6.2.20071003150752.01efd6c0@mum.edu> Message-ID: <004901c80604$deb4dd10$5001a8c0@home> Yes you can have multiple scales...however rrdtool cannot provide the scale. If you normalise the data so it all fits on a single y axis, i.e. normalise everything to 0-100 %, all you then need to do is superimpose a scale onto the rrdtrend. DO NOT LET rrd print the scale just the trend area. it is quite easy once you create a small picture (png) having the scale. Imagemagick enables you to combine two .png files together and is a command line editor/tool for data manipulation under linux or I think windows. I run it under a SLUG and linux ubuntu. Thus you could make a complex multiscale axis and you just use imagemagick to combine the rrd graph and scale. The image below is just the rrdgraph combined with a scale to show alarm setpoint etc however theres nothing stopping you creating an alternative scale. http://img.photobucket.com/albums/v319/rjconway/rob.png This is the scale file, image magic updates this with actual data then overlays the completed scale onto the trend...and in firefox it shows its transparency http://img.photobucket.com/albums/v319/rjconway/tempscale.png I use image magic to update the header on the web page below that shows real time data above the trend areas http://smg.photobucket.com/albums/v319/rjconway/?action=view¤t=newupdate.jpg RRDTool and imagemagick and you have all the required tools to create a really nice html interface for you data Imagemagic runs under linux and the NSLU2 slug .... rob, Australia ----- Original Message ----- From: Gregory Guthrie To: rrd-users at lists.oetiker.ch Sent: Thursday, October 04, 2007 6:08 AM Subject: [rrd-users] rrd graph - multiple graphs on one plot? I wanted to have multiple data on one graph, and also add an additional set of y-axis labels. I'd like to have wind speed and direction on one graph, and have the velocity labels as standard on the left Y-axis, but also add a direction indicator (N S E W) scale label over the scaled (0-360 => 20-30) wind direction on the right hand side of the graph. A site with a similar thing is at: http:/ / www.iwindsurf.com/windandwhere.iws?regionID=100&siteID=1399 but they use icons for direction, I wanted to just plot it as a graph. Thanks, Greg ----------------------------------------------- Gregory Guthrie MUM Faculty Mail - FM 1068 Fairfield, IA 52557 (641)472-7773 ------------------------------------------------ -------------------------------------------------------- Dr. Gregory Guthrie guthrie at mum.edu (641)472-1125 Fax: -1103 Professor and Dean Computer Science Department College of Computer Science and Mathematics Maharishi University of Management http://www.mum.edu/cs_dept -------------------------------------------------------- ------------------------------------------------------------------------------ _______________________________________________ rrd-users mailing list rrd-users at lists.oetiker.ch https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071004/9eba5a1a/attachment-0001.html From cos at aaaaa.org Thu Oct 4 19:34:27 2007 From: cos at aaaaa.org (Ofer Inbar) Date: Thu, 4 Oct 2007 13:34:27 -0400 Subject: [rrd-users] graphs with logarithmic Y-axis? Message-ID: <20071004173427.GH466@mip.polyamory.org> I sent this to the Ganglia list, but it might be an RRD Tool issue. Ganglia is a system that uses daemons to store metrics about system performance in RRD files, and has a web interface that generates graphs from those metrics using rrdgraph. It's written in PHP and the code passes command line options and graph definitions from variables to rrdgraph, though I don't haven't read all the code so I don't know how it works in detail. I seem to have a problem adding -o or --logarithmic to graphs that otherwise generate just fine; if I try that option no image is made. .......... I've defined a lot of custom graphs to view our Ganglia data, because most of the metrics we want to see are custom metrics. I just add new blocks to the if-else if... structure in graph.php, with new RRD graph definitions, and plug those into our custom PHP Ganglia front page, and it works quite well. There's one set of custom metrics we'd like to view with a logarithmic Y-axis. rrdgraph's documentation says it can generate a graph with a logarithmic Y-axis if you pass it the -o or --logarithmic option. I tried adding this to the $upper_limit or $extras variable in the block defining the graph in question, but when I do that no graph gets generated at all. Has anyone succesfully modified graph.php to add graphs with logarithmic Y-axis? How'd you do it? We're running: rrdtool 1.2.18 Ganglia 3.0.1 RedHat EL 4 -- Cos From jmack at wm7d.net Thu Oct 4 23:35:16 2007 From: jmack at wm7d.net (Joseph Mack NA3T) Date: Thu, 4 Oct 2007 14:35:16 -0700 (PDT) Subject: [rrd-users] AREA starting at value rather than at y=0 Message-ID: using rrdtool-1.0.45 I have a value with a +/- error margin . I'd like to have an AREA starting at (value-error) of height 2*error to give a band of color corresponding to the error limits. I've tried starting the bottom with AREA:(value-error) using black or white as the color of AREA, but these colors are obtrusive. I'd rather have a transparent AREA. (I then STACK 2*error.) Alternately I use LINE2 for the lower limit, the most likely value, and the upper limit, but then there are 3 lines and the viewer thinks I'm plotting 3 values. Any suggestions? Thanks Joe -- Joseph Mack NA3T EME(B,D), FM05lw North Carolina jmack (at) wm7d (dot) net - azimuthal equidistant map generator at http://www.wm7d.net/azproj.shtml Homepage http://www.austintek.com/ It's GNU/Linux! From g.s at arcor.de Fri Oct 5 08:06:09 2007 From: g.s at arcor.de (Gerhard Strangar) Date: Fri, 05 Oct 2007 08:06:09 +0200 Subject: [rrd-users] AREA starting at value rather than at y=0 In-Reply-To: References: Message-ID: <4705D451.7000909@arcor.de> Joseph Mack NA3T wrote (2007-10-04 23:35): > using rrdtool-1.0.45 > > I have a value with a +/- error margin . I'd like to have an > AREA starting at (value-error) of height 2*error to give a > band of color corresponding to the error limits. What I do is: Draw an area from 0 to maximum in grey, then draw an area from 0 to minumum in white and then draw a line in color - no stacking at all. (If your values can get negative, you have do little more, but it's possible as well) -- * Origin: Fido over IP (2:240/2188.575) From alex at ergens.op.het.net Fri Oct 5 13:14:39 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Fri, 5 Oct 2007 13:14:39 +0200 Subject: [rrd-users] AREA starting at value rather than at y=0 In-Reply-To: References: Message-ID: <20071005111439.GC13650@ergens.op.het.net> On Thu, Oct 04, 2007 at 02:35:16PM -0700, Joseph Mack NA3T wrote: > using rrdtool-1.0.45 > > I have a value with a +/- error margin . I'd like to have an > AREA starting at (value-error) of height 2*error to give a > band of color corresponding to the error limits. > > I've tried starting the bottom with AREA:(value-error) using > black or white as the color of AREA, but these colors are > obtrusive. I'd rather have a transparent AREA. (I then STACK > 2*error.) Alternately I use LINE2 for the lower limit, the > most likely value, and the upper limit, but then there are 3 > lines and the viewer thinks I'm plotting 3 values. > > Any suggestions? Keep it as is, except that you change the color from black or white into fully transparent. You can do this by leaving out the color, or by explicitly setting transparency to full: LINE1:valueminuserror AREA:errortwice#FF0000:errorrange:STACK or AREA:valueminuserror#12345600:legend AREA:errortwice#FF0000:errorrange:STACK or AREA:valueminuserror#1230:legend AREA:errortwice#F00:errorrange:STACK -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/ From jmack at wm7d.net Fri Oct 5 15:59:30 2007 From: jmack at wm7d.net (Joseph Mack NA3T) Date: Fri, 5 Oct 2007 06:59:30 -0700 (PDT) Subject: [rrd-users] AREA starting at value rather than at y=0 In-Reply-To: <4705D451.7000909@arcor.de> References: <4705D451.7000909@arcor.de> Message-ID: On Fri, 5 Oct 2007, Gerhard Strangar wrote: > Joseph Mack NA3T wrote (2007-10-04 23:35): >> using rrdtool-1.0.45 >> >> I have a value with a +/- error margin . I'd like to have an >> AREA starting at (value-error) of height 2*error to give a >> band of color corresponding to the error limits. > > What I do is: > Draw an area from 0 to maximum in grey, then draw an area from 0 to > minumum in white I tried something like this and found the grey instrusive - it obliterated the striped background for one thing. As well if I have multiple sets of data, they can be overwritten. Thanks though Joe -- Joseph Mack NA3T EME(B,D), FM05lw North Carolina jmack (at) wm7d (dot) net - azimuthal equidistant map generator at http://www.wm7d.net/azproj.shtml Homepage http://www.austintek.com/ It's GNU/Linux! From jmack at wm7d.net Fri Oct 5 16:01:49 2007 From: jmack at wm7d.net (Joseph Mack NA3T) Date: Fri, 5 Oct 2007 07:01:49 -0700 (PDT) Subject: [rrd-users] AREA starting at value rather than at y=0 In-Reply-To: <20071005111439.GC13650@ergens.op.het.net> References: <20071005111439.GC13650@ergens.op.het.net> Message-ID: On Fri, 5 Oct 2007, Alex van den Bogaerdt wrote: > You can do this by leaving out the color, didn't know you could do that. Works perfectly! > or by explicitly setting > transparency to full: didn't know that you could do transparency. I have rrdtool-1.0.x. I see transparency is in 1.2.x. Maybe I should upgrade, but the no-color option is just fine for the moment Thanks Joe -- Joseph Mack NA3T EME(B,D), FM05lw North Carolina jmack (at) wm7d (dot) net - azimuthal equidistant map generator at http://www.wm7d.net/azproj.shtml Homepage http://www.austintek.com/ It's GNU/Linux! From g.s at arcor.de Fri Oct 5 17:19:09 2007 From: g.s at arcor.de (Gerhard Strangar) Date: Fri, 05 Oct 2007 17:19:09 +0200 Subject: [rrd-users] AREA starting at value rather than at y=0 In-Reply-To: <20071005111439.GC13650@ergens.op.het.net> References: <20071005111439.GC13650@ergens.op.het.net> Message-ID: <470655ED.2060008@arcor.de> Alex van den Bogaerdt wrote (2007-10-05 13:14): > On Thu, Oct 04, 2007 at 02:35:16PM -0700, Joseph Mack NA3T wrote: >> using rrdtool-1.0.45 > LINE1:valueminuserror > AREA:errortwice#FF0000:errorrange:STACK No, you can't stack areas ontop of lines in 1.0.45. -- * Origin: Fido over IP (2:240/2188.575) From plancashire at ci.portland.or.us Fri Oct 5 23:55:45 2007 From: plancashire at ci.portland.or.us (Lancashire, Pete) Date: Fri, 5 Oct 2007 14:55:45 -0700 Subject: [rrd-users] seg fault Message-ID: <7EF5BBDCF35CD944A9FDB06AB82AF39904568C88@CITYEMAIL2> not sure if this is new, in advertently happend when I made a mistake in a wrapper script. if you leave off the number of rows on a RRA the create will work but the update will seg fault. i have not had time to look any closer, maybe this weekend $ rrdtool create foo.rrd --start 1170000000 --step 10 DS:one:GAUGE:20:U:U RRA:AVERAGE:0.5:1 $ rrdtool update foo.rrd 1191600622:1 Segmentation fault $ rrdtool version RRDtool 1.2.23 Copyright 1997-2007 by Tobias Oetiker Compiled Jul 10 2007 21:41:00 x86, fedora 7 cat /proc/version Linux version 2.6.22.9-91.fc7 (kojibuilder at xenbuilder1.fedora.redhat.com) (gcc version 4.1.2 20070502 (Red Hat 4.1.2-12)) #1 SMP Thu Sep 27 23:10:59 EDT 2007 rpm -qa rrdtool rrdtool-1.2.23-6.fc7 i also downloaded the latest stable, compiled it, and same result -pete From mario.carassale at sohonet.co.uk Thu Oct 11 12:46:54 2007 From: mario.carassale at sohonet.co.uk (mario.carassale) Date: Thu, 11 Oct 2007 11:46:54 +0100 Subject: [rrd-users] RRDtool directory missing when compiling Ntop3.3 Message-ID: <470DFF1E.6020204@sohonet.co.uk> Hi, when i try to compile ntop3.3 under debian lenny i got an error saying that the directory /usr/local/rrdtool cannot be found. Any ideas? Thanks Mario From david.carmean at netapp.com Fri Oct 12 22:01:10 2007 From: david.carmean at netapp.com (David Carmean) Date: Fri, 12 Oct 2007 13:01:10 -0700 Subject: [rrd-users] Find first non-NaN value in RRD? Message-ID: <20071012200110.GP63249@netapp.com> I need to programmatically choose between dozens, if not a hundred, pairs of RRD files to determine which have the "best" or the "most" data. One of the things I'm doing is an rrddump and fgrep to count the number of non-NaN samples. But I also need to find the first and last non-NaN values in the the RRD. There's only one ds per RRD (Zenoss). rrdfirst is not behaving as I'd expect. It gives me the first/oldest slot, no matter whether it has a valid sample or not. Somebody has to have solved this problem? Thanks. -- David Carmean Network Appliance, Inc Infosystems Architect, 495 E. Java Drive Java (Sunnyvale) Engineering Lab Services Sunnyvale, CA 94089 From alex at ergens.op.het.net Sat Oct 13 01:08:41 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Sat, 13 Oct 2007 01:08:41 +0200 Subject: [rrd-users] Find first non-NaN value in RRD? In-Reply-To: <20071012200110.GP63249@netapp.com> References: <20071012200110.GP63249@netapp.com> Message-ID: <20071012230841.GA4384@ergens.op.het.net> On Fri, Oct 12, 2007 at 01:01:10PM -0700, David Carmean wrote: > I need to programmatically choose between dozens, if not a hundred, > pairs of RRD files to determine which have the "best" or the "most" > data. One of the things I'm doing is an rrddump and fgrep to count > the number of non-NaN samples. But I also need to find the first > and last non-NaN values in the the RRD. There's only one ds per > RRD (Zenoss). rrdfirst is not behaving as I'd expect. It gives > me the first/oldest slot, no matter whether it has a valid sample or > not. > > Somebody has to have solved this problem? Doesn't rrdtool 1.2 VDEF FIRST do what you want? See man rrdgraph_rpn Then print the time using PRINT (not GPRINT) and use it in whatever script you are building. See man rrdgraph_graph HTH -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/ From ladaan at iptel.org Sun Oct 14 10:43:47 2007 From: ladaan at iptel.org (Ladislav Andel) Date: Sun, 14 Oct 2007 10:43:47 +0200 Subject: [rrd-users] remote rrdtool access Message-ID: <4711D6C3.4000103@iptel.org> Hello, I have setup my rrdsrv service via inetd and telneting to the meta-server is working fine. Now I just need to read rrd files and work with them locally. Could you give me a hint where to start or give me an example? Does anybody have a working code to fetch rrd files remotely with perl sockets or netcat? I have never worked with perl sockets or netcat(probably somehow through pipes). Thank you hints. Lada From ladaan at iptel.org Sun Oct 14 11:06:19 2007 From: ladaan at iptel.org (Ladislav Andel) Date: Sun, 14 Oct 2007 11:06:19 +0200 Subject: [rrd-users] remote rrdtool access In-Reply-To: <4711D6C3.4000103@iptel.org> References: <4711D6C3.4000103@iptel.org> Message-ID: <4711DC0B.2060301@iptel.org> I'm sorry for the confusion, but here is exactly what I want. I have a web pages which reads graphs created by rrdgraph but some of the rrd archives are on remote machines and I need to get either created graphs or the rrd archives and make the graphs locally where the webserver is running. So possibly I could transfer rrd archives via scp to local machine and create those graphs. I don't know if this is the best approach I could go but if you give any comments to any good and easy working approach I'd be very glad. Thanks. Lada Ladislav Andel wrote: > Hello, > I have setup my rrdsrv service via inetd and telneting to the > meta-server is working fine. > Now I just need to read rrd files and work with them locally. > Could you give me a hint where to start or give me an example? > Does anybody have a working code to fetch rrd files remotely with perl > sockets or netcat? > I have never worked with perl sockets or netcat(probably somehow through > pipes). > > Thank you hints. > > Lada > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > From davidtball at gmail.com Sun Oct 14 16:36:01 2007 From: davidtball at gmail.com (David Ball) Date: Sun, 14 Oct 2007 08:36:01 -0600 Subject: [rrd-users] remote rrdtool access In-Reply-To: <4711DC0B.2060301@iptel.org> References: <4711D6C3.4000103@iptel.org> <4711DC0B.2060301@iptel.org> Message-ID: <8d4861b00710140736m44ecf5c6kbe6e1f85df4dd23e@mail.gmail.com> Since RRD files are architecture-dependant, you might be safer to generate the graphs on the original machine and copy them over, OR, make use of the RRDtool 'dump' command on the original machine and scp the .xml files that are created. You'd then need to use the rrdtool 'restore' command to (re)generate the .rrd files on the local machine, then generate your graphs from those. That said, if you have a homogeneous server environment (same architecture and OS on your machines) you might get away with simply copying the .rrd files directly. David On 14/10/2007, Ladislav Andel wrote: > I'm sorry for the confusion, but here is exactly what I want. > I have a web pages which reads graphs created by rrdgraph but some of > the rrd archives > are on remote machines and I need to get either created graphs or the > rrd archives and make the graphs locally > where the webserver is running. So possibly I could transfer rrd > archives via scp to local machine and create those graphs. > I don't know if this is the best approach I could go but if you give any > comments to any good and easy working approach > I'd be very glad. > > Thanks. > Lada > > > Ladislav Andel wrote: > > Hello, > > I have setup my rrdsrv service via inetd and telneting to the > > meta-server is working fine. > > Now I just need to read rrd files and work with them locally. > > Could you give me a hint where to start or give me an example? > > Does anybody have a working code to fetch rrd files remotely with perl > > sockets or netcat? > > I have never worked with perl sockets or netcat(probably somehow through > > pipes). > > > > Thank you hints. > > > > Lada > > > > _______________________________________________ > > rrd-users mailing list > > rrd-users at lists.oetiker.ch > > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > From tobi at oetiker.ch Sun Oct 14 17:14:47 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Sun, 14 Oct 2007 17:14:47 +0200 (CEST) Subject: [rrd-users] remote rrdtool access In-Reply-To: <8d4861b00710140736m44ecf5c6kbe6e1f85df4dd23e@mail.gmail.com> References: <4711D6C3.4000103@iptel.org> <4711DC0B.2060301@iptel.org> <8d4861b00710140736m44ecf5c6kbe6e1f85df4dd23e@mail.gmail.com> Message-ID: David and friends, have you considered running the rrd storage on a central server and just transfering the data for input into rrd update to that server? this is the model I am using in the new master/slave mode of smokeping. if you need a local copy of the rrd files, you could also update the local copy and transfer the update info to the central server at the same time. cheers tobi -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From salatiel.filho at gmail.com Wed Oct 17 05:02:24 2007 From: salatiel.filho at gmail.com (Salatiel Filho) Date: Wed, 17 Oct 2007 00:02:24 -0300 Subject: [rrd-users] rrdtool seg fault using GPRINT Message-ID: Hi guys , i am trying to use rrdtool but everytime i try to use some GPRINT function i get segmentation fault. This limits a lot the use of rrdtool. Could anyone lead me to a solution. I had de debian etch binary 1.2.15 for arm , and now i have 1.2.23 compiled by myself. Both segfault. # uname -a Linux LS-GL7D6 2.6.12.6-arm1 #77 Tue Dec 5 22:23:46 CST 2006 armv5tejl GNU/Linux # rrdtool -v RRDtool 1.2.23 Copyright 1997-2007 by Tobias Oetiker Compiled Oct 16 2007 23:04:34 # /usr/bin/rrdtool graph /tmp/a.png --imgformat=PNG --start=-86400 --end=-300 --title="LSPRO - Load Average" --rigid --base=1000 --height=120 --width=500 --alt-autoscale-max --lower-limit=0 --units-exponent=0 --vertical-label="processes in the run queue" --slope-mode DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_load_1min_12.rrd":load_1min:AVERAGE DEF:b="/usr/share/cacti-0.8.6j/rra/lspro_load_5min_14.rrd":load_5min:AVERAGE DEF:c="/usr/share/cacti-0.8.6j/rra/lspro_load_15min_13.rrd":load_15min:AVERAGE CDEF:cdefg=TIME,1192584729,GT,a,a,UN,0,a,IF,IF,TIME,1192584729,GT,b,b,UN,0,b,IF,IF,TIME,1192584729,GT,c,c,UN,0,c,IF,IF,+,+ AREA:a#EACC00:"1 Minute Average" GPRINT:a:LAST:" Current\:%8.2lf\n" AREA:b#EA8F00:"5 Minute Average":STACK GPRINT:b:LAST:" Current\:%8.2lf\n" AREA:c#FF0000:"15 Minute Average":STACK GPRINT:c:LAST:"Current\:%8.2lf\n" LINE1:cdefg#000000:"Total" Segmentation fault # strace /usr/bin/rrdtool graph /tmp/a.png --imgformat=PNG --start=-86400 --end=-300 --title="LSPRO - Load Average" --rigid --base=1000 --height=120 --width=500 --alt-autoscale-max --lower-limit=0 --units-exponent=0 --vertical-label="processes in the run queue" --slope-mode DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_load_1min_12.rrd":load_1min:AVERAGE DEF:b="/usr/share/cacti-0.8.6j/rra/lspro_load_5min_14.rrd":load_5min:AVERAGE DEF:c="/usr/share/cacti-0.8.6j/rra/lspro_load_15min_13.rrd":load_15min:AVERAGE CDEF:cdefg=TIME,1192584729,GT,a,a,UN,0,a,IF,IF,TIME,1192584729,GT,b,b,UN,0,b,IF,IF,TIME,1192584729,GT,c,c,UN,0,c,IF,IF,+,+ AREA:a#EACC00:"1 Minute Average" GPRINT:a:LAST:" Current\:%8.2lf\n" AREA:b#EA8F00:"5 Minute Average":STACK GPRINT:b:LAST:" Current\:%8.2lf\n" AREA:c#FF0000:"15 Minute Average":STACK GPRINT:c:LAST:"Current\:%8.2lf\n" LINE1:cdefg#000000:"Total" execve("/usr/bin/rrdtool", ["/usr/bin/rrdtool", "graph", "/tmp/a.png", "--imgformat=PNG", "--start=-86400", "--end=-300", "--title=LSPRO - Load Average", "--rigid", "--base=1000", "--height=120", "--width=500", "--alt-autoscale-max", "--lower-limit=0", "--units-exponent=0", "--vertical-label=processes in th"..., "--slope-mode", ...], [/* 23 vars */]) = 0 uname({sys="Linux", node="LS-GL7D6", ...}) = 0 brk(0) = 0x11000 access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT (No such file or directory) mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40015000 access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory) open("/etc/ld.so.cache", O_RDONLY) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=20010, ...}) = 0 mmap2(NULL, 20010, PROT_READ, MAP_PRIVATE, 3, 0) = 0x40016000 close(3) = 0 access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT (No such file or directory) open("/usr/lib/librrd.so.2", O_RDONLY) = 3 read(3, "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0hV\0\0004"..., 512) = 512 fstat64(3, {st_mode=S_IFREG|0755, st_size=578967, ...}) = 0 mmap2(NULL, 311844, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x4001e000 mprotect(0x4005d000, 53796, PROT_NONE) = 0 mmap2(0x40065000, 16384, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x3f) = 0x40065000 mmap2(0x40069000, 4644, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40069000 close(3) = 0 access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT (No such file or directory) open("/usr/lib/libpng12.so.0", O_RDONLY) = 3 read(3, "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\0045\0\000"..., 512) = 512 fstat64(3, {st_mode=S_IFREG|0644, st_size=127124, ...}) = 0 mmap2(NULL, 158800, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x4006b000 mprotect(0x4008a000, 31824, PROT_NONE) = 0 mmap2(0x40091000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1e) = 0x40091000 close(3) = 0 access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT (No such file or directory) open("/usr/lib/libz.so.1", O_RDONLY) = 3 read(3, "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\254\26\0"..., 512) = 512 fstat64(3, {st_mode=S_IFREG|0644, st_size=84988, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40092000 mmap2(NULL, 116668, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x40093000 mprotect(0x400a8000, 30652, PROT_NONE) = 0 mmap2(0x400af000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x14) = 0x400af000 close(3) = 0 access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT (No such file or directory) open("/lib/libm.so.6", O_RDONLY) = 3 read(3, "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\3042\0\000"..., 512) = 512 fstat64(3, {st_mode=S_IFREG|0644, st_size=78836, ...}) = 0 mmap2(NULL, 110572, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x402a6000 mprotect(0x402b9000, 32748, PROT_NONE) = 0 mmap2(0x402c0000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x12) = 0x402c0000 close(3) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x402c1000 mprotect(0x40233000, 8192, PROT_READ) = 0 mprotect(0x40123000, 4096, PROT_READ) = 0 mprotect(0x4001c000, 4096, PROT_READ) = 0 munmap(0x40016000, 20010) = 0 brk(0) = 0x11000 brk(0x32000) = 0x32000 open("/etc/localtime", O_RDONLY) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 fstat64(3, {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40016000 read(3, "TZif\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\3\0\0\0\3\0"..., 4096) = 276 close(3) = 0 munmap(0x40016000, 4096) = 0 open("/usr/lib/locale/locale-archive", O_RDONLY|O_LARGEFILE) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=1449376, ...}) = 0 mmap2(NULL, 1449376, PROT_READ, MAP_PRIVATE, 3, 0) = 0x402c2000 close(3) = 0 time(NULL) = 1192589982 time(NULL) = 1192589982 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 time(NULL) = 1192589982 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 time(NULL) = 1192589982 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=276, ...}) = 0 open("/usr/share/fonts/truetype/ttf-dejavu/DejaVuSansMono.ttf", O_RDONLY) = 3 fcntl64(3, F_SETFD, FD_CLOEXEC) = 0 fstat64(3, {st_mode=S_IFREG|0644, st_size=258332, ...}) = 0 mmap2(NULL, 258332, PROT_READ, MAP_PRIVATE, 3, 0) = 0x40424000 close(3) = 0 open("/usr/lib/gconv/gconv-modules.cache", O_RDONLY) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=21584, ...}) = 0 mmap2(NULL, 21584, PROT_READ, MAP_SHARED, 3, 0) = 0x40464000 close(3) = 0 munmap(0x40424000, 258332) = 0 open("/usr/share/fonts/truetype/ttf-dejavu/DejaVuSansMono.ttf", O_RDONLY) = 3 fcntl64(3, F_SETFD, FD_CLOEXEC) = 0 fstat64(3, {st_mode=S_IFREG|0644, st_size=258332, ...}) = 0 mmap2(NULL, 258332, PROT_READ, MAP_PRIVATE, 3, 0) = 0x40424000 close(3) = 0 munmap(0x40424000, 258332) = 0 open("/usr/share/cacti-0.8.6j/rra/lspro_load_1min_12.rrd", O_RDONLY) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=47840, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40424000 read(3, "RRD\0000003\0\0\0\0C+\37[/%\300\307\1\0\0\0\10\0\0\0,\1"..., 4096) = 4096 _llseek(3, 0, [4096], SEEK_CUR) = 0 _llseek(3, 4096, [4096], SEEK_SET) = 0 read(3, "4\24\324?\36;\225\3718\326\345?\365JY\206\336\2\347?\27"..., 4096) = 4096 _llseek(3, 0, [0], SEEK_SET) = 0 read(3, "RRD\0000003\0\0\0\0C+\37[/%\300\307\1\0\0\0\10\0\0\0,\1"..., 4096) = 4096 close(3) = 0 munmap(0x40424000, 4096) = 0 open("/usr/share/cacti-0.8.6j/rra/lspro_load_5min_14.rrd", O_RDONLY) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=47840, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40424000 read(3, "RRD\0000003\0\0\0\0C+\37[/%\300\307\1\0\0\0\10\0\0\0,\1"..., 4096) = 4096 _llseek(3, 0, [4096], SEEK_CUR) = 0 _llseek(3, 4096, [4096], SEEK_SET) = 0 read(3, "1-\307?\17R\335I0*\321?}a2U\0\336\326?Y\27\267\321d\205"..., 4096) = 4096 _llseek(3, 0, [0], SEEK_SET) = 0 read(3, "RRD\0000003\0\0\0\0C+\37[/%\300\307\1\0\0\0\10\0\0\0,\1"..., 4096) = 4096 close(3) = 0 munmap(0x40424000, 4096) = 0 open("/usr/share/cacti-0.8.6j/rra/lspro_load_15min_13.rrd", O_RDONLY) = 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=47840, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40424000 read(3, "RRD\0000003\0\0\0\0C+\37[/%\300\307\1\0\0\0\10\0\0\0,\1"..., 4096) = 4096 _llseek(3, 0, [4096], SEEK_CUR) = 0 _llseek(3, 4096, [4096], SEEK_SET) = 0 read(3, "\267\33\307?\215\224@]~\214\311?x\v$(&S\315?\246\275\301"..., 4096) = 4096 _llseek(3, 0, [0], SEEK_SET) = 0 read(3, "RRD\0000003\0\0\0\0C+\37[/%\300\307\1\0\0\0\10\0\0\0,\1"..., 4096) = 4096 close(3) = 0 munmap(0x40424000, 4096) = 0 mmap2(NULL, 114532352, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x4046a000 time(NULL) = 1192589982 +++ killed by SIGSEGV +++ any help will be really appreciated . Thanks in advance. -- []'s Salatiel "O maior prazer do inteligente ? bancar o idiota diante de um idiota que banca o inteligente". -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071017/2f7a76c6/attachment-0001.html From alex at ergens.op.het.net Wed Oct 17 13:13:08 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Wed, 17 Oct 2007 13:13:08 +0200 Subject: [rrd-users] rrdtool seg fault using GPRINT In-Reply-To: References: Message-ID: <20071017111308.GA21384@ergens.op.het.net> On Wed, Oct 17, 2007 at 12:02:24AM -0300, Salatiel Filho wrote: > Hi guys , i am trying to use rrdtool but everytime i try to use some GPRINT > function i get segmentation fault. This limits a lot the use of rrdtool. > Could anyone lead me to a solution. I had de debian etch binary 1.2.15 for > arm , and now i have 1.2.23 compiled by myself. Both segfault. The script you show is much to complicated for debugging purposes. Remove elements from your script until the problem doesn't show itself, or until you end up without results. Start by using only one of the DSes, for instance only "a". Maybe the problem disappears if you use less memory; then we may need to hunt down some code problem. Another thing to try: in the trouble tickets I notice a report claiming segfaults when LANG=C. Changing this to anything else (LANG=banana) would fix the problem. Unfortunately this bug was solved as "worksforme", probably because its submitter didn't respond anymore. Try printing less information, such as only "%8.2lf", or only "Current", and so on. Does this make a difference? Try changing the options to printf. And so on. -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/ From ari197516 at yahoo.com Wed Oct 17 13:44:23 2007 From: ari197516 at yahoo.com (ariel mastracchio) Date: Wed, 17 Oct 2007 04:44:23 -0700 (PDT) Subject: [rrd-users] rrdtool seg fault using GPRINT In-Reply-To: Message-ID: <560349.54634.qm@web51004.mail.re2.yahoo.com> Hi, I have segmentation fault too with rrdcgi (rrdgraph) when the .rrd file is created, but empty. (rrdtools 1.2.15, ibm aix 5.2) hugs --- Salatiel Filho wrote: > Hi guys , i am trying to use rrdtool but everytime i > try to use some GPRINT > function i get segmentation fault. This limits a lot > the use of rrdtool. > Could anyone lead me to a solution. I had de debian > etch binary 1.2.15 for > arm , and now i have 1.2.23 compiled by myself. Both > segfault. > > # uname -a > Linux LS-GL7D6 2.6.12.6-arm1 #77 Tue Dec 5 22:23:46 > CST 2006 armv5tejl > GNU/Linux > > > # rrdtool -v > RRDtool 1.2.23 Copyright 1997-2007 by Tobias > Oetiker > Compiled Oct 16 2007 23:04:34 > > > > # /usr/bin/rrdtool graph /tmp/a.png --imgformat=PNG > --start=-86400 > --end=-300 --title="LSPRO - Load Average" --rigid > --base=1000 --height=120 > --width=500 --alt-autoscale-max --lower-limit=0 > --units-exponent=0 > --vertical-label="processes in the run queue" > --slope-mode > DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_load_1min_12.rrd":load_1min:AVERAGE > DEF:b="/usr/share/cacti-0.8.6j/rra/lspro_load_5min_14.rrd":load_5min:AVERAGE > DEF:c="/usr/share/cacti-0.8.6j/rra/lspro_load_15min_13.rrd":load_15min:AVERAGE > CDEF:cdefg=TIME,1192584729,GT,a,a,UN,0,a,IF,IF,TIME,1192584729,GT,b,b,UN,0,b,IF,IF,TIME,1192584729,GT,c,c,UN,0,c,IF,IF,+,+ > AREA:a#EACC00:"1 Minute Average" GPRINT:a:LAST:" > Current\:%8.2lf\n" > AREA:b#EA8F00:"5 Minute Average":STACK > GPRINT:b:LAST:" Current\:%8.2lf\n" > AREA:c#FF0000:"15 Minute Average":STACK > GPRINT:c:LAST:"Current\:%8.2lf\n" > LINE1:cdefg#000000:"Total" > Segmentation fault > > # strace /usr/bin/rrdtool graph /tmp/a.png > --imgformat=PNG --start=-86400 > --end=-300 --title="LSPRO - Load Average" --rigid > --base=1000 --height=120 > --width=500 --alt-autoscale-max --lower-limit=0 > --units-exponent=0 > --vertical-label="processes in the run queue" > --slope-mode > DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_load_1min_12.rrd":load_1min:AVERAGE > DEF:b="/usr/share/cacti-0.8.6j/rra/lspro_load_5min_14.rrd":load_5min:AVERAGE > DEF:c="/usr/share/cacti-0.8.6j/rra/lspro_load_15min_13.rrd":load_15min:AVERAGE > CDEF:cdefg=TIME,1192584729,GT,a,a,UN,0,a,IF,IF,TIME,1192584729,GT,b,b,UN,0,b,IF,IF,TIME,1192584729,GT,c,c,UN,0,c,IF,IF,+,+ > AREA:a#EACC00:"1 Minute Average" GPRINT:a:LAST:" > Current\:%8.2lf\n" > AREA:b#EA8F00:"5 Minute Average":STACK > GPRINT:b:LAST:" Current\:%8.2lf\n" > AREA:c#FF0000:"15 Minute Average":STACK > GPRINT:c:LAST:"Current\:%8.2lf\n" > LINE1:cdefg#000000:"Total" > execve("/usr/bin/rrdtool", ["/usr/bin/rrdtool", > "graph", "/tmp/a.png", > "--imgformat=PNG", "--start=-86400", "--end=-300", > "--title=LSPRO - Load > Average", "--rigid", "--base=1000", "--height=120", > "--width=500", > "--alt-autoscale-max", "--lower-limit=0", > "--units-exponent=0", > "--vertical-label=processes in th"..., > "--slope-mode", ...], [/* 23 vars > */]) = 0 > uname({sys="Linux", node="LS-GL7D6", ...}) = 0 > brk(0) = 0x11000 > access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT > (No such file or > directory) > mmap2(NULL, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = > 0x40015000 > access("/etc/ld.so.preload", R_OK) = -1 ENOENT > (No such file or > directory) > open("/etc/ld.so.cache", O_RDONLY) = 3 > fstat64(3, {st_mode=S_IFREG|0644, st_size=20010, > ...}) = 0 > mmap2(NULL, 20010, PROT_READ, MAP_PRIVATE, 3, 0) = > 0x40016000 > close(3) = 0 > access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT > (No such file or > directory) > open("/usr/lib/librrd.so.2", O_RDONLY) = 3 > read(3, > "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0hV\0\0004"..., > 512) = > 512 > fstat64(3, {st_mode=S_IFREG|0755, st_size=578967, > ...}) = 0 > mmap2(NULL, 311844, PROT_READ|PROT_EXEC, > MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = > 0x4001e000 > mprotect(0x4005d000, 53796, PROT_NONE) = 0 > mmap2(0x40065000, 16384, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x3f) = > 0x40065000 > mmap2(0x40069000, 4644, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = > 0x40069000 > close(3) = 0 > access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT > (No such file or > directory) > open("/usr/lib/libpng12.so.0", O_RDONLY) = 3 > read(3, > "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\0045\0\000"..., > 512) > = 512 > fstat64(3, {st_mode=S_IFREG|0644, st_size=127124, > ...}) = 0 > mmap2(NULL, 158800, PROT_READ|PROT_EXEC, > MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = > 0x4006b000 > mprotect(0x4008a000, 31824, PROT_NONE) = 0 > mmap2(0x40091000, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1e) = > 0x40091000 > close(3) = 0 > access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT > (No such file or > directory) > open("/usr/lib/libz.so.1", O_RDONLY) = 3 > read(3, > "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\254\26\0"..., > 512) = > 512 > fstat64(3, {st_mode=S_IFREG|0644, st_size=84988, > ...}) = 0 > mmap2(NULL, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = > 0x40092000 > mmap2(NULL, 116668, PROT_READ|PROT_EXEC, > MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = > 0x40093000 > mprotect(0x400a8000, 30652, PROT_NONE) = 0 > mmap2(0x400af000, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x14) = > 0x400af000 > close(3) = 0 > access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT > (No such file or > directory) > open("/lib/libm.so.6", O_RDONLY) = 3 > read(3, > "\177ELF\1\1\1a\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\3042\0\000"..., > 512) > = 512 > fstat64(3, {st_mode=S_IFREG|0644, st_size=78836, > ...}) = 0 > mmap2(NULL, 110572, PROT_READ|PROT_EXEC, > MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = > 0x402a6000 > mprotect(0x402b9000, 32748, PROT_NONE) = 0 > mmap2(0x402c0000, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x12) = > 0x402c0000 > close(3) = 0 > mmap2(NULL, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = > 0x402c1000 > mprotect(0x40233000, 8192, PROT_READ) = 0 > mprotect(0x40123000, 4096, PROT_READ) = 0 > mprotect(0x4001c000, 4096, PROT_READ) = 0 > munmap(0x40016000, 20010) = 0 > brk(0) = 0x11000 > brk(0x32000) = 0x32000 > open("/etc/localtime", O_RDONLY) = 3 > fstat64(3, {st_mode=S_IFREG|0644, st_size=276, ...}) > = 0 > fstat64(3, {st_mode=S_IFREG|0644, st_size=276, ...}) > = 0 > mmap2(NULL, 4096, PROT_READ|PROT_WRITE, > MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = > 0x40016000 > read(3, > "TZif\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\3\0\0\0\3\0"..., > 4096) = > 276 > close(3) = 0 > munmap(0x40016000, 4096) = 0 > open("/usr/lib/locale/locale-archive", > O_RDONLY|O_LARGEFILE) = 3 > fstat64(3, {st_mode=S_IFREG|0644, st_size=1449376, > ...}) = 0 > mmap2(NULL, 1449376, PROT_READ, MAP_PRIVATE, 3, 0) = > 0x402c2000 > close(3) = 0 > time(NULL) = 1192589982 > time(NULL) = 1192589982 > stat64("/etc/localtime", {st_mode=S_IFREG|0644, > st_size=276, === message truncated ===> _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From salatiel.filho at gmail.com Wed Oct 17 16:18:06 2007 From: salatiel.filho at gmail.com (Salatiel Filho) Date: Wed, 17 Oct 2007 11:18:06 -0300 Subject: [rrd-users] rrdtool seg fault using GPRINT In-Reply-To: <20071017111308.GA21384@ergens.op.het.net> References: <20071017111308.GA21384@ergens.op.het.net> Message-ID: On 10/17/07, Alex van den Bogaerdt wrote: > On Wed, Oct 17, 2007 at 12:02:24AM -0300, Salatiel Filho wrote: > > > Hi guys , i am trying to use rrdtool but everytime i try to use some GPRINT > > function i get segmentation fault. This limits a lot the use of rrdtool. > > Could anyone lead me to a solution. I had de debian etch binary 1.2.15 for > > arm , and now i have 1.2.23 compiled by myself. Both segfault. > > The script you show is much to complicated for debugging purposes. Sorry. Now i have one simpler. usr/bin/rrdtool graph /tmp/a.png \ --imgformat=PNG \ --start=-86400 \ --end=-300 \ --title="LSPRO - Fan Speed" \ --base=1000 \ --height=120 \ --width=500 \ --alt-autoscale-max \ --lower-limit=0 \ --vertical-label="FAN - RPM" \ --slope-mode \ DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd":fan:AVERAGE \ GPRINT:a:AVERAGE:"RPM\:%8.2lf %s" It still seg faults :/ As a said every time i try to use GPRINT it seg faults. > > Remove elements from your script until the problem doesn't show itself, > or until you end up without results. Start by using only one of the > DSes, for instance only "a". > > > Maybe the problem disappears if you use less memory; then we may need > to hunt down some code problem. > > Another thing to try: in the trouble tickets I notice a report claiming > segfaults when LANG=C. Changing this to anything else (LANG=banana) > would fix the problem. Unfortunately this bug was solved as "worksforme", > probably because its submitter didn't respond anymore. > # locale LANG=en_US.UTF-8 > Try printing less information, such as only "%8.2lf", or only "Current", > and so on. Does this make a difference? Try changing the options to printf. > And so on. > No. :/ > -- > Alex van den Bogaerdt > http://www.vandenbogaerdt.nl/rrdtool/ > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > -- []'s Salatiel "O maior prazer do inteligente ? bancar o idiota diante de um idiota que banca o inteligente". From alex at ergens.op.het.net Wed Oct 17 16:25:53 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Wed, 17 Oct 2007 16:25:53 +0200 Subject: [rrd-users] rrdtool seg fault using GPRINT In-Reply-To: References: <20071017111308.GA21384@ergens.op.het.net> Message-ID: <20071017142553.GA29054@ergens.op.het.net> On Wed, Oct 17, 2007 at 11:18:06AM -0300, Salatiel Filho wrote: > On 10/17/07, Alex van den Bogaerdt wrote: > > On Wed, Oct 17, 2007 at 12:02:24AM -0300, Salatiel Filho wrote: > > > > > Hi guys , i am trying to use rrdtool but everytime i try to use some GPRINT > > > function i get segmentation fault. This limits a lot the use of rrdtool. > > > Could anyone lead me to a solution. I had de debian etch binary 1.2.15 for > > > arm , and now i have 1.2.23 compiled by myself. Both segfault. > > > > The script you show is much to complicated for debugging purposes. > Sorry. Now i have one simpler. > > usr/bin/rrdtool graph /tmp/a.png \ > --imgformat=PNG \ > --start=-86400 \ > --end=-300 \ > --title="LSPRO - Fan Speed" \ > --base=1000 \ > --height=120 \ > --width=500 \ > --alt-autoscale-max \ > --lower-limit=0 \ > --vertical-label="FAN - RPM" \ > --slope-mode \ > DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd":fan:AVERAGE \ > GPRINT:a:AVERAGE:"RPM\:%8.2lf %s" I was thinking of: LANG=x export LANG cd /tmp cp /usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd a.rrd rrdtool graph a.png DEF:a=a.rrd:fan:AVERAGE GPRINT:a:AVERAGE:%lf -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/ From salatiel.filho at gmail.com Wed Oct 17 16:29:34 2007 From: salatiel.filho at gmail.com (Salatiel Filho) Date: Wed, 17 Oct 2007 11:29:34 -0300 Subject: [rrd-users] rrdtool seg fault using GPRINT In-Reply-To: <20071017142553.GA29054@ergens.op.het.net> References: <20071017111308.GA21384@ergens.op.het.net> <20071017142553.GA29054@ergens.op.het.net> Message-ID: On 10/17/07, Alex van den Bogaerdt wrote: > On Wed, Oct 17, 2007 at 11:18:06AM -0300, Salatiel Filho wrote: > > On 10/17/07, Alex van den Bogaerdt wrote: > > > On Wed, Oct 17, 2007 at 12:02:24AM -0300, Salatiel Filho wrote: > > > > > > > Hi guys , i am trying to use rrdtool but everytime i try to use some GPRINT > > > > function i get segmentation fault. This limits a lot the use of rrdtool. > > > > Could anyone lead me to a solution. I had de debian etch binary 1.2.15 for > > > > arm , and now i have 1.2.23 compiled by myself. Both segfault. > > > > > > The script you show is much to complicated for debugging purposes. > > Sorry. Now i have one simpler. > > > > usr/bin/rrdtool graph /tmp/a.png \ > > --imgformat=PNG \ > > --start=-86400 \ > > --end=-300 \ > > --title="LSPRO - Fan Speed" \ > > --base=1000 \ > > --height=120 \ > > --width=500 \ > > --alt-autoscale-max \ > > --lower-limit=0 \ > > --vertical-label="FAN - RPM" \ > > --slope-mode \ > > DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd":fan:AVERAGE \ > > GPRINT:a:AVERAGE:"RPM\:%8.2lf %s" > > I was thinking of: > > LANG=x > export LANG > cd /tmp > cp /usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd a.rrd > rrdtool graph a.png DEF:a=a.rrd:fan:AVERAGE GPRINT:a:AVERAGE:%lf # LANG=x # export LANG # cd /tmp # cp /usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd a.rrd # rrdtool graph a.png DEF:a=a.rrd:fan:AVERAGE GPRINT:a:AVERAGE:%lf Segmentation fault # > > -- > Alex van den Bogaerdt > http://www.vandenbogaerdt.nl/rrdtool/ > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > -- []'s Salatiel "O maior prazer do inteligente ? bancar o idiota diante de um idiota que banca o inteligente". From ml at t-b-o-h.net Wed Oct 17 23:34:22 2007 From: ml at t-b-o-h.net (Tuc at T-B-O-H.NET) Date: Wed, 17 Oct 2007 17:34:22 -0400 (EDT) Subject: [rrd-users] Not supposed to autoscale Message-ID: <200710172134.l9HLYMIT013074@himinbjorg.tucs-beachin-obx-house.com> Hi, I'm trying to understand why a graph we are producing from a program is autoscaling, even though the author specifically put code in NOT to autoscale. When it goes to rrdgraph, it sends : /usr/local/wwwcust/noccgi/mrtg/b-12-ping-1.png --start -36h --units-exponent 0 --color MGRID#000000 --color GRID#000000 --color BACK#F4F4F4 --color CANVAS#FFFFFF --color FONT#000000 --color FRAME#000000 --color ARROW#FF0000 --color SHADEA#C0C0C0 --color SHADEB#808080 --width 400 --height 150 -v Pings -t B-12 Pings (Hourly) --x-grid HOUR:1:DAY:1:HOUR:2:0:%H DEF:in=/usr/local/etc/mrtg/docs/b-12-ping.rrd:ds0:AVERAGE DEF:out=/usr/local/etc/mrtg/docs/b-12-ping.rrd:ds1:AVERAGE DEF:min=/usr/local/etc/mrtg/docs/b-12-ping.rrd:ds0:MAX DEF:mout=/usr/local/etc/mrtg/docs/b-12-ping.rrd:ds1:MAX CDEF:fin=in CDEF:fout=out CDEF:kin=in,60,* CDEF:kout=out,60,* CDEF:tin=in,100,/ CDEF:tout=out,100,/ CDEF:fmin=min CDEF:fmout=mout AREA:fin#00cc00 LINE2:fin#0000FF\l COMMENT:\l GPRINT:in:AVERAGE:Average Pings\: %2.2lf / 10\c GPRINT:in:MAX:Max Pings \: %2.2lf / 10\c GPRINT:in:LAST:Current Pings\: %2.2lf / 10\c Some of the graphs go from 8 to 10, some 6 to 10, others 0 to 10. I understand WHY it would present like that, why go to 0 if the lowest ping was 8... But thats what is wanted (Always seeing 0 to 10). Thanks, Tuc From ml at t-b-o-h.net Thu Oct 18 00:17:33 2007 From: ml at t-b-o-h.net (Tuc at T-B-O-H.NET) Date: Wed, 17 Oct 2007 18:17:33 -0400 (EDT) Subject: [rrd-users] Not supposed to autoscale (SOLVED) Message-ID: <200710172217.l9HMHXiX013905@himinbjorg.tucs-beachin-obx-house.com> Hi, Sorry, found "lower limit". Thanks, Tuc From will at nextlevelinternet.com Thu Oct 18 02:33:13 2007 From: will at nextlevelinternet.com (William Cassis) Date: Wed, 17 Oct 2007 17:33:13 -0700 Subject: [rrd-users] Fetching 5 minute max values from RRDs Message-ID: <88BC6885D33A9D42A1CCB45E8749525E364789@pigeon.sandiego.nextlevelinternet.com> Hello, I'm trying to extract the max values for each 5 minute sample from an rrd file, but I've noticed that the values returned for max are the same as the values for average: rrdtool fetch fastethernet0_2.rrd AVERAGE -r 300 -s 12:00 -e 12:15 ds0 ds1 1192647900: 9.4423089282e+04 3.4053380473e+04 1192648200: 7.0958851492e+04 3.3312245232e+04 1192648500: 4.5601612854e+04 1.5465940745e+04 1192648800: 7.6042505630e+04 1.5733790918e+04 rrdtool fetch fastethernet0_2.rrd MAX -r 300 -s 12:00 -e 12:15 ds0 ds1 1192647900: 9.4423089282e+04 3.4053380473e+04 1192648200: 7.0958851492e+04 3.3312245232e+04 1192648500: 4.5601612854e+04 1.5465940745e+04 1192648800: 7.6042505630e+04 1.5733790918e+04 When I choose a resolution larger than 300 seconds (1800 seconds, for example), the values returned for max and average are different (I'm guessing because it's not using the 5 minute sample section of the rrd): rrdtool fetch fastethernet0_2.rrd MAX -r 1800 -s 12:00 -e 13:00 ds0 ds1 1192649400: 9.6160742247e+04 3.4053380473e+04 1192651200: 7.1751123418e+04 3.3134429366e+04 1192653000: 1.5926465216e+05 1.1399776686e+05 rrdtool fetch fastethernet0_2.rrd AVERAGE -r 1800 -s 12:00 -e 13:00 ds0 ds1 1192649400: 7.2030585524e+04 2.3909270980e+04 1192651200: 5.4137476921e+04 2.3342711488e+04 1192653000: 1.2425490630e+05 6.5662137257e+04 MRTG used to store the average in/out and then the max in/out for each 5 minute sample in the log file. Is there any way to get the 5 minute max from an rrd file? I'm using MRTG version 2.15.2 and rrdtool version 1.2.23 on FreeBSD 6.2. Any help would be greatly appreciated. Thank you! Regards, William Cassis will at nextlevelinternet.com From alex at ergens.op.het.net Thu Oct 18 10:48:53 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Thu, 18 Oct 2007 10:48:53 +0200 Subject: [rrd-users] Fetching 5 minute max values from RRDs In-Reply-To: <88BC6885D33A9D42A1CCB45E8749525E364789@pigeon.sandiego.nextlevelinternet.com> References: <88BC6885D33A9D42A1CCB45E8749525E364789@pigeon.sandiego.nextlevelinternet.com> Message-ID: <20071018084853.GA5836@ergens.op.het.net> On Wed, Oct 17, 2007 at 05:33:13PM -0700, William Cassis wrote: > Hello, > > I'm trying to extract the max values for each 5 minute sample from an > rrd file, but I've noticed that the values returned for max are the same > as the values for average: [snip] > MRTG used to store the average in/out and then the max in/out for each 5 > minute sample in the log file. OK. > Is there any way to get the 5 minute max > from an rrd file? time hh:00, time hh:05 are in the log file (mrtg) or the RRD. You update at time hh:57, hh:02 and hh:07. The resulting rate between hh:57 and hh:02 is 3000. The resulting rate between hh:02 and hh:07 is 0. Normalization then computes: 2 out of 5 minutes @3000, 3 out of 5 @0. This is 2/5*3000 + 3/5*0 = 1200. This one rate is then "consolidated" into an RRA which uses one PDP per CDP: max(1200) = 1200 avg(1200) = 1200 But what you want is possible! set step to 1 second. create an AVERAGE and a MAX RRA, using 300 steps per PDP. This means normalization doesn't alter those values. Consolidation will then store 3000 in the MAX RRA. --step 1 DS:in:COUNTER:600:0:12500000 DS:out:COUNTER:600:0:12500000 RRA:AVERAGE:0.5:1*300:600 RRA:MAX:0.5:1*300:600 RRA:AVERAGE:0.5:6*300:700 RRA:MAX:0.5:6*300:700 RRA:AVERAGE:0.5:24*300:775 RRA:MAX:0.5:24*300:775 RRA:AVERAGE:0.5:288*300:797 RRA:MAX:0.5:288*300:797 You need to fill in 86400 instead of 288*300 and so on. I just wrote it like this for demonstration purposes. Normally you would have 797 rows of 288 steps of 300 seconds each, now you're going to create 797 rows of 86400 steps of 1 second each. The amount of time is the same: 797 * 288 * 300 = 797 * 86400 * 1. The amount of work for RRDtool is not much more (if at all). time hh:00, time hh:05 are in the RRD. You update at time hh:57, hh:02 and hh:07. The resulting rate between hh:57 and hh:02 is 3000. The resulting rate between hh:02 and hh:07 is 0. These rates are valid for all ${step} intervals. Now RRDtool is going to consolidate 300 of such intervals into one PDP: max(3000,3000,3000, ... , 0, 0, 0) = 3000 avg(3000,3000,3000, ... , 0, 0, 0) = 1200 HTH -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/ From vlad at vlad.hu Fri Oct 19 19:02:13 2007 From: vlad at vlad.hu (Pal, Laszlo) Date: Fri, 19 Oct 2007 19:02:13 +0200 (CEST) Subject: [rrd-users] multiple target question for dummies Message-ID: <2836.80.99.239.65.1192813333.squirrel@vlad.hu> Hi, I know this kind of question is a typical RTFM question, but after surfing the manual I decided to ask you :) I have a site where I have two circuits connecting to different PE nodes of an MPLS network. I would like to have the in/out utilization of these two links on one graph, so I can see the utilization difference. Basically I'm using the great routers2.cgi script to create rrd files and making graphs from it. I tried to search an easy solution to use the same tool for this kind of graph but I cannot find anything on this subject Now, maybe I have to create some custom webpage for this kind of graphs, but for this I have to pass the right parameters to rrdtool... Maybe one of you already did the same and can post a sample config for this. Thank you Laszlo Pal From jorge at enlaweba.com Fri Oct 19 22:34:35 2007 From: jorge at enlaweba.com (Jorge Cuevas) Date: Fri, 19 Oct 2007 15:34:35 -0500 Subject: [rrd-users] Help on update made on future date Message-ID: <278BACDA-C822-4611-B0CE-EE0C2DA6282D@enlaweba.com> Hi, I had an issue that caused my server to reboot at 1am. Problem is the hardware clock was set to a future date, Nov 26, 2007. Cacti started polling with that base date and erased a whole month of data and replaced it with NaN values for dates from today till Nov 26th. Is there a way to go back to the current date preserving the past data and erasing any data past today's date? Thanks for your help Jorge From sudhirmd at yahoo.com Sat Oct 20 08:39:06 2007 From: sudhirmd at yahoo.com (Sitaram Dhuri) Date: Fri, 19 Oct 2007 23:39:06 -0700 (PDT) Subject: [rrd-users] [librrdupd.la] Error 1 Message-ID: <53847.41901.qm@web39503.mail.mud.yahoo.com> Hi, while making rrdtool-1.2.23 i got following error on solaris 10 x86 make[2]: *** [librrdupd.la] Error 1 make[2]: Leaving directory `/tmp/pkg/rrdtool-1.2.23/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/tmp/pkg/rrdtool-1.2.23' make: *** [all] Error 2 Regards __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071019/8e013076/attachment.html From gore_jarold at yahoo.com Mon Oct 22 00:47:46 2007 From: gore_jarold at yahoo.com (Gore Jarold) Date: Sun, 21 Oct 2007 15:47:46 -0700 (PDT) Subject: [rrd-users] quick sanity check on my config ... not sure if AVERAGE is appropriate Message-ID: <957294.20572.qm@web63008.mail.re1.yahoo.com> Hello, I keep two ipfw counters of bandwidth - one inbound and one outbound: 10 399000 569950699 count ip from any to any via fxp0 in 11 258417 15855661 count ip from any to any via fxp0 out I am simply trying to create a simple, standard green/blue in/out bandwidth graph, just like MRTG, measured in BITS/s. I am just not sure if I have created the RRD correctly. ----- I created the RRD with: rrdtool create traffic.rrd DS:inbound:COUNTER:600:0:U DS:outbound:COUNTER:600:0:U RRA:AVERAGE:0.5:1:288 (my cron job will check it every five minutes) I update with: rrdtool update traffic.rrd N:`/sbin/ipfw show|grep ^00010 |awk '{print $3}'`:`/sbin/ipfw show|grep ^00011 |awk '{print $3}'` So is this correct ? I am not sure if AVERAGE is the correct aggregate function ? The 'count' number I am grabbing from ipfw never goes down to zero - it is just bigger and bigger every time it checks it (unless the system reboots). And it checks it every five minutes. Also, the 'ipfw count' value I am grabbing is in BYTES, so do I need to be dividing by 8 somewhere here? I know there are fancy scripts for grabbing ipfw, but I am really doing nothing but measuring ipfw count for inbound and outbound, so I should be able to just plug those numbers right in without much complication ... Thanks. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From linux at thehobsons.co.uk Mon Oct 22 08:52:15 2007 From: linux at thehobsons.co.uk (Simon Hobson) Date: Mon, 22 Oct 2007 07:52:15 +0100 Subject: [rrd-users] quick sanity check on my config ... not sure if AVERAGE is appropriate In-Reply-To: <957294.20572.qm@web63008.mail.re1.yahoo.com> References: <957294.20572.qm@web63008.mail.re1.yahoo.com> Message-ID: At 15:47 -0700 21/10/07, Gore Jarold wrote: >Hello, > >I keep two ipfw counters of bandwidth - one inbound >and one outbound: > >10 399000 569950699 count ip from any to any via >fxp0 in >11 258417 15855661 count ip from any to any via >fxp0 out > >I am simply trying to create a simple, standard >green/blue in/out bandwidth graph, just like MRTG, >measured in BITS/s. > >I am just not sure if I have created the RRD >correctly. >I created the RRD with: > >rrdtool create traffic.rrd DS:inbound:COUNTER:600:0:U >DS:outbound:COUNTER:600:0:U RRA:AVERAGE:0.5:1:288 > >(my cron job will check it every five minutes) > >I update with: > >rrdtool update traffic.rrd N:`/sbin/ipfw show|grep >^00010 |awk '{print $3}'`:`/sbin/ipfw show|grep ^00011 >|awk '{print $3}'` > >So is this correct ? I am not sure if AVERAGE is the >correct aggregate function ? This looks reasonable, it's similar to what I do (except I use iptables directly to read the counters). Is average correct ? Well that depends on what you want to graph in the future. Personally I would suggest having MAX as well - you can't derive max from average nor average from max so if you might ever want max (or min for that matter) then store it now and you won't be coming back in 6 months asking how to get it ! But you WILL be wanting average once your data gets consolidated, using max gives a very distorted view of historical traffic. You have only defined one RRA there, keeping 288x5min samples - if you want to show historical traffic then you'll need additional lower resolution RRAs to keep the data. >The 'count' number I am grabbing from ipfw never goes >down to zero - it is just bigger and bigger every time >it checks it (unless the system reboots). And it >checks it every five minutes. Well if you leave it long enough it will wrap around to 0 ! I check my counters every minute and let rrd take care of the numbers - unless you take a lot of care, you won't be checking exactly on rrd's sample boundaries and it'll be doing some adjustments anyway. It also means that if I reload the firewall config (which resets the counters) then I don't lose a whole sample period and it's less of a blip in the graphs. >Also, the 'ipfw count' value I am grabbing is in >BYTES, so do I need to be dividing by 8 somewhere >here? It's multiply by 8 to get from bytes/s to bits/s. You can do it as you store the data, or when you do the graphs - for such simple graphs there is negligable overhead in doing it when you draw the graphs. From alex at ergens.op.het.net Mon Oct 22 11:43:53 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Mon, 22 Oct 2007 11:43:53 +0200 Subject: [rrd-users] quick sanity check on my config ... not sure if AVERAGE is appropriate In-Reply-To: <957294.20572.qm@web63008.mail.re1.yahoo.com> References: <957294.20572.qm@web63008.mail.re1.yahoo.com> Message-ID: <20071022094352.GA11701@ergens.op.het.net> On Sun, Oct 21, 2007 at 03:47:46PM -0700, Gore Jarold wrote: > I created the RRD with: > > rrdtool create traffic.rrd DS:inbound:COUNTER:600:0:U > DS:outbound:COUNTER:600:0:U RRA:AVERAGE:0.5:1:288 This is an unlimited rate. If your counter wraps to zero, that's probably fine as it will most likely do so at 2^32 or 2^32. However, if your counter resets to zero, and if you don't tell rrdtool, then rrdtool will think it is a wrap. For instance: The counter reached 12,345,678 The system reboots. The counter is now 0 RRDtool does: counter increase = 0-12345678 this is negative --> assume counter wrap counter increase = 2^32+0-12345678 divide by 300 resulting rate: 14,275,405.39333 Bps which is 114,203,243.14666 bps. This is not the reality and, depending on your network interface speed, it may even go undetected. But even for 10Mbps this could work: Counter value: 4,000,000,000 Reboot. Counter value: 0 counter increase = 0 - 4000000000 this is negative --> assume counter wrap counter increase = 2^32+0-4000000000 divide by 300 resulting rate: 983,224 Bps which is 7,865,792 bps. You need to: a) set a sane upper limit and b) update with "U" when the counters are reset or use DERIVE instead of COUNTER. Keep the lower limit as zero. True counter wraps are no longer compensated for, but also you won't have a problem with resets HTH -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/ From pgrill at tesnet-group.de Wed Oct 24 21:04:25 2007 From: pgrill at tesnet-group.de (Patrik Grill) Date: Wed, 24 Oct 2007 21:04:25 +0200 Subject: [rrd-users] Question concerning Archives and creating graphs Message-ID: <001701c81670$b4c4be40$1e4e3ac0$@de> Hi all together, i'm using JRobin but i guess the problem is the same - so I transfer my problem from jrobin code to rrdtool code: I've created a rrd database with one datasource and two archives, both of the consolidation function AVERAGE, but with different steps and rows, like this: DS:input:GAUGE:600:0:U RRA:AVERAGE:0.5:6:700 RRA:AVERAGE:0.5:2:300 Now I want to generate a graph for each archive, but with: DEF:in=traffic.rrd:input:AVERAGE I always get the first archive plotted (RRA:AVERAGE:0.5:6:700)--- how can I plot the second archive (RRA:AVERAGE:0.5:2:300) ? Or am I complete wrong in my understanding of the RRDTool ? Thanks for your help, Best regards, Patrik -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071024/608b5230/attachment.html From Mandy_Hughes at reyrey.com Wed Oct 24 21:27:04 2007 From: Mandy_Hughes at reyrey.com (Hughes, Mandy F) Date: Wed, 24 Oct 2007 15:27:04 -0400 Subject: [rrd-users] Getting .rrd no such file or directory when trying to graph in Cacti Message-ID: <7752B110D1ABB0449E4D171CC202684503C55EE5@OH18CL07VE1.ad.reyrey.com> I have recently integrated Cacti to open source Groundwork. After adding a device in Cacti and creating few graphs for the device, the graphs are not displayed. So I turned on the debug mode and got the following messages. Not sure if this is permission issue. As you can see I am unfamiliar with Cacti as well as RRDTool. Any help on this is deeply appreciated. Thank you. The setting for /usr/local/groundwork/apache2/htdocs/cacti/rra is: drwxr-xr-x 2 cacti cacti 4096 Oct 24 14:22 rra Debug messages: RRDTool Command: /usr/local/groundwork/bin/rrdtool graph - \ --imgformat=PNG \ --start=-86400 \ --end=-300 \ --title="lnmmgttm001 - Traffic - 10.206.1.20 (eth0)" \ --rigid \ --base=1000 \ --height=120 \ --width=500 \ --alt-autoscale-max \ --lower-limit=0 \ --vertical-label="bits per second" \ --slope-mode \ DEF:a="/usr/local/groundwork/apache2/htdocs/cacti/rra/lnmmgttm001_traffi c_in_17.rrd":traffic_in:AVERAGE \ DEF:b="/usr/local/groundwork/apache2/htdocs/cacti/rra/lnmmgttm001_traffi c_in_17.rrd":traffic_out:AVERAGE \ CDEF:cdefa=a,8,* \ CDEF:cdefe=b,8,* \ AREA:cdefa#00CF00:"Inbound" \ GPRINT:cdefa:LAST:" Current\:%8.2lf %s" \ GPRINT:cdefa:AVERAGE:"Average\:%8.2lf %s" \ GPRINT:cdefa:MAX:"Maximum\:%8.2lf %s\n" \ LINE1:cdefe#002A97:"Outbound" \ GPRINT:cdefe:LAST:"Current\:%8.2lf %s" \ GPRINT:cdefe:AVERAGE:"Average\:%8.2lf %s" \ GPRINT:cdefe:MAX:"Maximum\:%8.2lf %s" RRDTool Says: ERROR: opening '/usr/local/groundwork/apache2/htdocs/cacti/rra/lnmmgttm001_traffic_in_1 7.rrd': No such file or directory RRDTool Command: /usr/local/groundwork/bin/rrdtool graph - \ --imgformat=PNG \ --start=-86400 \ --end=-300 \ --title="lnmmgttm001 - Traffic - 10.206.1.20 (eth0)" \ --rigid \ --base=1000 \ --height=120 \ --width=500 \ --alt-autoscale-max \ --lower-limit=0 \ --vertical-label="bits per second" \ --slope-mode \ DEF:a="/usr/local/groundwork/apache2/htdocs/cacti/rra/lnmmgttm001_traffi c_in_17.rrd":traffic_in:AVERAGE \ DEF:b="/usr/local/groundwork/apache2/htdocs/cacti/rra/lnmmgttm001_traffi c_in_17.rrd":traffic_out:AVERAGE \ CDEF:cdefa=a,8,* \ CDEF:cdefe=b,8,* \ AREA:cdefa#00CF00:"Inbound" \ GPRINT:cdefa:LAST:" Current\:%8.2lf %s" \ GPRINT:cdefa:AVERAGE:"Average\:%8.2lf %s" \ GPRINT:cdefa:MAX:"Maximum\:%8.2lf %s\n" \ LINE1:cdefe#002A97:"Outbound" \ GPRINT:cdefe:LAST:"Current\:%8.2lf %s" \ GPRINT:cdefe:AVERAGE:"Average\:%8.2lf %s" \ GPRINT:cdefe:MAX:"Maximum\:%8.2lf %s" RRDTool Says: ERROR: opening '/usr/local/groundwork/apache2/htdocs/cacti/rra/lnmmgttm001_traffic_in_1 7.rrd': No such file or directory -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071024/fd0ee37a/attachment-0001.html From linux at thehobsons.co.uk Wed Oct 24 22:53:30 2007 From: linux at thehobsons.co.uk (Simon Hobson) Date: Wed, 24 Oct 2007 21:53:30 +0100 Subject: [rrd-users] Question concerning Archives and creating graphs In-Reply-To: <001701c81670$b4c4be40$1e4e3ac0$@de> References: <001701c81670$b4c4be40$1e4e3ac0$@de> Message-ID: Patrik Grill wrote: >I've created a rrd database with one datasource and two archives, >both of the consolidation function AVERAGE, but with different steps >and rows, like this: > >DS:input:GAUGE:600:0:U >RRA:AVERAGE:0.5:6:700 >RRA:AVERAGE:0.5:2:300 > >Now I want to generate a graph for each archive, but with: > >DEF:in=traffic.rrd:input:AVERAGE > >I always get the first archive plotted (RRA:AVERAGE:0.5:6:700)--- >how can I plot the second archive (RRA:AVERAGE:0.5:2:300) ? > >Or am I complete wrong in my understanding of the RRDTool ? You have the wrong understanding. The tools will use the best fit RRA that is available. So if you plot a graph that exceeds the time range encompassed in the second RRA, the first will be used. Also (not applicable in this case), if the start/end times fit one better than another then this will have an effect. You should specify the start and end times as exact times which are multiples of your aggregate period rather than just "now" and "now-x" - eg "now - now % step" as end and "end - n x step" as start. If the steps weren't multiples of each other - eg you had used 2 and 5, then there would be times where the x5 periods would give a better match to a time than the x2 periods - eg from 5 to 15s would be exactly 2x5, but be 1/2x2 plus 4x2 plus 1/2x2. So taking your figures, always make sure your start and end times are multiples of 20 minutes (600x2 s), and end-start is less than 100 hours (600x2x300 s) - then RRD tools should use the higher resolution RRA. From salatiel.filho at gmail.com Wed Oct 24 22:59:26 2007 From: salatiel.filho at gmail.com (Salatiel Filho) Date: Wed, 24 Oct 2007 17:59:26 -0300 Subject: [rrd-users] rrdtool seg fault using GPRINT In-Reply-To: References: <20071017111308.GA21384@ergens.op.het.net> <20071017142553.GA29054@ergens.op.het.net> Message-ID: no more ideas ? On 10/17/07, Salatiel Filho wrote: > On 10/17/07, Alex van den Bogaerdt wrote: > > On Wed, Oct 17, 2007 at 11:18:06AM -0300, Salatiel Filho wrote: > > > On 10/17/07, Alex van den Bogaerdt wrote: > > > > On Wed, Oct 17, 2007 at 12:02:24AM -0300, Salatiel Filho wrote: > > > > > > > > > Hi guys , i am trying to use rrdtool but everytime i try to use some GPRINT > > > > > function i get segmentation fault. This limits a lot the use of rrdtool. > > > > > Could anyone lead me to a solution. I had de debian etch binary 1.2.15 for > > > > > arm , and now i have 1.2.23 compiled by myself. Both segfault. > > > > > > > > The script you show is much to complicated for debugging purposes. > > > Sorry. Now i have one simpler. > > > > > > usr/bin/rrdtool graph /tmp/a.png \ > > > --imgformat=PNG \ > > > --start=-86400 \ > > > --end=-300 \ > > > --title="LSPRO - Fan Speed" \ > > > --base=1000 \ > > > --height=120 \ > > > --width=500 \ > > > --alt-autoscale-max \ > > > --lower-limit=0 \ > > > --vertical-label="FAN - RPM" \ > > > --slope-mode \ > > > DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd":fan:AVERAGE \ > > > GPRINT:a:AVERAGE:"RPM\:%8.2lf %s" > > > > I was thinking of: > > > > LANG=x > > export LANG > > cd /tmp > > cp /usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd a.rrd > > rrdtool graph a.png DEF:a=a.rrd:fan:AVERAGE GPRINT:a:AVERAGE:%lf > > # LANG=x > # export LANG > # cd /tmp > # cp /usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd a.rrd > # rrdtool graph a.png DEF:a=a.rrd:fan:AVERAGE GPRINT:a:AVERAGE:%lf > Segmentation fault > # > > > > -- > > Alex van den Bogaerdt > > http://www.vandenbogaerdt.nl/rrdtool/ > > > > _______________________________________________ > > rrd-users mailing list > > rrd-users at lists.oetiker.ch > > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > > > > -- > []'s > Salatiel > > "O maior prazer do inteligente ? bancar o idiota > diante de um idiota que banca o inteligente". > -- []'s Salatiel "O maior prazer do inteligente ? bancar o idiota diante de um idiota que banca o inteligente". From geoff.garside at openhosting.co.uk Thu Oct 25 14:37:00 2007 From: geoff.garside at openhosting.co.uk (Geoff Garside) Date: Thu, 25 Oct 2007 13:37:00 +0100 Subject: [rrd-users] xport vdef's Message-ID: Is there any way, or any plans to support, the exporting of vdefs from an rrd. The main thing I need is to be able to export 95th percentile information from an rrd over a set time period. Currently I am having to generate an SVG file as output then parse that for the 95th percentile information. This solution just feels like a cludge though. Any other ideas how to do this? Regards, Geoff Garside Open Hosting Ltd -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071025/b11f658a/attachment.html From tobi at oetiker.ch Thu Oct 25 15:12:18 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Thu, 25 Oct 2007 15:12:18 +0200 (CEST) Subject: [rrd-users] xport vdef's In-Reply-To: References: Message-ID: Hi Geoff, well you can use graph and PRINT for this ... just make sure that you use the right scalingk, so that the individual 5 minute measurments do not get consolidated prior the the 95% calculation ... the trick is the use PRINT and not GRAPHING commands at all ... cheers tobi Today Geoff Garside wrote: > Is there any way, or any plans to support, the exporting of vdefs from an rrd. > The main thing I need is to be able to export 95th percentile information from > an rrd over a set time period. Currently I am having to generate an SVG file > as output then parse that for the 95th percentile information. This solution > just feels like a cludge though. > > Any other ideas how to do this? > > Regards, > Geoff Garside > > Open Hosting Ltd > -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From geoff.garside at openhosting.co.uk Thu Oct 25 16:05:24 2007 From: geoff.garside at openhosting.co.uk (Geoff Garside) Date: Thu, 25 Oct 2007 15:05:24 +0100 Subject: [rrd-users] xport vdef's In-Reply-To: References: Message-ID: <76055562-0F13-4B64-998B-C075602C5325@openhosting.co.uk> Thanks Tobi that works fine with the cli. I get a segfault when trying it through the Ruby bindings though, but I can just call the cli from Ruby so should be fine. Thanks again, Geoff Garside Open Hosting Ltd On 25 Oct 2007, at 14:12, Tobias Oetiker wrote: > Hi Geoff, > > well you can use > > graph and PRINT for this ... just make sure that you use the right > scalingk, so that the individual 5 minute measurments do not get > consolidated prior the the 95% calculation ... > > the trick is the use PRINT and not GRAPHING commands at all ... > > > cheers > tobi > > Today Geoff Garside wrote: > >> Is there any way, or any plans to support, the exporting of vdefs >> from an rrd. >> The main thing I need is to be able to export 95th percentile >> information from >> an rrd over a set time period. Currently I am having to generate >> an SVG file >> as output then parse that for the 95th percentile information. >> This solution >> just feels like a cludge though. >> >> Any other ideas how to do this? >> >> Regards, >> Geoff Garside >> >> Open Hosting Ltd >> > > -- > Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten > http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071025/01e527d3/attachment.html From salatiel.filho at gmail.com Thu Oct 25 22:34:39 2007 From: salatiel.filho at gmail.com (Salatiel Filho) Date: Thu, 25 Oct 2007 17:34:39 -0300 Subject: [rrd-users] ERROR: unknown freetype pixel mode: 0 Message-ID: I just cross-compiled a static version with gcc soft-float to run on my arm box , the compilation was successfully but when i try to graph on the target box i get: # ./rrdtool graph /tmp/a.png --imgformat=PNG --start=-86400 --end=-300 --title="LSPRO - Fan Speed" --base=1000 --height=120 --width=500 --alt-autoscale-max --lower-limit=0 --vertical-label="FAN - RPM" --slope-mode DEF:a="/usr/share/cacti-0.8.6j/rra/lspro_fan_27.rrd":fan:AVERAGE GPRINT:a:AVERAGE:"RPM\:%8.2lf %s" 597x199 ERROR: unknown freetype pixel mode: 0 any ideas how to fix this ? -- []'s Salatiel "O maior prazer do inteligente ? bancar o idiota diante de um idiota que banca o inteligente". From wernli at in2p3.fr Fri Oct 26 09:47:38 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Fri, 26 Oct 2007 09:47:38 +0200 Subject: [rrd-users] I/O Read overkill Message-ID: <20071026074738.GZ31429@ccswiss.in2p3.fr> Hi, Yesterday I migrated my dual-xeon 4GB RAID1 rrd to a quad-opteron 8GB RAID10 server, in the hope to reduce the load. I am managing 18k RRDs, each being updated randomly every ~10minutes over ssh, which makes roughly 30 file updates/s. I had many performance problems with the new server, which I was able to manage (=server stays alive) using xfs instead of ext3, raid10 instead of raid5. Underlying disk technology is the same (SCSI 10k drives atatched to LSI 1030 controller), but box brand is different. Graphs speak for themselves. http://swissunix.free.fr/rrd/rrd_perf.png Before: 294% iowait 4.9MB/s read (hardware raid1 hidden to linux) 1.1MB/s write 4.6 loadavg After: 318% iowait 57.9MB/s read (linux raid10 -> values have to be div by 2) 2.3MB/s write 8.9 loadavg tests made with rrdtool 1.2.10 and 1.2.19. Now where the heck do those 57.9/2 = 28.95MB/s READ come from??? I do not understand how the Read IOs were multiplied by a factor of 6!!! From wernli at in2p3.fr Fri Oct 26 13:38:47 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Fri, 26 Oct 2007 13:38:47 +0200 Subject: [rrd-users] I/O Read overkill In-Reply-To: <20071026074738.GZ31429@ccswiss.in2p3.fr> References: <20071026074738.GZ31429@ccswiss.in2p3.fr> Message-ID: <20071026113847.GA19216@ccswiss.in2p3.fr> On Fri, Oct 26, 2007 at 09:47:38AM +0200, Fabien Wernli wrote: > Before: 294% iowait > 4.9MB/s read (hardware raid1 hidden to linux) > 1.1MB/s write > 4.6 loadavg > > After: 318% iowait > 57.9MB/s read (linux raid10 -> values have to be div by 2) > 2.3MB/s write > 8.9 loadavg > > tests made with rrdtool 1.2.10 and 1.2.19. I forgot to mention that old server was a 2.4 32bit kernel, new one being a 2.6 64bit. also, I just tested rrdtool 1.3beta (w/mmap) and my problem seems to disappear totally. I changed "use lib" in update script to use rrdtool 1.3b at 12:25 (see graph). http://swissunix.free.fr/rrd/rrd_perf_1.3beta.png Now I don't understand why on earth it used to work so well on the old server and rrdtool 1.2.10 ... From jfesler at gigo.com Fri Oct 26 14:53:33 2007 From: jfesler at gigo.com (Jason Fesler) Date: Fri, 26 Oct 2007 05:53:33 -0700 (PDT) Subject: [rrd-users] I/O Read overkill In-Reply-To: <20071026113847.GA19216@ccswiss.in2p3.fr> References: <20071026074738.GZ31429@ccswiss.in2p3.fr> <20071026113847.GA19216@ccswiss.in2p3.fr> Message-ID: <20071026055055.X93855@goat.gigo.com> > Now I don't understand why on earth it used to work so well on the old > server and rrdtool 1.2.10 ... I wonder if it is the defaults for readahead. "man blockdev" and search for readahead. You might read what your defaults were on the old kernel; and temporarilly apply them to your drives under the new kernel, and see what happens with 1.2.10. From paul.rimmer at gmail.com Fri Oct 26 23:13:45 2007 From: paul.rimmer at gmail.com (Paul Rimmer) Date: Fri, 26 Oct 2007 15:13:45 -0600 Subject: [rrd-users] Need help with a CDEF/AREA setup Message-ID: Hi, I've been using rrdtool to graph temps around my house for a couple of months now. I setup a CDEF that shows night vs. day based on sunrise and sunset temps. Its been working great however now that the below zero have arrived, I notice that the AREA does not cross the x-axis. You can see what I mean here: http://ronin-tech.com/Content/pid=27.html The script I use to generate the graphs is here: http://ronin-tech.com/Content/pid=26.html#graphtemp The line that defines the CDEF used for night is: CDEF:night=LTIME,86400,%,$1,LT,INF,LTIME,86400,%,$2,GT,INF,UNKN,Basement,*,IF,IF \ Any idea how I can change this so that the night AREA covers both sides of the x-axis? Cheers, Paul -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071026/ac28b662/attachment.html From erik at oezie.org Sat Oct 27 11:33:29 2007 From: erik at oezie.org (Erik de Mare) Date: Sat, 27 Oct 2007 11:33:29 +0200 Subject: [rrd-users] Need help with a CDEF/AREA setup In-Reply-To: References: Message-ID: <472305E9.1000301@oezie.org> Hi, INF only graphs from 0 to infinity, so negative values will never be covered. you can add an other CDEF for the negative part with NEGINF. http://oss.oetiker.ch/rrdtool/doc/rrdgraph_rpn.en.html INF, NEGINF Pushes a positive or negative infinite value on the stack. When such a value is graphed, it appears at the top or bottom of the graph, no matter what the actual value on the y-axis is. so your other CDEF would be like: CDEF:night=LTIME,86400,%,$1,LT,NEGINF,LTIME,86400,%,$2,GT,NEGINF,UNKN,Basement,*,IF,IF Mvg, Erik Paul Rimmer schreef: > Hi, > > The line that defines the CDEF used for night is: > > > CDEF:night=LTIME,86400,%,$1,LT,INF,LTIME,86400,%,$2,GT,INF,UNKN,Basement,*,IF,IF > \ > Cheers, > Paul > > > ------------------------------------------------------------------------ > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users From as at modeemi.fi Sat Oct 27 19:27:45 2007 From: as at modeemi.fi (Anssi Saari) Date: Sat, 27 Oct 2007 20:27:45 +0300 Subject: [rrd-users] Putting a degree sign in an SVG? Message-ID: <20071027172745.GB7575@jolt.modeemi.cs.tut.fi> I've been playing with generating SVGs with rrdgraph recently, but I can't seem to be able to include a degree sign in an SVG graph. I'd like to have vertical label like -v "Temperature (?C)". So far I've tried a literal ISO-8859-1 and UTF-8 in the script, but that seems to generate bad XML, or at least Firefox doesn't like it. That works for PNGs though. Enc-Erature (&#-1374389534;���P!��� It's also interesting that Temperature becomes Enc-Erature... My other idea was using the XML notatation i.e. ° but unfortunately that's replaced with &#x00b0; so I end up with the literal string ° in the graph instead of the degree sign... So, any ideas? Using RRDtool 1.2.15 in Debian Etch. Anssi From tobi at oetiker.ch Sun Oct 28 00:18:43 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Sun, 28 Oct 2007 00:18:43 +0200 (CEST) Subject: [rrd-users] Putting a degree sign in an SVG? In-Reply-To: <20071027172745.GB7575@jolt.modeemi.cs.tut.fi> References: <20071027172745.GB7575@jolt.modeemi.cs.tut.fi> Message-ID: Hi Anssi, you may want to try rrdtool 1.3beta ... it uses cairo to generate the svg ... this should fix this problem I guess cheers tobi Yesterday Anssi Saari wrote: > > I've been playing with generating SVGs with rrdgraph recently, but I > can't seem to be able to include a degree sign in an SVG graph. I'd like > to have vertical label like -v "Temperature (?C)". So far I've tried a > literal ISO-8859-1 and UTF-8 in the script, but that seems to generate > bad XML, or at least Firefox doesn't like it. That works for PNGs though. > > Enc-Erature > (&#-1374389534;���P!��� > > It's also interesting that Temperature becomes Enc-Erature... > > My other idea was using the XML notatation i.e. ° but > unfortunately that's replaced with &#x00b0; so I end up with the > literal string ° in the graph instead of the degree sign... > > So, any ideas? Using RRDtool 1.2.15 in Debian Etch. > > Anssi > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From wernli at in2p3.fr Sun Oct 28 10:09:06 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Sun, 28 Oct 2007 10:09:06 +0100 Subject: [rrd-users] Putting a degree sign in an SVG? In-Reply-To: References: <20071027172745.GB7575@jolt.modeemi.cs.tut.fi> Message-ID: <20071028090906.GG3044@mail.faxm0dem.org> On Sun, Oct 28, 2007 at 12:18:43AM +0200, Tobias Oetiker wrote: > you may want to try rrdtool 1.3beta ... > > it uses cairo to generate the svg ... this should fix this problem and it'll generate fewer IOs to draw the symbol ;-) From wernli at in2p3.fr Sun Oct 28 11:00:10 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Sun, 28 Oct 2007 11:00:10 +0100 Subject: [rrd-users] I/O Read overkill In-Reply-To: <20071026055055.X93855@goat.gigo.com> References: <20071026074738.GZ31429@ccswiss.in2p3.fr> <20071026113847.GA19216@ccswiss.in2p3.fr> <20071026055055.X93855@goat.gigo.com> Message-ID: <20071028100010.GA32644@mail.faxm0dem.org> Hi, On Fri, Oct 26, 2007 at 05:53:33AM -0700, Jason Fesler wrote: > I wonder if it is the defaults for readahead. "man blockdev" and search > for readahead. You might read what your defaults were on the old kernel; > and temporarilly apply them to your drives under the new kernel, and see > what happens with 1.2.10. yeah I was planning on doing that, however now my production system is up and running again so... ;) Here are the values before (RRDs on sda): # /sbin/blockdev --getra /dev/sda 120 and after (RRDs on md10 raid1+0): # /sbin/blockdev --getra /dev/sd[b-e] /dev/md[01] /dev/md10 256 256 256 256 256 256 512 (siectorsiwe is 512 in both cases) So I went from 120 to 512. Does that mean 4.266 times more data was being read off the drive? This doesn't account for the difference though, as I went from average read of 730KB/s to 36.2MB/s (sorry my previous values were wrong I mistook Avg for Max), which is a factor of more than 50(!). So unless the response is non-linear, I guess I have to look elsewhere... I still suspect a 64bit issue, what do you think? Now that I use rrdtool 1.2.99999[...] the read value dropped even further: 84KB/s, which is consistent with the announcement by the author (10x more updates per second). From jfesler at gigo.com Sun Oct 28 15:34:47 2007 From: jfesler at gigo.com (Jason Fesler) Date: Sun, 28 Oct 2007 07:34:47 -0700 (PDT) Subject: [rrd-users] I/O Read overkill In-Reply-To: <20071028100010.GA32644@mail.faxm0dem.org> References: <20071026074738.GZ31429@ccswiss.in2p3.fr> <20071026113847.GA19216@ccswiss.in2p3.fr> <20071026055055.X93855@goat.gigo.com> <20071028100010.GA32644@mail.faxm0dem.org> Message-ID: <20071028073259.N93855@goat.gigo.com> > So I went from 120 to 512. Does that mean 4.266 times more data was being read > off the drive? Yes, but only when the OS decided you were doing sequential enough reads to gamble doing so. > Now that I use rrdtool 1.2.99999[...] the read value dropped even further: > 84KB/s, which is consistent with the announcement by the author (10x more > updates per second). I haven't gone to this yet (I'm no longer responsible for the ~250 ide disks running rrd..) but it it sure does sound promising. From anssi.saari at modeemi.fi Sun Oct 28 17:19:26 2007 From: anssi.saari at modeemi.fi (Anssi Saari) Date: Sun, 28 Oct 2007 18:19:26 +0200 Subject: [rrd-users] Putting a degree sign in an SVG? In-Reply-To: References: <20071027172745.GB7575@jolt.modeemi.cs.tut.fi> Message-ID: <20071028161926.GB32105@jolt.modeemi.cs.tut.fi> On Sun, Oct 28, 2007 at 12:18:43AM +0200, Tobias Oetiker wrote: > Hi Anssi, > > you may want to try rrdtool 1.3beta ... > > it uses cairo to generate the svg ... this should fix this problem > I guess Uhm, no. Or it might, but since I can't make out any of the text, I can't tell :) Other downsides apart from the unreadable text is that the 1.3beta generates much larger SVGs that really bog down Firefox, both filesize and when rendered by Firefox. I put up a pair of examples, both generated with the rrdcgi script below. http://www.modeemi.fi/~as/ctxt_new.svg was generated by 1.3beta, http://www.modeemi.fi/~as/ctxt_old.svg was generated by 1.2.15. The filesize increase in this one is huge, over 10x. I notice there are no text elements in the SVG generated by 1.3beta. #! /usr/bin/rrdcgi Simple RRDTool SVG graph

Simple RRDTool SVG graph

From wernli at in2p3.fr Sun Oct 28 18:11:17 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Sun, 28 Oct 2007 18:11:17 +0100 Subject: [rrd-users] Putting a degree sign in an SVG? In-Reply-To: <20071028161926.GB32105@jolt.modeemi.cs.tut.fi> References: <20071027172745.GB7575@jolt.modeemi.cs.tut.fi> <20071028161926.GB32105@jolt.modeemi.cs.tut.fi> Message-ID: <20071028171117.GF32644@mail.faxm0dem.org> On Sun, Oct 28, 2007 at 06:19:26PM +0200, Anssi Saari wrote: > http://www.modeemi.fi/~as/ctxt_new.svg was generated by 1.3beta, Opera doesn't like it either BTW From ephedrine at rogers.com Mon Oct 29 03:24:50 2007 From: ephedrine at rogers.com (Lex) Date: Sun, 28 Oct 2007 22:24:50 -0400 Subject: [rrd-users] Getting Started with RRD Message-ID: <000c01c819d2$e2cbe920$3201a8c0@ONE> Hello RRDers, I've a question or two about getting started, and appreciate any time you can offer! I have a system, that periodically reports server load, HDD load as well as a few other stats...to get started however I'd like to make a simple graph that plots server load, and ping response time. In a tabular format, my data would look something like this: unix timestamp ping-to response time cpu load 1193624363 0.02 0.1 1193624363 0.001 0.05 Is it possible to plot both of these bits of information on a same RRDTool graph? Thanks! Alex -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.oetiker.ch/pipermail/rrd-users/attachments/20071028/0f45416b/attachment.html From Rene at deinheld.de Mon Oct 29 11:35:46 2007 From: Rene at deinheld.de (Rene Fritzsche) Date: Mon, 29 Oct 2007 11:35:46 +0100 Subject: [rrd-users] Getting Started with RRD Message-ID: <1840586889@web.de> Hi, > I have a system, that periodically reports server load, HDD load as well as a few other stats...to get started however I'd like to make a simple graph that plots server load, and ping response time. [...] > unix timestamp ping-to response time cpu load > 1193624363 0.02 0.1 > 1193624363 0.001 0.05 > Is it possible to plot both of these bits of information on a same RRDTool graph? nope, unless you change the interval of value-readings to a minimum of 1 second before trying to create a rrd out of these informations. ;-) lets assume you check your server every minute (60s), you'd have to create a rrd with some lines like this: rrdtool create serverstats.rrd --step 60 \ DS:ping:ABSOLUTE:120:0:U \ DS:load:ABSOLUTE:120:0:U \ [and then the needed RRA-Definitions...] but you should check the rrdtool-webpage for more detailed informations and documentation how to use rrdtool at all. (www.rrdtool.org) HTH Rene __________________________________________________________________________ Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach! Mehr Infos unter http://produkte.web.de/club/?mc=021131 From vlad at vlad.hu Mon Oct 29 13:02:37 2007 From: vlad at vlad.hu (Pal, Laszlo) Date: Mon, 29 Oct 2007 13:02:37 +0100 (CET) Subject: [rrd-users] [Fwd: Re: Getting Started with RRD] Message-ID: <52334.80.99.239.65.1193659357.squirrel@vlad.hu> ---------------------------- Original Message ---------------------------- Subject: Re: [rrd-users] Getting Started with RRD From: "Pal, Laszlo" Date: Mon, October 29, 2007 1:01 pm To: "Rene Fritzsche" -------------------------------------------------------------------------- I've a very similar issue. I've my data in two separate rrd file (these are router if traffic). The interval are the same (rrd created by router2.cgi) and I have two values for each 5 minutes (in/out) how can I put the data on one graph? I would see a graph which shows in/out values for both circuits (maybe different colours) Thank you Laszlo >......< Rene Fritzsche<......> > Hi, > >> I have a system, that periodically reports server load, HDD load as >> well as a few other stats...to get started however I'd like to make >> a simple graph that plots server load, and ping response time. > [...] >> unix timestamp ping-to response time cpu load >> 1193624363 0.02 0.1 >> 1193624363 0.001 0.05 >> Is it possible to plot both of these bits of information on a same >> RRDTool graph? > > nope, unless you change the interval of value-readings to a minimum of 1 > second > before trying to create a rrd out of these informations. ;-) > > lets assume you check your server every minute (60s), you'd have to create > a rrd > with some lines like this: > > rrdtool create serverstats.rrd --step 60 \ > DS:ping:ABSOLUTE:120:0:U \ > DS:load:ABSOLUTE:120:0:U \ > [and then the needed RRA-Definitions...] > > but you should check the rrdtool-webpage for more detailed informations > and documentation > how to use rrdtool at all. (www.rrdtool.org) > > HTH > Rene > > > > __________________________________________________________________________ > Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach! > Mehr Infos unter http://produkte.web.de/club/?mc=021131 > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > From henrik-rrdusers at hswn.dk Mon Oct 29 13:32:37 2007 From: henrik-rrdusers at hswn.dk (Henrik Stoerner) Date: Mon, 29 Oct 2007 13:32:37 +0100 Subject: [rrd-users] [Fwd: Re: Getting Started with RRD] In-Reply-To: <52334.80.99.239.65.1193659357.squirrel@vlad.hu> References: <52334.80.99.239.65.1193659357.squirrel@vlad.hu> Message-ID: <20071029123237.GA15414@hswn.dk> > I've a very similar issue. I've my data in two separate rrd file (these > are router if traffic). The interval are the same (rrd created by > router2.cgi) and I have two values for each 5 minutes (in/out) Assuming you have two RRD files, file1.rrd and file2.rrd, and each of these have an "inbytes" and "outbytes" dataset, then you can create a graph with data from both files like this: rrdtool graph somefile.png -s e-48h \ "DEF:in1=file1.rrd:inbytes:AVERAGE" \ "DEF:in2=file2.rrd:inbytes:AVERAGE" \ "DEF:out1=file1.rrd:outbytes:AVERAGE" \ "DEF:out2=file2.rrd:outbytes:AVERAGE" \ "LINE2:in1#00CC00:Bytes in router 1" \ "LINE2:in2#CC0000:Bytes in router 2" \ "LINE2:out1#0000CC:Bytes out router 1" \ "LINE2:out2#CCCCCC:Bytes out router 2" I.e. you just define the different data you want to graph, then you can add as many lines as you like using different colors for each. See the docs for the "rrdgraph" command for details. Henrik From ephedrine at rogers.com Mon Oct 29 17:16:56 2007 From: ephedrine at rogers.com (Lex) Date: Mon, 29 Oct 2007 12:16:56 -0400 Subject: [rrd-users] Getting Started with RRD References: <1840298116@web.de> Message-ID: <001201c81a47$209c6620$3201a8c0@ONE> So is that a pre-requirement to RRD? That all data have an even step? The reason I ask, is that the series of data I have doesn't necessarily have information available at regular intervals; consider reachability issues and the like.. thanks Alex ----- Original Message ----- From: To: "Lex" ; Sent: Monday, October 29, 2007 4:05 AM Subject: Re: [rrd-users] Getting Started with RRD Hi, > I have a system, that periodically reports server load, HDD load as > well as a few other stats...to get started however I'd like to make a > simple graph that plots server load, and ping response time. [...] > unix timestamp ping-to response time cpu load > 1193624363 0.02 0.1 > 1193624363 0.001 0.05 > Is it possible to plot both of these bits of information on a same > RRDTool graph? nope, unless you change the interval of value-readings to a minimum of 1 second before trying to create a rrd out of these informations. ;-) lets assume you check your server every minute (60s), you'd have to create a rrd with some lines like this: rrdtool create serverstats.rrd --step 60 \ DS:ping:ABSOLUTE:120:0:U \ DS:load:ABSOLUTE:120:0:U \ [and then the needed RRA-Definitions...] but you should check the rrdtool-webpage for more detailed informations and documentation how to use rrdtool at all. (www.rrdtool.org) HTH Rene __________________________________________________________________________ Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach! Mehr Infos unter http://produkte.web.de/club/?mc=021131 From wernli at in2p3.fr Tue Oct 30 16:46:06 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Tue, 30 Oct 2007 16:46:06 +0100 Subject: [rrd-users] 1.3b and timestamps Message-ID: <20071030154606.GB31518@ccswiss.in2p3.fr> Hi, I just noticed that timestamps of RRDs stopped to be updated since my upgrade to 1.2.99907080300 Any idea why? From tobi at oetiker.ch Tue Oct 30 18:51:28 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Tue, 30 Oct 2007 18:51:28 +0100 (CET) Subject: [rrd-users] 1.3b and timestamps In-Reply-To: <20071030154606.GB31518@ccswiss.in2p3.fr> References: <20071030154606.GB31518@ccswiss.in2p3.fr> Message-ID: Hi Fabien, hmmm this would indicate the linux does not bother to update the timestamp of an mmaped file ... guess this is not optimal for backups and such ... anyone got any insight into this ? it might be good from a performance standpoint though, since this will reduce filesystem meta data updates, potentially some more writing saved ... cheers tobi Today Fabien Wernli wrote: > Hi, > > I just noticed that timestamps of RRDs stopped to be updated > since my upgrade to 1.2.99907080300 > > Any idea why? > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From andy.riebs at hp.com Tue Oct 30 18:59:50 2007 From: andy.riebs at hp.com (Andy Riebs) Date: Tue, 30 Oct 2007 13:59:50 -0400 Subject: [rrd-users] 1.3b and timestamps In-Reply-To: References: <20071030154606.GB31518@ccswiss.in2p3.fr> Message-ID: <1193767190.11176.38.camel@ariebs> Fabien, Is your disk mounted with the "noatime" attribute, by any chance? /andy On Tue, 2007-10-30 at 17:51 +0000, Tobias Oetiker wrote: > Hi Fabien, > > hmmm this would indicate the linux does not bother to update the > timestamp of an mmaped file ... guess this is not optimal for > backups and such ... > > anyone got any insight into this ? > > it might be good from a performance standpoint though, since this > will reduce filesystem meta data updates, potentially some more > writing saved ... > > cheers > tobi > > Today Fabien Wernli wrote: > > > Hi, > > > > I just noticed that timestamps of RRDs stopped to be updated > > since my upgrade to 1.2.99907080300 > > > > Any idea why? > > > > _______________________________________________ > > rrd-users mailing list > > rrd-users at lists.oetiker.ch > > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > > > > > -- > Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten > http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users From wernli at in2p3.fr Tue Oct 30 22:20:51 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Tue, 30 Oct 2007 22:20:51 +0100 Subject: [rrd-users] 1.3b and timestamps In-Reply-To: <1193767190.11176.38.camel@ariebs> References: <20071030154606.GB31518@ccswiss.in2p3.fr> <1193767190.11176.38.camel@ariebs> Message-ID: <20071030212051.GD21901@mail.faxm0dem.org> On Tue, Oct 30, 2007 at 01:59:50PM -0400, Andy Riebs wrote: > Is your disk mounted with the "noatime" attribute, by any chance? I already checked: it's not! /dev/md10 /srv xfs rw,nosuid,nodev,sunit=128,swidth=256 0 0 From ephedrine at rogers.com Wed Oct 31 06:16:03 2007 From: ephedrine at rogers.com (Lex) Date: Wed, 31 Oct 2007 01:16:03 -0400 Subject: [rrd-users] [Fwd: Re: Getting Started with RRD] References: <52334.80.99.239.65.1193659357.squirrel@vlad.hu> <20071029123237.GA15414@hswn.dk> Message-ID: <000c01c81b7d$228e6b40$3201a8c0@ONE> Thanks for the previous help, I've managed to get everything working! Now comes to the time to 'beautify' the graphs themselves, and I would like to add a gradient to the area below the graph line. In one case, the graph is a percentage, where: - 0 to 60% should be gree - 61 to 90% yellow - 90 to 100% red ...with a gradient in between... for example, the Temperature graph on the examples page at that links to http://pampa.tche.br/temperatura/temperatura.cgi would do just perfectly. Unfortunately...I can't find any examples of how this is done. Any guidance appreciated! Alex ----- Original Message ----- From: "Henrik Stoerner" To: "Pal, Laszlo" Cc: Sent: Monday, October 29, 2007 8:32 AM Subject: Re: [rrd-users] [Fwd: Re: Getting Started with RRD] >> I've a very similar issue. I've my data in two separate rrd file (these >> are router if traffic). The interval are the same (rrd created by >> router2.cgi) and I have two values for each 5 minutes (in/out) > > Assuming you have two RRD files, file1.rrd and file2.rrd, and each of > these have an "inbytes" and "outbytes" dataset, then you can create a > graph with data from both files like this: > > rrdtool graph somefile.png -s e-48h \ > "DEF:in1=file1.rrd:inbytes:AVERAGE" \ > "DEF:in2=file2.rrd:inbytes:AVERAGE" \ > "DEF:out1=file1.rrd:outbytes:AVERAGE" \ > "DEF:out2=file2.rrd:outbytes:AVERAGE" \ > "LINE2:in1#00CC00:Bytes in router 1" \ > "LINE2:in2#CC0000:Bytes in router 2" \ > "LINE2:out1#0000CC:Bytes out router 1" \ > "LINE2:out2#CCCCCC:Bytes out router 2" > > > I.e. you just define the different data you want to graph, then you can > add as many lines as you like using different colors for each. > > See the docs for the "rrdgraph" command for details. > > > Henrik > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users From wernli at in2p3.fr Wed Oct 31 07:03:53 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Wed, 31 Oct 2007 07:03:53 +0100 Subject: [rrd-users] 1.3b and timestamps In-Reply-To: References: <20071030154606.GB31518@ccswiss.in2p3.fr> <1193767190.11176.38.camel@ariebs> <20071030212051.GD21901@mail.faxm0dem.org> Message-ID: <20071031060353.GC4557@mail.faxm0dem.org> On Wed, Oct 31, 2007 at 01:27:03AM +0100, Tobias Oetiker wrote: > I don't think it has todo with noatime ... it is the mtime which > does not get updated, right ? I have seen the same happening with > samba database files ... I know - and yes, the atime _is_ being modified. It's the mtime which stays the same. Just to rule out anything on my side, can anyone reproduce this behaviour? > now I could add code to rrdtool which forces a timestamp update, > but then again, this would cause additional disk io, which is > probably not what we want ... yeah. If this is "normal" mmap behaviour, I'll modify my own code to touch the mtime From tobi at oetiker.ch Wed Oct 31 07:47:59 2007 From: tobi at oetiker.ch (Tobias Oetiker) Date: Wed, 31 Oct 2007 07:47:59 +0100 (CET) Subject: [rrd-users] 1.3b and timestamps In-Reply-To: <20071031060353.GC4557@mail.faxm0dem.org> References: <20071030154606.GB31518@ccswiss.in2p3.fr> <1193767190.11176.38.camel@ariebs> <20071030212051.GD21901@mail.faxm0dem.org> <20071031060353.GC4557@mail.faxm0dem.org> Message-ID: Hi Fabien, look at this ... http://lkml.org/lkml/2007/1/9/310 it seems to be a kernel bug ... cheers tobi Today Fabien Wernli wrote: > On Wed, Oct 31, 2007 at 01:27:03AM +0100, Tobias Oetiker wrote: > > I don't think it has todo with noatime ... it is the mtime which > > does not get updated, right ? I have seen the same happening with > > samba database files ... > > I know - and yes, the atime _is_ being modified. It's the mtime which stays > the same. Just to rule out anything on my side, can anyone reproduce this > behaviour? > > > now I could add code to rrdtool which forces a timestamp update, > > but then again, this would cause additional disk io, which is > > probably not what we want ... > > yeah. If this is "normal" mmap behaviour, I'll modify my own code to touch > the mtime > > _______________________________________________ > rrd-users mailing list > rrd-users at lists.oetiker.ch > https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users > > -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902 From alex at ergens.op.het.net Wed Oct 31 13:20:14 2007 From: alex at ergens.op.het.net (Alex van den Bogaerdt) Date: Wed, 31 Oct 2007 13:20:14 +0100 Subject: [rrd-users] [Fwd: Re: Getting Started with RRD] In-Reply-To: <000c01c81b7d$228e6b40$3201a8c0@ONE> References: <52334.80.99.239.65.1193659357.squirrel@vlad.hu> <20071029123237.GA15414@hswn.dk> <000c01c81b7d$228e6b40$3201a8c0@ONE> Message-ID: <20071031122014.GA24168@ergens.op.het.net> On Wed, Oct 31, 2007 at 01:16:03AM -0400, Lex wrote: > Now comes to the time to 'beautify' the graphs themselves, and I would like > to add a gradient to the area below the graph line. > > In one case, the graph is a percentage, where: > > - 0 to 60% should be gree > - 61 to 90% yellow > - 90 to 100% red > > ...with a gradient in between... for example, the Temperature graph on the > examples page at that links to > http://pampa.tche.br/temperatura/temperatura.cgi would do just perfectly. > Unfortunately...I can't find any examples of how this is done. Look in the documentation and find tutorials for rrdtool, cdef, rpn. Then convert the following basic idea to CDEF and AREA statements: show everything in red if (value is above 90%) then show 90%, else everything in yellow if (value is above 60%) then show 60%, else everything in grey This means you plot everying in red, overlay part of the value with yellow, and part of it with grey. You end up with grey, perhaps with a bit of yellow above it, and perhaps with a bit of red above that. In CDEF, an if-then-else looks like: a,b,c,IF where (a) is (boolean, such as: value is above 90%) where (b) is (if true, e.g. 90% of value) where (c) is (if false, e.g. value) Thus: if (value>9000000) then 9000000 else value becomes: value,9000000,GT,9000000,value Double if: if (val>1) then if (val<10) then val else 10 else 1 Split into two if-then-else statements, convert and combine: a,b,c,IF this is: if (a) then (b) else (c) x,y,z,IF a,x,y,z,IF,c,IF this is: if (a) then {if(x) then (y) else (z)} else (c) a: val>1 b: result of other if c: 1 x: val<10 y: val z: 10 convert a and x: a: val,1,GT x: val,10,LT convert abc: val,1,GT,b,1,IF (todo: b) convert xyz: val,10,LT,val,10,IF substitute b: val,1,GT,val,10,LT,val,10,IF,1,IF How do you find out what 90% is? This depends on what you want. Sometimes, 90% will be 9,000,000 (for 10Mbps ethernet). Sometimes it will be 4,500,000 (for 10Mbps ethernet, still usable). Or maybe you have {blue,green,red} for {too cold,comfortable,too warm}. Sometimes it should be what is visible on the graph itself. You will need to use rrdtool v1.2 (or above) which knows VDEF: VDEF:max=temperature,MAXIMUM and then use "max" instead of a number in your CDEFs. You may even want to setup your databases not only with averages but also with minimum and maximum rates, and then plot only the part between these two rates: difference=maximum-minimum plot minimum using transparency stack the difference using a colour and if you want different colours: difference=maximum-minimum if (maximum>90%) then y1=difference, y2=blank, y3=blank else if (maximum>60%) then y1=blank, y2=difference, y3=blank else y1=blank,y2=blank,y3=difference use "UNKN" (unknown) for blank, this will not show up on the graph. plot minimum using transparency, stack y1 using red plot minimum using transparency, stack y2 using yellow plot minimum using transparency, stack y3 using grey Transparency is achieved by not setting a colour, or by setting its opacity to zero (v1.2 and above only): AREA:value or AREA:value#FFFFFF00 (any colour RRGGBBAA will do, as long as AA is zero) The resulting set of CDEFs and AREAs will look complex but really is nothing more than straight forward programming. Some code for v1.2 (untested, you debug and/or modify): DEF:min=temp.rrd:temperature:MIN DEF:max=temp.rrd:temperature:MAX CDEF:dif=max,min,- CDEF:y1=max,9000000,GT,dif,UNKN,IF CDEF:y2=max,6000000,GT,max,9000000,LE,dif,UNKN,IF,UNKN,IF CDEF:y3=max,6000000,LE,dif,UNKN,IF AREA:min AREA:y1#F00:busy:STACK AREA:min AREA:y2#0C0:normal:STACK AREA:min AREA:y3#00F:idle:STACK If you want 90% of the current maximum, replace 9000000 with the appropriate VDEF example: VDEF:maxseen=max,MAXIMUM then replace "9000000" with "maxseen,0.9,*" and so on HTH -- Alex van den Bogaerdt http://www.vandenbogaerdt.nl/rrdtool/