[rrd-users] memory corruption

[Kiss Gabor (Bitman)]

> > *** glibc detected *** realloc(): invalid pointer: 0xbfb01af1 ***
> > Aborted
> O.K. (Thank god newer libc are more picky about such memory inconsistencies ; )
> At first which version of rrdtool do you use ?

$ rrdtool --version
RRDtool 1.2.15  Copyright 1997-2006 by Tobias Oetiker <tobi at oetiker.ch>
               Compiled Dec 14 2006 22:05:35
$ dpkg -l *rrd*
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Name           Version        Description
+++-==============-==============-============================================
ii  librrd2        1.2.15-0.3     Time-series data storage and display system 
ii  librrds-perl   1.2.15-0.3     Time-series data storage and display system 
ii  rrdtool        1.2.15-0.3     Time-series data storage and display system 



> Maybe you should upgrade to a more recent version?

I'm afraid it is not possible. This is a production machine and I'm not
sysadmin on this host.

> 
> Or - as a workaround - set the environment variable MALLOC_CHECK_ to 0
> But I recommed a - possible - upgrade of the rrdtool. I bet this issue
> has been fixed ;)

If you say only rrddump is affected I don't care with it. :)

> 
> > 1. I don't know if the first command creating database is correct or not.
> > Please help.
> If it gives no error it should be created correct.
> 
> > 3. I guess rrddump should not crash but check file more strictly.
> > "rrdtool fetch /var/lib/cricket//power/klimagepek/fh1.rrd AVERAGE" works well.
> Well if the creation works without error _AND_ you can fill the database
> with - sane - values. Well I think the database is fine.

Thanks.
This is enough for me.

Gabor