[mrtg] statistics during server halted

[srtt.be - Michel Lombart]

Hello all,

I'm a newbie in mrtg graphics. I've installed MRTG in some Debian 
servers and firewalls, right now mainly to monitor ethernet and memory 
usage. I monitor also latency in firewalls.

Last day I had to manualy stop ( simply the 'halt' comment in a shh 
console ) a customer's server due to a main electrical installation 
scheduled maintenance. The server was halted during about one hour.

After the reboot and a full functionnal check, I've looked at the MRTG 
statistics. I was surprised to see that values during the interruption 
was not null but seems to be the same as the last measure juste before 
the halt. Values was constant during the whole period.

It's not a real issue but I wonder to know why and if there is a smart 
solution to avoid these false data ?

Because each server hosts his own graph, I do not use SNMP but some 
short perl or bash scripts reading in /proc/*** files.

Thank for you help.

-- 

Michel Lombart
Gérant

srtt.be - Informatique et internet professionnels

tel : + 32 4 264 86 00
fax : + 32 4 248 12 36
mail : michel at srtt.be
web : http://www.srtt.be/

srtt.be sprl
RPM : 0 863 982 166
Entreprise eTIC #06 - http://www.labeletic.be/

siège social
clos Saint Roch, 3
B-4683 - Vivegnis
Belgique

-------------- next part --------------
A non-text attachment was scrubbed...
Name: subs.vcf
Type: text/x-vcard
Size: 271 bytes
Desc: not available
Url : http://lists.oetiker.ch/pipermail/mrtg/attachments/20070421/cbbf6eb7/attachment.vcf