Graphs breaking on Cacti 0.8.7b

[Nabeel]

Hi guys,

We are getting breaks in many graphs. Attaching a snapshot of the problematic graph.

This is happening for many hosts.

Is this a known issue/bug ?
Could it be due to excessive system load ? although this does not happens for the hosts which were added earlier.


Regards,
Nabeel

[Nabeel]

We are using a linux system with following specs,

Dell Poweredge 1950
2 x Intel(R) Xeon(R) CPU 5160 @ 3.00GHz
4GB Fully Buffered Dimm RAM (Matched Pair)
2 x 73 GB SCSI HDD

Thanks again.

[BSOD2600]

Cacti 0.8.7e is the latest now.

Those gaps are due to a lack of data from various reasons. Follow http://docs.cacti.net/manual:087:4_help ... #debugging to help find out why.

[gandalf]

Please see 2nd link of my sig as a start
Reinhard

[Nabeel]

Guys,

I have done the debugging, but don't understand the ds.MIN & ds.MAX values from the following

filename = "kar-pe1-ssg1_5min_cpu_204066.rrd"
rrd_version = "0003"
step = 300
last_update = 1248776703
ds[5min_cpu].type = "GAUGE"
ds[5min_cpu].minimal_heartbeat = 600
ds[5min_cpu].min = 0.0000000000e+00
ds[5min_cpu].max = 1.0000000000e+02
ds[5min_cpu].last_ds = "78"
ds[5min_cpu].value = 2.3400000000e+02
ds[5min_cpu].unknown_sec = 0

Can anyone explain how these are being calculated ?
Should MAX be increased or not ?

[mcutting]

Possibly a duplicate poller issue - have you checked our your crontab for all users ?

[jxc]

check to ensure that the poller actually completes polling within 300s - check for sth like this line in your cacti.log

:07/28/2009 11:04:59 AM - POLLER: Poller[0] Maximum runtime of 298 seconds exceeded. Exiting.

If so, try increasing the setting for 'Maximum Concurrent Poller Processes'

[Nabeel]

Thanks for the input.

yes, this might be the issue. There were a lot of entries like this. I have increased the threads from 10 to 20. Observing it now, lets see if that helps.