Zenoss ZenTech Community

I am getting missed runs on some new collectors I setup for zencommand. These new collectors are built on new hardware with 24 cpu cores and 142GB of ram. Memory/cpu usage is not crazy. Why am I getting missed runs Is there a way to find out what is being missed I have tried upping the maxparellel in the zencommand.conf to 100 but I am still getting increasing missed runs on every zencommand run. Any help would be much appreciated as my collector today stopped working for command data source. Restarting the zencommand daemon resets the missed run but it continues to rise. Thanks!

2014-05-14 22:49:41,138 INFO zen.maintenance: Performing periodic maintenance
2014-05-14 22:49:41,139 INFO zen.zencommand: Counter eventCount, value 1156757
2014-05-14 22:49:41,143 INFO zen.zencommand: 347 devices processed (285798 datapoints)
2014-05-14 22:49:41,163 INFO zen.collector.scheduler: Tasks: 422 Successful_Runs: 12313 Failed_Runs: 0 Missed_Runs: 37 Queued_Tasks: 0 Running_Tasks: 4

2014-05-14 22:52:11,477 INFO zen.maintenance: Performing periodic maintenance
2014-05-14 22:52:11,478 INFO zen.zencommand: Counter eventCount, value 804095
2014-05-14 22:52:11,482 INFO zen.zencommand: 396 devices processed (184468 datapoints)
2014-05-14 22:52:11,492 INFO zen.collector.scheduler: Tasks: 426 Successful_Runs: 10917 Failed_Runs: 0 Missed_Runs: 52 Queued_Tasks: 0 Running_Tasks: 2

Subject:	zencommand missed_runs
Author:	Jan Garaj
Posted:	2014-05-22 14:15

Try to increase maxparallel setting for your zencommand and just temporary logseverity for more details in your logs.
What is your collection cycle time setting (5 minute or less) What is your CPU load (!= CPU usage) Which type of datasources/templates are you using (ssh, ...) and how much average time usually they need for completion

Devops Monitoring Expert advice: Dockerize/automate/monitor all the things.

DevOps stack: Docker / Kubernetes / Mesos / Zabbix / Zenoss / Grafana / Puppet / Ansible / Vagrant / Terraform / Elasticsearch

Subject:	Hi,
Author:	Charles Bueche
Posted:	2015-06-12 09:54

Hi,

I do have the same problem on Zenoss Core 4.2.5 with SP457 installed. I poll about 1'900 network devices, for about 800'000 data points every 5 minutes. CPU and RAM are ok. I have "maxparallel 1000" in zenperfsnmp.conf

My question is about Missed_Runs, is it going up constantly as a counter from the start of the zenperfsnmp
or showing the last cycle value like a gauge

Thx,
Charles

Subject:	IMO: it's a COUNTER.
Author:	Jan Garaj
Posted:	2015-06-14 18:36

IMO: it's a COUNTER.

Subject:	Thanks Jan, but then, what
Author:	Charles Bueche
Posted:	2015-06-17 03:55

Thanks Jan, but then, what could be the cause of such a behavior
https://dl.dropboxusercontent.com/u/1683666/zenoss_missed_runs.png

Subject:	Missed run is only symptom
Author:	Jan Garaj
Posted:	2015-06-17 04:19

Missed run is only symptom and root cause can be anything (network latency, device load, ....).

Subject:	ok thanks for your help !
Author:	Charles Bueche
Posted:	2015-06-17 07:30

ok thanks for your help !

zencommand missed_runs