![]() |
![]() |
Subject: | zenperfsnmp timeout but commandline zenperfsnmp works |
Author: | [Not Specified] |
Posted: | 2014-04-02 15:00 |
I'm having an issue were I get timeouts in the zenperfsnmp.log file for a set of devices but if I run
zenperfsnmp run -v10 -d mydevice
from command line as zenoss user the process runs and updates my graph data.
Subject: | Version 4.2.4 |
Author: | [Not Specified] |
Posted: | 2014-04-02 15:07 |
forgot to provide the zenoss version.
Subject: | Try changing the |
Author: | [Not Specified] |
Posted: | 2014-04-03 07:39 |
Try changing the "zMaxOIDPerRequest" from 40 to 10 for those devices or in their Device Tree to see if that helps.
Hydruid
Subject: | change to zMaxOIDPerRequest |
Author: | [Not Specified] |
Posted: | 2014-04-03 13:32 |
Thanks for the response! I changed the zMaxOIDPerRequest to 10 but issue still exists.
Subject: | How many devices and try |
Author: | [Not Specified] |
Posted: | 2014-04-03 13:35 |
How many devices and try pinging those from the zenoss web UI, are the ping times good
Hydruid
Subject: | additional info |
Author: | [Not Specified] |
Posted: | 2014-04-03 14:03 |
I do see a lot of Task mydevice skipped because it was not idle in the zenperfsnmp.log file. What causes this. are there configuration values I should set to fix this.
Subject: | pings response in 1ms |
Author: | [Not Specified] |
Posted: | 2014-04-03 15:54 |
pings work ~1ms for each device. It's 10 out of 27 devices that have this issue including the zenoss server.
Subject: | What version of SNMP are you |
Author: | [Not Specified] |
Posted: | 2014-04-04 08:45 |
What version of SNMP are you using What OS is the Zenoss server What is the Zenoss server specs how many total devices are you monitoring
From the Zenoss server CLI, start a continuous ping to one of the devices that keeps timing out, let it timeout a time or two with SNMP and then check the results of the ping to see if any were dropped.
Hydruid
Subject: | SNMP Version |
Author: | [Not Specified] |
Posted: | 2014-04-04 11:31 |
Using SNMP V3
Kernel = 2.6.32-358.el6.x86_64
RAM = 16G
CPU = 8 cores
27 devices are being monitored. Only ones with issue are linux devices.
will provide results of ping in another comment.
Subject: | ping results |
Author: | [Not Specified] |
Posted: | 2014-04-07 13:07 |
Here are ping results:
70 packets transmitted, 70 received, 0% packet loss, time 69766ms
rtt min/avg/max/mdev = 1.097/1.151/1.213/0.038 ms
2014-04-07 13:58:17,340 DEBUG zen.zenperfsnmp: 3 consecutive timeouts, abandoning run for mydevice [x.x.x.x]
Subject: | More from log |
Author: | [Not Specified] |
Posted: | 2014-04-07 13:10 |
2014-04-07 13:58:17,341 DEBUG zen.zenperfsnmp: Device mydevice [x.x.x.x] 0 of 31 OIDs scanned successfully
2014-04-07 13:58:17,341 DEBUG zen.zenperfsnmp: Device mydevice [x.x.x.x] has 0 good oids, 0 bad oids and 31 untested oids out of 31 configured
Subject: | have you seen this post: http |
Author: | [Not Specified] |
Posted: | 2014-04-07 14:15 |
have you seen this post: http://community.zenoss.org/message/73439
I haven't really been able to find any info on it. Try lowering the OID's down to 5 and then restating zenoss
Hydruid
Subject: | http post |
Author: | [Not Specified] |
Posted: | 2014-04-07 14:56 |
Thanks for the post. Hadn't seen that post. I have lowered the OID down to 5 and restarted zenoss. will see if that works for these devices. Will also request the infrastructure team to validate the engineid's of these devices as well.
Subject: | Any more suggestions |
Author: | [Not Specified] |
Posted: | 2014-04-09 15:49 |
I deleted all the devices that were having the issue except for the zenoss server but still getting timeout when getting performance info for the zenoss server. zenperfsnmp runs fine from command line.
Subject: | Solved by switching to SNMP v2c |
Author: | [Not Specified] |
Posted: | 2014-04-21 15:46 |
Solved issue by switching to using SNMP v2c rather than V3.
Subject: | Hmm |
Author: | [Not Specified] |
Posted: | 2014-04-22 13:37 |
That is very interesting.......did you have something weird in your V3 config
Hydruid
Subject: | Hi, |
Author: | [Not Specified] |
Posted: | 2016-08-03 01:31 |
Hi,
I am having the same issue and by changing to v2c doesn't solve the issue. I also tried to reduce maxoid's and it didn't work out for me. I came to foung that zenperfsnmp is configured to run for 6osec by default globally. Is there a way i can setup zenperfsnmp config for each device depending on device collection rather than doing trial and error method
Any answers will be appreciated.
Thanks
Subject: | You don't say what version of |
Author: | Jane Curry |
Posted: | 2016-08-11 10:27 |
You don't say what version of Zenoss you are on For some long time (since 4.0 ) there has been a zSnmpCollectionInterval Configuration Property that you can set at either device class or at an individual device level,
Cheers,
Jane
Email: jane.curry@skills-1st.co.uk Web: https://www.skills-1st.co.uk
< |
Previous import mib problem |
Next Alert on failure to collect SNMP counter |
> |