IPMI flooding connection failure on log file

[leonardomachado]

Zabbix 2.0.1

I have configured one server as IPMI
Then, created a trigger to query Ambient Temperature
It's working fine, but the logs is being flooded by theses messages:

10593:20120803:170448.974 IPMI item [ambient_temp] on host [myserverVM01] failed: another network error, wait for 15 seconds
10593:20120803:170504.980 IPMI item [ambient_temp] on host [myserverVM01] failed: another network error, wait for 15 seconds
10593:20120803:170529.627 resuming IPMI checks on host [myserverVM01]: connection restored
10617:20120803:170620.412 IPMI item [ambient_temp] on host [myserverVM01] failed: first network error, wait for 15 seconds
10593:20120803:170642.660 IPMI item [ambient_temp] on host [myserverVM01] failed: another network error, wait for 15 seconds
10593:20120803:170658.672 IPMI item [ambient_temp] on host [myserverVM01] failed: another network error, wait for 15 seconds
10593:20120803:170718.566 resuming IPMI checks on host [myserverVM01]: connection restored
10617:20120803:171026.206 IPMI item [ambient_temp] on host [myserverVM01] failed: first network error, wait for 15 seconds
10593:20120803:171048.649 IPMI item [ambient_temp] on host [myserverVM01] failed: another network error, wait for 15 seconds
10593:20120803:171104.661 IPMI item [ambient_temp] on host [myserverVM01] failed: another network error, wait for 15 seconds



I tried to send many queries simultaneously to this server, and sometimes, I get

Get Session Challenge command failed: Node busy
Error: Unable to establish LAN session
Get Device ID command failed
Unable to open SDR for reading

I think it's a normal behavior, the problem is that Zabbix is flooding my server with IPMI query (even though I have just 1 trigger). How do you handle this problem? How to avoid zabbix from quering the host so frequently. I have configure 120s of polling interval. How do I change this retry interval to something greater than 15seconds??

[rgrignon]

Great question.. Installed 2.0.2 and seeing the same exact thing... Anyone got a solution?

[bradh]

Same issue here with zabbix 2.0.3

I was thinking that maybe Dell has some problems with their iDRAC. In fact, I have an issue open with them now where pulling IPMI information *seems* to cause the Dell server to report power supplies missing (that were never installed)

Is this Dell or something else?

[Hansi]

Same problem with Fujitsu servers

I get the same problems with fujitsu servers (RX200 S7).
Bug in Zabbix ipmi implementation? openipmi in shell is always working.

[Hansi]

Changing algorythm to RMCP+

According to bug report ZBX-3243 I changed the algorythm in IPMI setting of host to RMCP+ and now it works!

[nowoe]

Had the same issue with 2.2.6 and a Supermicro motherboard.
I had imported a template with items for another, similar Supermicro mainboard. However, some items did not apply for my mainboard.
I observed that one must not have any one item that cannot be evaluated!
Having an item that checks for an unavailable sensor id causes those network-failure logs and breaks further processing of other items, resulting in missing values and a gap in the graphs.