A short time after sucessfully adding a server to HP SIM 7.2, the daily Identify System job and manual Identify Systems job for that server fails with this message:
********* Significant issues and overall target discovery/identification status: Normal: The system has valid hardware data (model, serial number, and unique identifier) and has been identified properly for SIM to manage. Critical: The system(s) can not be identified by either WBEM or SNMP
I only have physical HP servers and I am only using SNMP. I have our SNMP community string setup as a GLOBAL protocal credential. The jobs work for a while (days, maybe a week) and then start to fail. Eventually ALL server are affected by this problem.
The only solution is to remove the servers from SIM and then re-add them. This fixes them for a few days/week and then they break again. I have tried restarting the SIM services and restart the SIM server - nothing works except for removing and re-adding the servers to SIM. Since I can sucessfully add the servers to SIM and they work for a few days, I don't think there is anything wrong with SNMP/HP Agents/Networking/Firewall/Etc. on either the SIM server or clients.
that's actually the same problem I'm having since some weeks now...
The servers have been in the HP SIM for 2-3 years now without problems but suddenly it starts to show me also this message (Critical: The system(s) can not be identified by either WBEM or SNMP) when identifying or discovering systems...
For me it looks like the monitoring and alerting works still fine (as you can see, the overall status is "The system has valid hardware data (model, serial number, and unique identifier) and has been identified properly for SIM to manage") but something makes HP SIM thinking that he isn't able to communicate with the systems using SNMP... And also when sending a test trap, it always shows up correctly in HP SIM.
Same here....everything else works such as alerting. I want to say the problem started happening after I upgraded to 7.2 but I am not 100% positive on the exact timing of when it started.
I am not sure if this is completely related, but I found that when I do a discovery of a new Gen8 server, and it then auto-discovers the iLO for the server, it hangs on the identification of the iLO. Further investigation has shown that if the iLO SNMP is set to "Agentless Management", this is the cause of the hang. If I set the iLO to "SNMP pass-through", the iLO discovery is happy. I am not using agentless management for any server so I am guessing is the problem.
In terms of the Identify System job hanging - my guess is that somewhere in my inventory I have a few Gen8 server iLOs that are still configured to use Agentless Management for SNMP and when the Identify job runs and hits them it hangs and throws the whole system into a bad state. I am not sure why this would affect other servers, so mabye my theory is way off - but so far that is the only thing I have found.
