Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Possible Duplicate:
System load by HP-SNMP-Agents is high

After receiving our HP ProLiant DL360 I'd like to integrate the machine into our Munin system and thus enable ProLiant specific information to be exposed via SNMP.

I'm running Debian Lenny with kernel 2.6.26-2-vserver-amd64 . I've followed http://downloads.linux.hp.com/SDR/getting_started.html and the HP repository has been added to /etc/apt/sources.list.d/HP-ProLiantSupportPack.list .

Setting up Lenny SNMP itself is not a problem, I configure it to have a public v1 community string to read all data and it works. I install hp-snmp-agents and run hpsnmpconfig and it adds additional lines to the top of /etc/snmp/snmpd.conf :

dlmod cmaX /usr/lib64/libcmaX64.so

snmpd gets restarted.

Via lsof I can see that libcmaX64 was loaded and is used by snmpd, put I do not get any additional information out of snmp. I use snmpwalk -v 1 -c public ... and I can see many OIDs but I do not see the new ones I'd expect, most notably temperatures, fan speed and such.

The OIDs I'm expecting are e.g. 1.3.6.1.4.1.232.6.2.6.8.1.4.1 , this is from the existing munin plugins from http://exchange.munin-monitoring.org/plugins/snmp__hp_temp/version/1 .

snmpd[19007]: cmaX: Parsing shared as a type was unsucessful
snmpd[19007]: cmaX: listening for subagents on port 25375
snmpd[19007]: cmaX: subMIB 1 handler has disconnected
snmpd[19007]: cmaX: subMIB 2 handler has disconnected
snmpd[19007]: cmaX: subMIB 3 handler has disconnected
snmpd[19007]: cmaX: subMIB 5 handler has disconnected
snmpd[19007]: cmaX: subMIB 6 handler has disconnected
snmpd[19007]: cmaX: subMIB 8 handler has disconnected
snmpd[19007]: cmaX: subMIB 9 handler has disconnected
snmpd[19007]: cmaX: sent ColdStarts on ports 25376 to 25393
snmpd[19007]: cmaX: subMIB 10 handler has disconnected
snmpd[19007]: cmaX: subMIB 11 handler has disconnected
snmpd[19007]: cmaX: subMIB 14 handler has disconnected
snmpd[19007]: cmaX: subMIB 15 handler has disconnected
snmpd[19007]: cmaX: subMIB 16 handler has disconnected
snmpd[19007]: cmaX: subMIB 21 handler has disconnected
snmpd[19007]: cmaX: subMIB 22 handler has disconnected
snmpd[19007]: cmaX: subMIB 23 handler has disconnected
snmpd[19007]: cmaX: subMIB  1 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB  2 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB  3 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB  5 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB  6 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB  8 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB  9 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 10 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 11 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 14 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 15 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 16 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 21 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 22 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 23 will be sent on port 25376 to hp Advanced Server Management_Peer
snmpd[19007]: cmaX: subMIB 18 handler has disconnected
snmpd[19007]: cmaX: subMIB 18 will be sent on port 25393 to cpqnicd
snmpd[19007]: NET-SNMP version 5.4.1

This doesn't look particular bad for me, it's just informational I guess.

I've compared the walking OID output with and without the module and there's no difference in the OID served back at all. Are there any other prerequisites I'm missing?

I've also noticed that from the time I installed hp-snmp-agents it adds a lot of additional daemons and that my load suddenly jumps to 1. I've uninstalled the package for now. Is this expected behavior?

Load spike

share|improve this question

marked as duplicate by ewwhite, Chris S Sep 12 '12 at 12:55

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

1 Answer 1

I'm experiencing the same, my load also stays put at 1 after starting the agents. I've found a solution in this thread, which reduces the load back to at least 0.6:

http://forums13.itrc.hp.com/service/forums/questionanswer.do?admit=109447627+1295820132076+28353475&threadId=1332334

Also make sure you have different community strings for the various options. Otherwise you'll keep seeing 'error in subcontainer' messages in /var/log/net-snmpd.log


Update: I'd asked about the same question before finding yours (just with a completely different title); System load by HP-SNMP-Agents is high You'll find the answer there. Quoting:

In short, you'll want to stop the HP management agents with /etc/init.d/hp-snmp-agents stop. Edit the exclude line in /opt/hp/hp-snmp-agents/cma.conf from:

exclude

to

exclude cmasm2d cpqriisd cmarackd

Start your agents again with /etc/init.d/hp-snmp-agents start

share|improve this answer
    
You're more likely to get an "answered" tag if you just post the solution here. Also, if this ground is already covered, it's entirely reasonable to vote to close this question as a duplicate. –  gWaldo Sep 25 '11 at 17:41

Not the answer you're looking for? Browse other questions tagged or ask your own question.