Thank you for your response(s).
I think I have up to date firmware/disk/esh firmware versions running; machines are audited and maintained frequently. I see shelf/disk ids when I'm running a 'environment status' or any other sysstat command and there is no error message at all.
However, someone from the list guided me to enclChannelShelfAddr. It's not what I really expected, but it's something to work with.
For completion and answering the other questions I've received off-list:
Collecting those temperature values over CLI for monitoring is not a solution - but getting it over SNMP IS. If you query your filers, you'll get something similar like this:
CLI: http://pastey.net/88923 SNMP: http://pastey.net/88922-ep7t
As you can see the CLI and SNMP values are ordered different. After reading enclChannelShelfAddr I know, that all SNMP values are totally mixed up. There is not even a cut between local and partner shelves. However, you can work with those values to have a persistent key for your monitoring system (nagios, tivoli, cacti, rrd, whatever) and/or to locate the shelves physically. ;-)
enclChannelShelfAddr output: http://pastey.net/88925
Thanks a lot, -Stefan.
Webster, Stetson wrote:
I can't help but to wonder if you are encountering symptoms of outdated shelf and disk firmware which cause lots of false enclosure services errors if neglected. I ask this because the intensity of these errors can be directly proportionate to how old the firmware is.
If you are indeed having lots of these errors which are causing you to be concerned about this, I recommend calling the Support Center for guidance on what firmware you need and the exact process for updating it. Some of the firmware can be upgraded non-disruptively, and some will require an outage, there are contingencies that are best resolved at the Support Center.
Hopefully this helps to resolve a larger problem (the source of these messages) you may be having than just the SNMP monitoring of these very messages.