SUMMARY: How to [temporarily] disable messages about a hardware problem?



Duh - I am a dummy! One person asked me about /etc/syslog.conf ... and this
reminded me about what process was actually doing the writing into
/var/adm/messages!

So, I have removed kern.err messages in the /etc/syslog.conf file and
refreshed the syslogd process as follows:

# svcadm refresh svc:/system/system-log

This has stopped these fan failed messages (among other messages too, of
course) for tonight - I will restore the entry in the /etc/syslog.conf file
tomorrow and refresh the syslogd process again after I replace the fan.

Thanks to Joon Kim for the reminder!

Z

From: Syed Zaeem Hosain (Syed.Hosain@xxxxxxxxx)
Sent: Tuesday, September 28, 2010 10:17 PM
To: 'sunmanagers@xxxxxxxxxxxxxxx'
Subject: How to [temporarily] disable messages about a hardware problem?

Hi, all.

I am getting the following message on one of our V240 servers in the
/var/adm/messages file:

Sep 28 21:28:01 srv007 rmclomv: [ID 728264 kern.error] CPU_FAN @ MB.P1.F1.RS
has FAILED.

And, yes, I understand that the CPU fan needs replacement (the other one has
kicked in at higher speed to compensate) - one is on order right now for
delivery tomorrow!

However, this entry in the file is also going every hour to our pager system
(special app that monitors events like these) and keeping a person awake all
night (since they have to check for other, true, alarms). I'd rather they not
get disturbed by this particular issue tonight, since they will not stay alert
for other, more important, real-time problems then!


1. Is there a way to [temporarily] disable these "fan has failed"
messages from being written into the /var/adm/messages file?

2. Perhaps a mechanism to "acknowledge" receipt of the message till I
replace the fan?

3. What the heck is the "rmclomv" source for this message as mentioned
in the /var/adm/messages file? I cannot find any man page or info on it in the
documentation (unless I have missed something!).

I tried to see if the ALOM messages could be turned off and all I see so far
as a possible source of control, is the "sc_clieventlevel" parameter in the
ALOM, but setting it to 0 ("no alarms" setting) did not seem to do the trick -
unless I missed something there(?)

Any thoughts? And, yes, this is somewhat urgent ... if you see this about 12
to 13 hours from now (by 10 to 11am US Pacific time) , then it is too late for
this question, since I will have replaced the fan by then, I hope! :) But, I'd
still like to know if there is an answer ... for future occurrences about
errors written to the /var/adm/messages from the ALOM (we have had other V240
CPU fans die too!)

Regards,

Z
_______________________________________________
sunmanagers mailing list
sunmanagers@xxxxxxxxxxxxxxx
http://www.sunmanagers.org/mailman/listinfo/sunmanagers