Subject: Re: [sdss3-25mlog 866] 2.5m obslog 20121011 (MJD 56212)
From: "John K. Parejko"
Date: Fri, 12 Oct 2012 10:08:47 -0400
> 09:45Z Had a critical alert with the BOSS-icc heartbeat.  We resolved =
this by
> starting BOSS-icc again through stageManager.
Did you run the status commands that I requested? I can't find them in =
the logs. I've copied my request below. We need to determine whether the =
connection to the ICC is really going bad, or whether the alerts actor =
is not reseting itself properly. The below commands will help with that, =
and may clear the error without restarting the ICC.
Thanks.
John
In the hopes of figuring out where these errors are coming from, if you =
receive the "BOSS ICC's been dead for XXX seconds" alert, could you =
please run the following commands in sequence, and include the results =
in your night log?
boss ping
boss status
alerts ping
alerts status
boss camCheck
Once you've issued the camCheck, wait a minute or so to see if the =
heartbeat alert has cleared (it may). If it doesn't clear, then restart =
the BOSS ICC with stageManager. Running "boss iack" shouldn't be =
necessary.
You may already do this, but to make pasting the output of the above =
easier, you can set a log window with a filter for severity "None" "+ My =
Commands and Replies". Cuts down on the clutter if you're just looking =
for the results of your commands.