BB Unix Network Monitor - Message

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

{bb} Bogus disk full messages?



I'm running BB 19c (old, I know, but it seems to work just fine) on servers and clients running mostly Solaris 9.
I've got one machine that started sending bogus disk full messages a few days ago (see below)

-the filesystem is not full
-the filesystem is not out of inodes
-the directory is owned by the BB user
-BB is running as the correct user
-nothing else that I know of has changed on this machine in many months

I don't get the disk full message all the time; it is showing as OK roughly half the time, maybe a bit more
http://enki.gsd.harvard.edu/cgi-bin/bb-hist.sh?HISTFILE=atlas%2Cgsd%2Charvard%2Cedu.disk&ENTRIES=50

I have restarted BB but not rebooted the machine (as it is a production Oracle server)

There are a couple of errors in the system log from Tuesday but this problem started earlier and the syslog errors have not repeated

>>Oct  3 10:51:03 atlas genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap
space to grow stack for pid 4846 (bb)
>>Oct  3 10:51:03 atlas last message repeated 4 times
>>Oct  3 17:11:21 atlas genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap
space to grow stack for pid 19711 (bbrun)

what else to look at?

>Date: Thu, 5 Oct 2006 09:41:21 -0400 (EDT)
>From: Big Brother <bbrother@gsd.harvard.edu>
>To: bb_admin@gsd.harvard.edu
>Subject: !BB - 2889610! atlas.gsd.harvard.edu.disk - 100128103170036
>
>[2889610]  atlas.gsd.harvard.edu.disk red Thu Oct  5 09:41:20 EDT 2006 - Filesystem of /var/bbrother/bb19c/tmp is FULL
>
>Please see: http://enki.gsd.harvard.edu/cgi-bin/bb-hostsvc.sh?HOSTSVC=atlas,gsd,harvard,edu.disk

thanks, Betsy

Home | Main Index | Thread Index