Hello,
One of our systems decided to appear to halt for a brief period of time
tonight .. after running for a few days straight with no problem.. I wasn't
at the console, so I only have some messages from the syslog to show..
Jan 6 20:53:50 reverse kernel: IP: queue_glue: no memory for gluing queue
c81cfce0
Jan 6 20:53:50 reverse last message repeated 3 times
This message was repeated over and over in the syslog for about an hour ..
the system 'halted' for a brief period and then seemed to regain control ..
We only have a gig of memory in the system, which is pretty much completely
used, and a 4gig swap partition (2-2gig partitions actually), which hasn't
even been touched by the system.. Again, the kernel is 2.3.34, running a
high volume of apache web servers (well.. 200-300.. if you consider that
high), and using almost 100% of the available memory....
We're also getting a very high volume of messages of the following, in the
logs:
kernel: TCPv4 bad checksum from xx.xx.xx.xx:0576 to yy.yy.yy.yy:0050,
len=325/345
(where 'xx' is an 'outside' ip address, different each time .. and 'yy' is
one of our virtual hosts on the machine)
Any ideas? (and for those that remember my last post about the load
problems .. I don't have netfilter running anymore .. however, we are
running the real-media server, nfs-mounting an sgi [Irix 6.5] for the
content .. )
Thanks,
Terry
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jan 07 2000 - 21:00:08 EST