I'm getting occasional OOM faults apparently originating from clamd on debian.
Is there a simple way to track down why these are happening and preventing them? Top never shows any problems, and the machine is very lightly loaded. Message log begins: > Mar 3 21:18:27 rose kernel: clamd invoked oom-killer: gfp_mask=0x201da, > order=0, oom_adj=0 > Mar 3 21:18:28 rose kernel: Pid: 1790, comm: clamd Not tainted > 2.6.32.8-kvm-i386-20100212 #1 > Mar 3 21:18:28 rose kernel: Call Trace: > Mar 3 21:18:28 rose kernel: [<c147e2b7>] ? printk+0x18/0x1a > Mar 3 21:18:28 rose kernel: [<c107bfb7>] oom_kill_process+0x77/0x220 > Mar 3 21:18:28 rose kernel: [<c107c59a>] __out_of_memory+0xfa/0x150 > Mar 3 21:18:28 rose kernel: [<c107c634>] out_of_memory+0x44/0x90 > Mar 3 21:18:28 rose kernel: [<c107f2cf>] __alloc_pages_nodemask+0x4ff/0x510 > Mar 3 21:18:28 rose kernel: [<c108091a>] > __do_page_cache_readahead+0xba/0x1c0 etc. Any suggestions welcome. paul _______________________________________________ Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net http://www.clamav.net/support/ml
