Fatal segment fault

Hello,

I've recently upgraded from Redhat 5.1 --> Redhat 6.0 and while I was at
it I went from Squid 1.2Beta22 --> Squid 2.2Stable3 I've also enabled the
async-io option which I never used to have on.

I've managed to have an error like this on both my cache machines in the
last week:

FATAL: Received Segment Violation...dying.
CPU Usage: 4593.760 seconds
Maximum Resident Size: 0 KB
Page faults with physical i/o: 786
Memory usage for squid via mallinfo():
    total space in arena: 355318 KB
    Ordinary blocks: 350416 KB 8108 blks
    Small blocks: 0 KB 0 blks
    Holding blocks: 176 KB 1 blks
    Free Small blocks: 0 KB
    Free Ordinary blocks: 4901 KB
    Total in use: 350592 KB 99%
    Total free: 4901 KB 1%

I'm not using any special malloc() library and I have cache_mem at 256Mb
(there is 512mb of RAM in each server).

Any ideas on how I can avoid these segment faults?

Regards,

Leigh
Received on Mon Jun 21 1999 - 07:01:37 MDT