At 3:43 AM +1000 1/22/98, Dancer wrote:
>Are you bouncing against a resource-limit that might be generating the signal?
It's conceivable, but I don't think it's likely, for several reasons:
1) The kernel and system configuration has been tuned with
2) Adjusting the resource limits don't help the problem at all.
3) I've seen various resource limits reached on occassion, and they don't
generate a random HUP signal, they generate error messages consistent
with the resource limits being reached.
4) If it were just a matter of a signal being generated, squid should
gracefully restart like it does when I actually send it HUP signals.
Unfortunately, it dies without any further messages about why.
From Mr. Dace's original message:
>This from squid_cache.log:
>
>1998/01/20 13:56:24| reconfigure: SIGHUP received
>1998/01/20 13:56:24| Waiting 30 seconds for active connections to finish
>1998/01/20 13:56:24| FD 21 Closing HTTP connection
Usually an HUP signal doesn't cause squid to decide to die immediately.
Jordyn
|----------------------------------------------------------------|
|Jordyn A. Buchanan mailto:jordyn@bestweb.net |
|Bestweb Corporation http://www.bestweb.net |
|Senior System Administrator +1.914.271.4500 |
|----------------------------------------------------------------|
Received on Wed Jan 21 1998 - 10:03:50 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:38:28 MST