Re: Squid 2.3Stable1 stops serving WCCP #HELLO??

From: Garlic <garlic@dont-contact.us>
Date: Wed, 16 Feb 2000 08:33:50 -0800

We are using WCCP, Linux, and 2.3S1 here with no problems like this. How is the
squid server connected to the router? If the "route" varies then that is your
problem. Squid does not tolerate a changing route very well.

Ahsan Khan wrote:

> Hi Walter,
>
> Well I have observed some thiong more in this regards. I have two system
> running squid 2.3.STABLE1, Now an interesting thing is that wccp stop
> working on the system which is acting as a proxy server as well as a cache
> engin, Thats mean there must be some thing we hav eto define like
> filediscriptors. I am trying to debug that and hope soon will get the
> Probelm and will kill that . any progress from Your side.?
>
> With Regards
> Ahsan Khan
> Sr. System Admin
> Internet Division (OneNet)
> Sun Communication Pvt. Ltd.
> http://www.one.net.pk
>
> ----- Original Message -----
> From: "Walter Klomp" <walter@swiftech.net.sg>
> To: <squid-users@ircache.net>
> Sent: Tuesday, February 15, 2000 9:08 AM
> Subject: Squid 2.3Stable1 stops serving WCCP #HELLO??
>
> > Hello,
> >
> > Is everybody so flabbergasted that the list remains silent regarding this
> > problem? :-) I got 1 private reply of somebody who has the same problem
> and
> > is also waiting for a solution..
> >
> > Maybe the silence is because I forgot to mention that I am running squid
> on
> > RedHat Linux, Kernel 2.2.14, WCCP with the proprietry wccp.c module, with
> > async-io & cache-digests. The hardware is HP Netserver... The router is
> > CISCO 720x with IOS 11.2.
> >
> > I hope to get some pointers on how to resolve this soon...
> >
> > Thanks in advance,
> > Walter Klomp.
> >
> >
> > -----Original Message-----
> > From: Walter Klomp [mailto:walter@swiftech.net.sg]
> > Sent: Monday, February 14, 2000 11:17 PM
> > To: Squid-Users@Ircache. Net
> > Subject: Squid 2.3Stable1 stops serving WCCP
> >
> >
> > Hi,
> >
> > I am running 2 squid boxes with WCCP, but on a regular basis, one of them
> > stops serving, and the cisco router stops using it, although it is marked
> > usable...
> >
> > The only thing I can do to get the box to work again is squid -k shutdown,
> > after which the RunCache will start it up again...
> >
> > It seems that after a crash, the WCCP doesn't work anymore, until it is
> then
> > shutdown again... here's the message from the last crash which made the
> wccp
> > stop...
> >
> > 2000/02/14 20:42:24| sslReadServer: FD 85: read failure: (104) Connection
> > reset by peer
> > 2000/02/14 21:45:19| storeAufsOpenDone: (2) No such file or directory
> > 2000/02/14 21:45:19| /usr/local/squid/cache3/a/15/42/00056133
> > 2000/02/14 21:45:19| assertion failed: store_swapin.c:80: "errflag <= 0"
> > 2000/02/14 21:45:30| Starting Squid Cache version 2.3.STABLE1 for
> > i686-pc-linux-gnu...
> > 2000/02/14 21:45:30| Process ID 1397
> > 2000/02/14 21:45:30| With 4096 file descriptors available
> > 2000/02/14 21:45:30| Performing DNS Tests...
> > 2000/02/14 21:45:30| Successful DNS name lookup tests...
> > 2000/02/14 21:45:30| DNS Socket created on FD 5
> > 2000/02/14 21:45:30| idnsParseResolvConf: nameserver 203.123.8.8
> > 2000/02/14 21:45:30| idnsAddNameserver: Added nameserver #0: 203.123.8.8
> >
> >
> > Anybody an idea what could be the problem ?
> >
> > Thanks
> > Walter.
> >
Received on Wed Feb 16 2000 - 09:42:51 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:51:17 MST