Re: [squid-users] cache.log: this be aioCancel

From: Hans Juergen von Lengerke <lengerkeh@dont-contact.us>
Date: Wed, 27 Mar 2002 11:58:09 +0100 (CET)

I got an off list reply for this from Marc Elsen (thx). Apparently this
was discussed before on the list but I just didn't find it in the
archive. So the answer is:

  Don't worry about the aioCancel, it is an informational message,
  which has a too low visible debug setting in SQUID.
  It is printed if a user aborts a request in the browser,(probably).

Hans Juergen von Lengerke <lengerkeh@sixt.de> on Mar 26, 2002:

> I just upgraded an accelerator squid to 2.4.STABLE6 with:
>
> ./configure --prefix=/home/www/squid --enable-removal-policies=heap,lru
> --enable-storeio=ufs,aufs --disable-ident-lookups --disable-wccp
> --disable-snmp --enable-poll --disable-internal-dns --enable-async-io
>
> This thing is doing up to 80Hits/s, the file system on the cache_dir is
> reiserfs. I'm using aufs and I'm getting these errors quite frequently
> in cache.log:
>
> 2002/03/26 18:51:07| this be aioCancel
> 2002/03/26 18:51:52| aio_queue_request: WARNING - Queue congestion
>
> Now, the latter I found in the archives where I read that it's probably
> OK and if I really care I need to use more aio threads. But what about
> the former? 'this be aioCancel' what does it mean? do I have to worry?
Received on Wed Mar 27 2002 - 03:58:32 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:07:07 MST