Re: comm_accept: FD 37

From: Jonathan Larmour <[email protected]>
Date: Fri, 27 Jun 1997 15:32:17 +0100

At 09:28 27/06/97 -0300, Marco Aurelio S. Lopes wrote:
>At 08:16 26/06/97 +0200, you wrote:
>>Hey,
>>
>>i see a lot of error messages like :
>>
>>97/06/26 07:55:27| asciiHandleConn: FD 37: accept failure: (71) Protocol
>>error
>>97/06/26 07:55:27| comm_accept: FD 37: accept failure: (71) Protocol error
>>
>>in my 'cache.log' !
>>
>>What did they mean ?

All this implies is that when squid did an accept() on a connection, it got
returned EPROTO. It is highly unlikely that squid can _cause_ this to
happen as this is being returned from your kernel protocol stack, and is
more likely to be either a bug in your OS, or some tuning issue - you could
be out of resources, maybe file descriptors (accept creates a socket, which
uses an FD), or some other configurable OS parameter.

To see squid's FD usage, use the cache manager. Obviously, you could run
out of FD's for the _whole_ of your system if your using your squid server
as a server for other things too.

Have a look in your OS docs, or maybe your man pages for accept (section 3
of most man pages). If nothing else, try temporarily increasing your FD
limits and see if the problem goes away/diminishes.

However, generally, EPROTO is due to things like the client setting bad IP
options - and it shouldn't slow down your system when it happens, so I'm
still not sure this is the correct answer.

Jonathan L.

Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355
------[ Do not think that every sad-eyed woman has loved and lost... ]------
April 12th! Ra!Ra!----[ she may have got him. -Anon ]-----April 12th! Ra!Ra!
Help fight spam! http://spam.abuse.net/ These opinions are all my own fault
Received on Fri Jun 27 1997 - 07:38:43 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:35:35 MST