[SQU] No buffer space available

From: Max V. Bouglacoff <vesemir@dont-contact.us>
Date: Thu, 8 Mar 2001 12:36:55 +0200 (EET)

  Hello,

Some time ago I've started getting this lines in cache.log

2001/03/08 01:59:04| Accepting HTTP connections at 0.0.0.0, port 3128, FD 6.
...
2001/03/08 07:25:40| comm_accept: FD 6: (105) No buffer space available
2001/03/08 07:25:40| httpAccept: FD 6: accept failure: (105) No buffer
 space available
2001/03/08 07:25:40| comm_accept: FD 6: (105) No buffer space available
2001/03/08 07:25:40| httpAccept: FD 6: accept failure: (105) No buffer
 space available
2001/03/08 07:25:40| comm_accept: FD 6: (105) No buffer space available
2001/03/08 07:25:40| httpAccept: FD 6: accept failure: (105) No buffer
 space available

And oftenly squid crashes after that. I have no idea where to look and what
to fix. This hapens with 2.4-pre1 and 2.4-pre2. Platform: Slackware-4.0,
2.2.17, 256M, 18G SCSI. File descriptors set to 16384.
# ./configure --enable-stacktraces --enable-underscores
  --disable-ident-lookups --enable-cache-digests --enable-forw-via-db
  --enable-htcp --enable-snmp --disable-wccp --enable-icmp
  --enable-removal-policies=heap --enable-storeio=diskd --enable-gnuregex
Also squid runs 10 `adzap' redirectors on port 8080 (FD 7)

  respects
---------------------------------------------
Max V. Bouglacoff <nospam-vesemir@iname.com>
Co-sysadmin, programmer of ISP "Farlep"
Odessa, Ukraine

--
To unsubscribe, see http://www.squid-cache.org/mailing-lists.html
Received on Thu Mar 08 2001 - 03:40:29 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:58:34 MST