www-apache-bugdb mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Per Elmdahl <erap...@lmera.ericsson.se>
Subject os-solaris/883: No record locks available - error.log fills up with messages
Date Fri, 18 Jul 1997 12:40:02 GMT

>Number:         883
>Category:       os-solaris
>Synopsis:       No record locks available - error.log fills up with messages
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    apache (Apache HTTP Project)
>State:          open
>Class:          sw-bug
>Submitter-Id:   apache
>Arrival-Date:   Fri Jul 18 05:40:01 1997
>Originator:     erapeel@lmera.ericsson.se
>Organization:
apache
>Release:        1.2.1
>Environment:
Server running on Solaris 2 (SunOS 5.5.1), file system NFS mounted
from Auspex server (uname -a gives: "SunOS lmeray 4.1.4 2 aushp")
Apache compiled with gcc. "gcc -v" says  "... 2.7-96Q3"
>Description:
After running the server lightly for a few days, the server stopped working.
 The error.log file filled up with:
  fcntl:F_SETLKW: No record locks available
  Error getting accept lock. Exiting.

The configuration automatically uses USE_FCNTL_SERIALIZED_ACCEPT as Apache
is compiled on a Solaris2 machine, BUT the file server is a SunOS
machine. Should I use USE_FLOCK_SERIALIZED_ACCEPT instead, as the release
notetes suggested as remedy for another problem?
in the release notes (
>How-To-Repeat:
I guess by having a SunOS4.1.4. machine NFS-serving a SunOS5.5.1 Web server.
Maybe Auspex has its own network code, so you need an Auspex........
>Fix:
If my idea is the "right" way to do it, make sure that it is documended in the
installation instructrions. If you happen to know what is going on in the locking
internals, please give a brief description in the documentation.

        THANKS!
           //Per
%0
>Audit-Trail:
>Unformatted:



Mime
View raw message