https://issues.apache.org/bugzilla/show_bug.cgi?id=47645
Summary: httpd-2.2.12 often hangs for hours
Product: Apache httpd-2
Version: 2.2.12
Platform: Sun
OS/Version: Solaris
Status: NEW
Severity: critical
Priority: P2
Component: All
AssignedTo: [email protected]
ReportedBy: [email protected]
--- Comment #0 from HWS <[email protected]> 2009-08-05 03:12:31 PDT ---
After upgrading from httpd 2.2.11 to 2.2.12, the server was blocked at least
once per day for up to six hours. In this state, neither any answer packets
were generated (tested with network sniffer) nor any log lines written. In one
such state, I restarted the server gracefully, then it worked again. The log
files do not show any errors. Tracing the running master httpd process with
Solaris "truss" showed a loop of pollsys(...) and waitid(P_ALL,...) calls, but
this is normal. Only downgrading to 2.2.11 helped.
Replacing the log-pipe syntax "|exec cronolog ..." with the new one,
"||cronolog ..." had no effect. Systematic testing will be difficult since I
cannot do this on our "production" web server and the bug cannot be triggered
at will.
System: Solaris 10 u7 (SPARC), httpd self-compiled with Sun Studio Forte 7
compiler, configure options: CFLAGS='-xO3 -xarch=v8 -xstrconst'
--disable-threads --enable-suexec --with-ssl=/usr/sfw (plus module options;
mpm=prefork [default]). Additionally, mod_fastcgi was compiled separately. All
this and the runtime configuration were identical to previous versions, which
always worked reliably.
Process hierarchy (log output is piped through cronolog):
httpd (master)
cronolog
cronolog
cronolog
httpd (fastcgi controller)
php-fcgi (master, PHP_FCGI_CHILDREN=2)
php-fcgi
php-fcgi
httpd
httpd
...
httpd
--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
