httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jeff Trawick <>
Subject Re: Solaris8 Problem with native compiler
Date Tue, 12 Feb 2002 18:22:55 GMT
jean-frederic clere <> writes:

> I have some problems with the httpd-2.0 (from CVS).
> Whe compiled with Sun compiler it hangs at startup:
> +++
> > $c                                                                            
>`_lwp_sema_wait+8(8d7f0, ff19e000, 0, 8d738, 2501c, 0)                 
>`_swtch+0x400(8d738, 8d738, ff19e000, 5, 671, cf3c0)              
>`_mutex_adaptive_lock+0x144(ff1a9944, 66666400, 4c00, 1, 4d58,    
> fffeffff)                                                                       
>`_cmutex_lock+0x50(ff29e448, ff, ff280fce, ff21a82c, ffbef680,    
> ff280fce)                                                                       
>`_get_hostserv_inetnetdir_byname+0x224(ffbef78c, ffbef794, 0,        
> ffbef794, ff29ae30, 93c09)                                                      
>`getipnodebyname+0x12c(0, 0, 8e6a8, ffbef80c, 93c08, cea38)          
>`get_addr+0xc4(0, 93c08, ffbef980, 1, ff31a000, 1e62)             
>`getaddrinfo+0x524(ff31ac0c, 93c08, 1e62, ffbefa1c, ff31a000,     
> ffbef980)                                                                       
>`apr_sockaddr_info_get+0x14c(d7cc4, 93c08, 0, 1e62, 0, 95930)        
> ap_mpm_pod_open+0xd8(95930, 8a66c, 49, 1, ff3e0000, ff370330)                   
> prefork_open_logs+0xb0(95930, bd9f8, bfa00, 975e0, bfa00, 0)                    
> ap_run_open_logs+0x98(95930, bd9f8, bfa00, 975e0, 0, 0)                         
> main+0x864(1, ffbefc9c, ffbefca4, 87800, 0, 0)                                  
> _start+0xb8(0, 0, 0, 0, 0, 0)                                                   
> +++

I've finally verified that there is a Solaris fix for this.  I think
the specific patch is 109326 (Sparc) or 109327 (i386), but in truth I
installed the current "Solaris recommended patch cluster" in its
entirety (which included that patch).  After patch installation and
reboot, I didn't hit the problem again.

(Funny: While you were encountering and reporting this and I was
trying to think of what it might be, I had a regression test blocked
for days and days in the same getaddrinfo() call on an unpatched Sun 8
box at the office.  Unfortunately, the sendmail cfg on that box is
hosed and I didn't see the error messages from it being hung.)

Now that the box is working better I see a couple of errors that
need further study:

1) cgi request handled by cgid right after a restart isn't working
   (I added a bunch of restarts in my regression test during the
   days that it was hung on Solaris, so probably this isn't related to
   the patchlevel)

2) POD issue with worker MPM:

[Tue Feb 12 12:01:46 2002] [notice] SIGUSR1 received.  Doing graceful
[Tue Feb 12 12:01:46 2002] [warn] (128)Network is unreachable: connect
to listener

Jeff Trawick | | PGP public key at web site:
             Born in Roswell... married an alien...

View raw message