[Varnish] #1287: Varnish 3.0.3 - segfault in libvarnish.so.

[Varnish]

#1287: Varnish 3.0.3 - segfault in libvarnish.so.
------------------------------------+--------------------
 Reporter:  robroy                  |       Owner:
     Type:  defect                  |      Status:  new
 Priority:  normal                  |   Milestone:
Component:  varnishd                |     Version:  trunk
 Severity:  normal                  |  Resolution:
 Keywords:  segfault libvarnish.so  |
------------------------------------+--------------------

Comment (by coennie):

 We just went live with Varnish 3.0.3 and I see the same results as
 described above:

 {{{
 Apr 12 16:23:32 lb1 varnishd[8572]: child (18461) Started
 Apr 12 16:23:32 lb1 varnishd[8572]: Child (18461) said Child starts
 Apr 12 16:23:41 lb1 kernel: [13748.309110] varnishd[18488]: segfault at 0
 ip 00007f48c9355504 sp 00007f48b3ee9320 error 4 in
 libvarnish.so[7f48c9349000+13000]
 Apr 12 16:23:41 lb1 varnishd[8572]: Child (18461) died signal=11
 Apr 12 16:23:41 lb1 varnishd[8572]: Child cleanup complete
 Apr 12 16:23:41 lb1 varnishd[8572]: child (18499) Started
 Apr 12 16:23:41 lb1 varnishd[8572]: Child (18499) said Child starts
 }}}
 I've traced the trashing back to PURGING. Everytime we call a PURGE from
 the backend, the varnishd crashes with the error above.
 In the VCL I've directly copied the purge commands from the manual:

 {{{
 sub vcl_hit {
     if (req.request == "PURGE" || req.url ~ "purge=true") {
         purge;
         error 200 "Purged.";
     }
 }

 sub vcl_miss {
     if (req.request == "PURGE" || req.url ~ "purge=true") {
         purge;
         error 200 "Purged.";
     }
 }
 }}}

 Please advise!

 Regards Coen

-- 
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1287#comment:3>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator