[Varnish] #93: management interface crashes varnishd, coredump with signal 6

[Varnish]

#93: management interface crashes varnishd, coredump with signal 6
----------------------+-----------------------------------------------------
 Reporter:  kang      |       Owner:  phk
     Type:  defect    |      Status:  new
 Priority:  normal    |   Milestone:     
Component:  varnishd  |     Version:     
 Severity:  normal    |    Keywords:     
----------------------+-----------------------------------------------------
 I'm running a 32-bit varnishd, I start varnishd with following command:

 {{{
 varnishd -a my-ip-addr:80 -f /usr/local/etc/test.vcl -s file,/usr/V,2g -T
 127.0.0.1:81
 }}}
 after varnishd started

 {{{
 telnet 127.0.0.1 81
 then press Ctrl+C

 }}}

 varnishd will crash with signal 6

 {{{
 kang# strings varnishd.core | tail -50
 <output omitted...>
 Ready
 CLI ready
 9       ((
 Y       (l
 Y       (X
         (file /usr/V size 214,
 ocks
 Assert error in mgt_cli_callback(), mgt_cli.c line 322:
   Condition(i <= cp->nbuf) not true.
   errno = 0 (Unknown error: 0)
 (void*)0              )>
 t child pid 44959
         (rolling(2)...
 Child starts
 managed to mmap 2147479552 bytes of 2147479552
 Ready
 CLI ready
  ]      (
         ( ]     (
         ( ]     (
 Y       (lS
         (lS
 varnishd
 <output omitted...>

 }}}

-- 
Ticket URL: <http://varnish.projects.linpro.no/ticket/93>
Varnish <http://varnish.projects.linpro.no/>
The Varnish HTTP Accelerator