[Varnish] #488: child process restart with "Missing errorhandling code in sma_alloc()" error

[Varnish]

#488: child process restart with "Missing errorhandling code in sma_alloc()"
error
----------------------+-----------------------------------------------------
 Reporter:  flyincat  |       Owner:  phk                     
     Type:  defect    |      Status:  new                     
 Priority:  high      |   Milestone:                          
Component:  varnishd  |     Version:  2.0                     
 Severity:  major     |    Keywords:  storage_malloc.c restart
----------------------+-----------------------------------------------------
 I'm running with varnish 2.0.4, as I found that my varnish restart the
 child process every few hours(or minutes). Meanwhile I got this error in
 /var/log/message.My cache server now get just 400-500req/s at the peak.
 {{{
 Apr 15 06:05:34 varnish /srv/vcache[9292]: Child (9293) died signal=6
 Apr 15 06:05:34 varnish /srv/vcache[9292]: Child (9293) Panic message:
 Missing errorhandling code in sma_alloc(), storage_malloc.c line 79:
 Condition((sma->s.ptr) != 0) not true.  errno = 12 (Cannot allocate
 memory)  thread = (cache-worker)sp = 0x71a42004 {   fd = 4218, id = 4218,
 xid = 2057707839,   client = 116.30.99.238:12844,   step = STP_LOOKUP,
 handling = hash,   ws = 0x71a4204c {      id = "sess",     {s,f,r,e} =
 {0x71a42534,,+672,(nil),+16384},   },     worker = 0x1f5f30d8 {     },
 vcl = {       srcname = {         "input",         "Default",       },
 }, },
 }}}

 BTW: I started varnish by

 {{{
 /usr/varnish/sbin/varnishd -n /srv/vcache -f
 /usr/varnish/etc/varnish/default.vcl -a 10.10.10.25:80 -s malloc,3328M -u
 www -w2,51200,30 -T 127.0.0.1:3500 -p client_http11 on -p thread_pools 4
 -p lru_interval 3600 -p obj_workspace 10240 -h classic,19997
 }}}

-- 
Ticket URL: <http://varnish.projects.linpro.no/ticket/488>
Varnish <http://varnish.projects.linpro.no/>
The Varnish HTTP Accelerator