Bug? Related to ticket 547

[Poul-Henning Kamp]

In message <D26F7E3F42600549A27A4433EB952E840403194B at sbs03.ecnext03.local>, "Ji
m Hayter" writes:
>I'm sorry, but I'm not sure how/where to report this.
>
>I'm running varnish-2.0.6 on Solaris 10.  I had the issue noted in
>http://varnish.projects.linpro.no/ticket/547 when testing 2.0.5, which
>was fixed in 2.0.6.  I just tried 2.0.6 on a production system and ran
>into the issues noted in the varnishd output below.  It appears to be a
>similar call to setsockopt to the one noted in ticket 547.  This error
>occurred every 3-6 minutes.

It seems to be Solaris returning EBADF if the connection is closed
before it gets around to do something to it.

Can you open a ticket with this report, I'll try to find a way to
add checks for this.

Poul-Henning

>
>Any input is welcomed.  We had hoped to run this on one of our
>production web servers all weekend as a prelude to putting it into full
>production.
>
>Thanks,
>Jim
>
>Varnishd output
>---------------
>...
>Child (21784) said ", 1, 3)
>Child (21784) not responding to ping, killing it.
>Child (21784) not responding to ping, killing it.
>Child (21784) not responding to ping, killing it.
>Child (21784) died signal=6 (core dumped)
>Child (21784) Panic message: Assert error in VCA_Prep(),
>cache_acceptor.c line 148:
>  Condition((setsockopt(sp->fd, 0xffff, 0x0080, &linger, sizeof linger))
>== 0) not true.
>errno = 9 (Bad file number)
>thread = (cache-worker)
>sp = 8f29974 {
>  fd = 84, id = 84, xid = 0,
>  client = 66.68.180.213:2086,
>  step = STP_FIRST,
>  handling = error,
>  restarts = 0, esis = 0
>  ws = 8f299c0 {
>    id = "sess",
>    {s,f,r,e} = {8f2a470,+19,0,+32768},
>  },
>  http[req] = {
>    ws = 0[]
>  },
>  worker = 731cdee0
>},
>
>
>Child cleanup complete
>child (21816) Started
>Child (21816) said Closed fds: 4 5 25 26 28 29
>...
>_______________________________________________
>varnish-misc mailing list
>varnish-misc at projects.linpro.no
>http://projects.linpro.no/mailman/listinfo/varnish-misc
>

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.