In message <[email protected]>, Paul 
Wright writes:
>On 13 February 2010 20:26, Poul-Henning Kamp <[email protected]> wrote:
>
>> I'm still not 100% convinced that EBADF is not me screwing something
>> up, so I have gone over the poll-waiter and made it even more
>> paranoid and conservative than before.
>>
>> Can I get you to give -trunk r4561 a swing, and see if that changes
>> anything ?
>
>I just ran r4561 for about 10 minutes and observed the same error with
>a client halfway through a KeepAlive session:

Is there any way (dtrace ?) you can log which systemcalls happen,
I would really like to be 100% sure that varnish does not close that
socket by mistake...

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
[email protected]         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.
_______________________________________________
varnish-misc mailing list
[email protected]
http://projects.linpro.no/mailman/listinfo/varnish-misc

Reply via email to