[zeromq-dev] Assertion failure in pull.cpp

[Pieter Hintjens]

OK, I was using the wrong socket (!) in one of the test steps, which
explains most of the problems I was seeing.  If one connects A rather
than B, it's normal that writing to B won't do much.

I'm now making a minimal test case for the assertion failure.

On Fri, Aug 20, 2010 at 1:41 PM, jon dyte <jon.dyte at gmail.com> wrote:
>
>
> On 20 August 2010 12:34, Pieter Hintjens <ph at imatix.com> wrote:
>>
>> On Fri, Aug 20, 2010 at 1:24 PM, jon dyte <jon.dyte at gmail.com> wrote:
>>
>> > It definately asserted on one run, (it also asserted somewhere else once
>> > running but that might have been memory)
>> > Have you tested that example I posted?
>>
>> No, sorry.  I've been focusing on trying to get a minimal test case.
>>
>
> ok not to worry. I'm still unclear whether it solves Oliver's issue....
>
>>
>> > Maybe we have both found an issue.
>
>
>>
>> I think we've found an issue.  I've got another test case here:
>
>
> looks that way, but last night I was unsure  and eventually I got something
> working....
>
>>
>> http://gist.github.com/540112.  This really carefully synchronizes
>> binds and connects so it can't possibly fail, but it does.
>>
>
> what if you put a sleep(3) or somesuch at line 49
>
>>
>> That shows the same problem with PUB/SUB sockets and I can provoke an
>> assertion failure in sub.cpp.
>>
>> It seems to affect 2.0.7 and master and wip-shutdown.
>>
>>
>> -
>> Pieter Hintjens
>> iMatix - www.imatix.com
>
>



-- 
-
Pieter Hintjens
iMatix - www.imatix.com