<p dir="ltr">The main thing I used to see was VM memory exhaustion due to 0MQ being just too damn fast. <br>
But that was more of a "No AQM in my code" issue :-)</p>
<p dir="ltr">Sent via smoke signals...<br>
Mahesh Paolini-Subramanya<br>
That tall bald Indian guy...</p>
<div class="gmail_quote">On Jan 30, 2014 5:46 PM, "Michael Truog" <<a href="mailto:mjtruog@gmail.com">mjtruog@gmail.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On 01/30/2014 02:22 PM, Garrett Smith wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Thu, Jan 30, 2014 at 4:20 PM, Benoit Chesneau <<a href="mailto:bchesneau@gmail.com" target="_blank">bchesneau@gmail.com</a>> wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Thu, Jan 30, 2014 at 10:58 PM, Garrett Smith <<a href="mailto:g@rre.tt" target="_blank">g@rre.tt</a>> wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Speed kills.<br>
</blockquote>
What do you mean? I like better to have a stable solution? How did you<br>
benchmarked it?<br>
</blockquote>
We are in agreement on stability :)<br>
<br>
There's a benchmark tool in the library.<br>
<br>
Garrett<br>
______________________________<u></u>_________________<br>
erlang-questions mailing list<br>
<a href="mailto:erlang-questions@erlang.org" target="_blank">erlang-questions@erlang.org</a><br>
<a href="http://erlang.org/mailman/listinfo/erlang-questions" target="_blank">http://erlang.org/mailman/<u></u>listinfo/erlang-questions</a><br>
<br>
</blockquote>
I guess your concerns about stability are confusing. The main issues are bugs within the ZeroMQ source code:<br>
<a href="https://zeromq.jira.com/browse/LIBZMQ-478" target="_blank">https://zeromq.jira.com/<u></u>browse/LIBZMQ-478</a><br>
<a href="https://zeromq.jira.com/browse/LIBZMQ-496" target="_blank">https://zeromq.jira.com/<u></u>browse/LIBZMQ-496</a><br>
<br>
The LIBZMQ-496 was fixed in 3.2.3 while LIBZMQ-478 remains unresolved. I see no reason to doubt the stability of erlzmq2, since these errors are handled in a way that causes no instability within the Erlang VM. The erlzmq2 is using a separate thread, so it doesn't have the Erlang scheduler lock-up issue that remains a problem for all long-running NIF/BIF functions.<br>
<br>
So, what instability are you referring to?<br>
______________________________<u></u>_________________<br>
erlang-questions mailing list<br>
<a href="mailto:erlang-questions@erlang.org" target="_blank">erlang-questions@erlang.org</a><br>
<a href="http://erlang.org/mailman/listinfo/erlang-questions" target="_blank">http://erlang.org/mailman/<u></u>listinfo/erlang-questions</a><br>
</blockquote></div>