[zeromq-dev] Improving zeromq in OOM conditions

[Martin Sustrik]

On 05/17/2011 09:36 AM, Blair Bethwaite wrote:
> On 17 May 2011 17:10, Martin Sustrik<sustrik at 250bpm.com>  wrote:
>> IMHO that kind of thing is so complex that it's not even worth trying,
>> but you are free to give it a shot.
>
> This may be a fair opinion, but there is a problem with this response
> in terms of development going forward. Does this indicate that future
> work (e.g. in master by you, Martin) wouldn't encompass handling OOM
> situations? If so, it would seem there is little point doing the work
> in the first place...

If I get functional patches, I will sure apply them. But I am a bit 
skeptical about the whole thing as with OOM even the recovery measures 
are likely to fail because of OOM :) As explained in previous email, I 
believe that the only real solution is to never hit OOM in the first place.

Martin