[rabbitmq-discuss] Long timeout if server host becomes unreachable

Tue Oct 8 10:58:00 BST 2013

On oct 8, 2013, at 1:49 p.m., Oleg Lyalikov <oleg.lyalikov at gmail.com> wrote:

> So for me it looks like the RabbitMQ client library should provide possibility to set such write timeout - maybe using the same Heartbeat thread (for now this thread is blocked on "writeFrame" method like all other threads waiting freeing the lock on outputStream object).
> 
> By the way I still cannot imagine any workaround for this issue but it's really critical for us. Do you think there are some?

Have you tried using heartbeats with a low value (say, single digit seconds)? It was created
exactly for the case when a peer goes away and TCP retransmission covers the fact.

RabbitMQ Java client pretty much has to synchronize writes to the socket. You can share connection
between threads w/o realizing so, e.g. by publishing from a consumer callback such
as handleDelivery.

MK

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20131008/14a20fc1/attachment.pgp>