Handling of conditions where server is down about java-gearman-service HOT 7 CLOSED

The client and workers have a "timeout period." They will not try to reconnect 
for that period of time. This is an attempt to avoid wasting resources, trying 
to reconnect on every submission. Look into setting the lost connection policy 
(http://java-gearman-service.googlecode.com/svn/javadoc/org/gearman/GearmanJobSe
rverPool.html#setLostConnectionPolicy(org.gearman.GearmanLostConnectionPolicy)) 
and setting the reconnect period 
(http://java-gearman-service.googlecode.com/svn/javadoc/org/gearman/GearmanJobSe
rverPool.html#setReconnectPeriod(long, java.util.concurrent.TimeUnit))

I'm 98% sure this is working correctly. However, I'm going to do a little 
testing on this functionality tonight, just to see if I can reproduce anything. 
If you could attach some source, it would be greatly appreciated.

The links did not turn out correctly...

Look into the following methods:

Setting the lost connection policy:
GearmanWorker#setLostConnectionPolicy(GearmanLostConnectionPolicy policy)
GearmanClient#setLostConnectionPolicy(GearmanLostConnectionPolicy policy)

Setting the default reconnect period:
GearmanWorker#setReconnectPeriod(long time, TimeUnit unit)
GearmanClient#setReconnectPeriod(long time, TimeUnit unit)

I'm able to reproduce this issue. I'll try to get a fix out as soon as possible.

This issue was closed by revision r93.

I've just released v0.5.2. Give that a shot, and please let me know if you find 
any more issues.


Thanks, omry. I appreciate you bringing this issue to my attention :)

going to try that out just now :).
thanks!

I will keep testing, but I`m likely to switch over anyway as it seems that you 
are actually developing this, unlike the owner of the other project which 
kinda' gave up.

Okay, now I consistently getting the correct error in my  GearmanSubmitHandler 
:FAILED_TO_CONNECT

you mentioned the reconnect policy.
does that apply to workers AND clients?
because you said clients should reconnect automatically, but I don't think its 
the behavior I am seeing.
if I am not trying to re-submit the job, I am seeing a single 
FAILED_TO_CONNECT, and then nothing.