Re: [infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-18 Thread Galder ZamarreƱo
You do see InterruptedException's actually: https://gist.github.com/1089760 On Jul 15, 2011, at 1:39 PM, Dan Berindei wrote: I was surprised because I would have thought your change would replace each IllegalStateException in the log with an InterruptedException, so I wouldn't have expected

Re: [infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-18 Thread Sanne Grinovero
2011/7/18 Galder ZamarreƱo gal...@redhat.com: You do see InterruptedException's actually: https://gist.github.com/1089760 Unless that test is explicitly verifying the behaviour when those exceptions are thrown, I see it seems to pass properly. So if it's a non-critical / recoverable condition,

Re: [infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-15 Thread Sanne Grinovero
Sorry for not being clear - I'm not assuming that this has anything to do with the InterruptedException/EOFs being still logged, just that the output is less noisy if I remove these which where pointless. Sanne 2011/7/14 Dan Berindei dan.berin...@gmail.com: TBH I'm not sure why the patch helped

Re: [infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-15 Thread Dan Berindei
I was surprised because I would have thought your change would replace each IllegalStateException in the log with an InterruptedException, so I wouldn't have expected the log to be any less noisy. I mean you don't stop the rehashing during invalidation, you just skip invalidation and go to the

[infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-14 Thread Sanne Grinovero
Hello, I was looking into the many stacktraces being thrown from the testsuite, this one caught my attention (see below). Can't we just discard such errors? if the cache is stopping, or stopped already, we shouldn't really care for invalidations - especially stacktraces and exceptions being

Re: [infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-14 Thread Manik Surtani
Patch looks good. Go ahead and issue a pull req. On 14 Jul 2011, at 16:26, Sanne Grinovero wrote: Hello, I was looking into the many stacktraces being thrown from the testsuite, this one caught my attention (see below). Can't we just discard such errors? if the cache is stopping, or

Re: [infinispan-dev] IllegalStateException when receiving invalidation messages on a stopped/stopping cache

2011-07-14 Thread Dan Berindei
TBH I'm not sure why the patch helped with the error messages in the log, I always thought that doing this would just replace the IllegalStateException with an InterruptedException (since do some more stuff after invalidation). Of course the best solution would be to not do rehashing at all when