Re: We still experience DIRMINA-1006 despite using 2.0.13

[Buddy Butterfly]

Hi,

if I understand it correct, you are telling me that there will be no
sultion for it? Or did I get it wrong. I am lost. This problem strikes
us heavily! Project responsible is over the top :-(.

We saw 4 lines of this type:

Line 41771: 2016-02-28 01:52:01,147 [   NioProcessor-2358] WARN
ing.AbstractPollingIoProcessor$Processor  run - Create a
new selector. Selected is 0, delta = 0

Line 71754: 2016-02-28 04:27:14,918 [   NioProcessor-3238] WARN
ing.AbstractPollingIoProcessor$Processor  run - Create a
new selector. Selected is 0, delta = 0

Line 74744: 2016-02-28 06:52:45,293 [   NioProcessor-4054] WARN
ing.AbstractPollingIoProcessor$Processor  run - Create a
new selector. Selected is 0, delta = 0

Line 87830: 2016-02-28 17:24:35,806 [   NioProcessor-7346] WARN
ing.AbstractPollingIoProcessor$Processor  run - Create a
new selector. Selected is 0, delta = 0


What would be a workaround?

Tnx a lot!
Buddy


Am 28.02.2016 um 11:33 schrieb Emmanuel Lécharny:
> Le 27/02/16 18:08, Buddy Butterfly a écrit :
>> Obviously netty fixed this with a workaround.
> 
> Thanks for the interesting analysis and pointers.
> 
> It seems that in some case, we are still looping on the select().
> 
> Can you do one more thing to confirm that it's teh real problem ? Set
> the log to WARN, you should get messages like :
> 
> Create a new selector. Selected is 0, delta = XXX
> 
> This would be the signal that we detected a stalled selector (ie a
> selector that return immediately on select(), even if no event has
> arrived, leading to a fast loop eating all the CPU). In thsi case, we
> recreate a new selector. In your case, and accordingly to the bug
> report, there might be a case where creating a new selector does not
> solve the issue.
> 
> Many thanks !
> 
> 

Ready for release 1.2 (?)

[elijah baley]

Hi everybody,

I feel we have reached a respectable threshold of features and bug fixes to 
warrant a 1.2 release - especially in view of the SSHD-639 and SSHD-646 fixes 
which hopefully will be released as 1.1.1 soon. I therefore would like to ask 
those of you who posted bugs or feature requests to clone the latest version 
from https://github.com/apache/mina-sshd and use it in order to make sure your 
specific bug has been fixed or that the feature you requested is now available. 
After a reasonable wait time, if no special bugs or features are reported I 
intend to call for a vote on releasing version 1.2 - I would of course 
appreciate positive feedbacks confirming the bug fix or feature request that 
each of you has posted.

Thanks,
Lyor G.

Re: We still experience DIRMINA-1006 despite using 2.0.13

[Emmanuel Lécharny]

Le 27/02/16 18:08, Buddy Butterfly a écrit :
> Obviously netty fixed this with a workaround.

Thanks for the interesting analysis and pointers.

It seems that in some case, we are still looping on the select().

Can you do one more thing to confirm that it's teh real problem ? Set
the log to WARN, you should get messages like :

Create a new selector. Selected is 0, delta = XXX

This would be the signal that we detected a stalled selector (ie a
selector that return immediately on select(), even if no event has
arrived, leading to a fast loop eating all the CPU). In thsi case, we
recreate a new selector. In your case, and accordingly to the bug
report, there might be a case where creating a new selector does not
solve the issue.

Many thanks !