You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@ignite.apache.org by VeenaMithare <v....@cmcmarkets.com> on 2020/04/13 22:32:35 UTC

Re: Local node terminated after segmentation

Hi Ilya, 

How can a node reachability resolver or Tcp Segmentation resolver help in
discovering segmentation due to GC pauses ? What is the best way to discover
segmentation on a node due to GC pauses ?



regards,
Veena



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: Local node terminated after segmentation

Posted by VeenaMithare <v....@cmcmarkets.com>.

Thanks Evgenii 



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: Local node terminated after segmentation

Posted by Evgenii Zhuravlev <e....@gmail.com>.

Hi Veena,

There is a message in the logs:
 [WARNING][jvm-pause-detector-worker][IgniteKernal] Possible too long JVM
pause: 8023 milliseconds.

In most cases, it is a sign of the long GC pause. Of course, this JVM pause
can be related to the problems with a virtual environment or something
else, but usually it's GC. You can collect GC logs to make sure that it's
GC.
If this JVM pause is longer than failureDetectionTimeout, then, node can be
kicked from the cluster.

Evgenii

пн, 13 апр. 2020 г. в 15:32, VeenaMithare <v....@cmcmarkets.com>:

> Hi Ilya,
>
> How can a node reachability resolver or Tcp Segmentation resolver help in
> discovering segmentation due to GC pauses ? What is the best way to
> discover
> segmentation on a node due to GC pauses ?
>
>
>
> regards,
> Veena
>
>
>
> --
> Sent from: http://apache-ignite-users.70518.x6.nabble.com/
>