You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ignite.apache.org by Dmitriy Pavlov <dp...@gmail.com> on 2018/09/02 07:10:48 UTC

Zookeeper mass timeouts

Hi,

First of all I apologize for mass emails came from zookeeper timeout
failures.

Both failures and the bot should be researched. I believe we can improve
the bot notification rules to avoid mass emails in case of flaky timeouts.

I believe it is better solution to make rules perfect instated of creating
separate channel.

Sincerely,
Dmitriy Pavlov

Re: Zookeeper mass timeouts

Posted by Alexey Goncharuk <al...@gmail.com>.
Dmitriy,

The zookeeper timeouts were caused by my commit (I looked through the wrong
PR when was checking tests), already reverted from master.

As for the bot, does it already have the logic to detect continuous
timeouts and send notification only after a successful run? If not, I guess
we should put it on our helper roadmap because this will not be the last
timeout change.

--AG

вс, 2 сент. 2018 г. в 10:11, Dmitriy Pavlov <dp...@gmail.com>:

> Hi,
>
> First of all I apologize for mass emails came from zookeeper timeout
> failures.
>
> Both failures and the bot should be researched. I believe we can improve
> the bot notification rules to avoid mass emails in case of flaky timeouts.
>
> I believe it is better solution to make rules perfect instated of creating
> separate channel.
>
> Sincerely,
> Dmitriy Pavlov
>