You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@pulsar.apache.org by Daniel Pono Takamori <po...@apache.org> on 2018/09/17 01:11:26 UTC
pulsar_precommit jobs taking up too much space
Hey folks,
Doing disk cleanups on our Jenkins nodes [0], and found about 40
jpulsar_precommit* jobs on a single node taking up 2-5gb each. We've
moved to purging 15 day old files [1] on Jenkins nodes. I'm wondering
if we can be more aggressive on purging these old precommit jobs, or
if there is something on your end which would also serve to reduce the
disk overhead on keeping them around.
Thanks for any ideas!
-Pono on behalf of Infra
[0] - https://issues.apache.org/jira/browse/INFRA-17024
[1] - https://github.com/apache/infrastructure-puppet/pull/1159/commits/8438937327a63c2c5999baa3fefe956d659dedec
Re: pulsar_precommit jobs taking up too much space
Posted by Sijie Guo <gu...@gmail.com>.
H Daniel,
I reduced the number of days to keep for the build. That should clean up
the job builds more aggressively.
- Sijie
On Sun, Sep 16, 2018 at 6:11 PM Daniel Pono Takamori <po...@apache.org>
wrote:
> Hey folks,
> Doing disk cleanups on our Jenkins nodes [0], and found about 40
> jpulsar_precommit* jobs on a single node taking up 2-5gb each. We've
> moved to purging 15 day old files [1] on Jenkins nodes. I'm wondering
> if we can be more aggressive on purging these old precommit jobs, or
> if there is something on your end which would also serve to reduce the
> disk overhead on keeping them around.
>
> Thanks for any ideas!
> -Pono on behalf of Infra
>
> [0] - https://issues.apache.org/jira/browse/INFRA-17024
> [1] -
> https://github.com/apache/infrastructure-puppet/pull/1159/commits/8438937327a63c2c5999baa3fefe956d659dedec
>
Re: pulsar_precommit jobs taking up too much space
Posted by Sijie Guo <gu...@gmail.com>.
H Daniel,
I reduced the number of days to keep for the build. That should clean up
the job builds more aggressively.
- Sijie
On Sun, Sep 16, 2018 at 6:11 PM Daniel Pono Takamori <po...@apache.org>
wrote:
> Hey folks,
> Doing disk cleanups on our Jenkins nodes [0], and found about 40
> jpulsar_precommit* jobs on a single node taking up 2-5gb each. We've
> moved to purging 15 day old files [1] on Jenkins nodes. I'm wondering
> if we can be more aggressive on purging these old precommit jobs, or
> if there is something on your end which would also serve to reduce the
> disk overhead on keeping them around.
>
> Thanks for any ideas!
> -Pono on behalf of Infra
>
> [0] - https://issues.apache.org/jira/browse/INFRA-17024
> [1] -
> https://github.com/apache/infrastructure-puppet/pull/1159/commits/8438937327a63c2c5999baa3fefe956d659dedec
>