You are viewing a plain text version of this content. The canonical link for it is here.
Posted to builds@apache.org by Hervé BOUTEMY <he...@free.fr> on 2014/04/19 11:14:34 UTC

Workspace size

Hi,

While looking at disk usage [1], which shows that ASF workspace is near 700GB 
(ouch!), I tried to find if some of the jobs I know are using more space than 
expected: so I looked at the second most resource intensive one = maven-wagon:
- Jobs = 58 MB
- Builds all = 58 MB
- Workspace = 72 GB

72 GB for workspace?
I suppose it is what is visible here  [2]

Does anybody know what is taking such a huge space? (I preferd not download 
the workspace using the provided link, since I suppose 72 GB download would 
cause harm to the whole Jenkins farm)

Is there something we should do on such disk usage?

Or is disk cheap and doesn't cause any harm on overall farm performance, so we 
should focus on other points?

Regards,

Hervé

[1] https://builds.apache.org/plugin/disk-usage/

[2] https://builds.apache.org/job/maven-wagon/

Re: Workspace size

Posted by Olivier Lamy <ol...@apache.org>.
On 19 April 2014 19:14, Hervé BOUTEMY <he...@free.fr> wrote:
> Hi,
>
> While looking at disk usage [1], which shows that ASF workspace is near 700GB

I wanted to write a groovy script to reconfigure all jobs for a
maximum of last 10 builds history.
Some are not configured to cleanup builds history.

> (ouch!), I tried to find if some of the jobs I know are using more space than
> expected: so I looked at the second most resource intensive one = maven-wagon:
> - Jobs = 58 MB
> - Builds all = 58 MB
> - Workspace = 72 GB
>
> 72 GB for workspace?
I fixed it to not use anymore a maven repo in the workspace
(.repository) but a shared one!
So should be better after the next build.

> I suppose it is what is visible here  [2]
>
> Does anybody know what is taking such a huge space? (I preferd not download
> the workspace using the provided link, since I suppose 72 GB download would
> cause harm to the whole Jenkins farm)
>
> Is there something we should do on such disk usage?
>
> Or is disk cheap and doesn't cause any harm on overall farm performance, so we
> should focus on other points?
>
> Regards,
>
> Hervé
>
> [1] https://builds.apache.org/plugin/disk-usage/
>
> [2] https://builds.apache.org/job/maven-wagon/



-- 
Olivier Lamy
Ecetera: http://ecetera.com.au
http://twitter.com/olamy | http://linkedin.com/in/olamy