Automatically installed tools get updated on every use

2021-04-14 Thread Bjoern Hinrichs 

Hi all,

since upgrading to Jenkins 2.263.4 we're experiencing the following issue:

We use Jenkins's feature to automatically install several build tools. 
In this case, in Global Tool Configuration we have a Maven installation 
"apache-maven-3.5.4" that uses "Extract *.zip/*.tar.gz" to install a zip 
file located on a samba share. When a job (we use "conventional" jobs 
here, not pipeline jobs) using this Maven installation runs, the zip 
file gets copied to the agent and unzipped inside the correct 
sub-directory  apache-maven-3.5.4.


Unfortunately, since our upgrade to 2.263.4, this happens every time, 
even if the installer has already been run on the same agent. Inside the 
sub-directory apache-maven-3.5.4 we can find a file .timestamp with the 
same timestamp as the zip file on our samba share.


Most of our agents use more than one executor. When two jobs run in 
parallel, we've seen it happen that one job "updates" the Maven 
installation while another job is actively using it. The job that's past 
the installation step then fails.


We didn't encounter this issue with Jenkins 2.107.3.

I couldn't find any hints on this issue in the changelog files for newer 
versions of Jenkins, nor was I able to find any hints on this on 
issues.jenkins.io.
Before I file a report myself: Has anyone else seen this behavior? Any 
workarounds?


Best regards
Björn Hinrichs

--
You received this message because you are subscribed to the Google Groups "Jenkins 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/6ef1d14d-2837-7ae5-fc05-f21a273846d8%40btc-es.de.

Re: Statistics plugin to help identify bottlenecks

2019-08-02 Thread Bjoern Hinrichs 

Hi Mark,

thanks for your input.
DevOptics seems promising, but as far as I can see DevOptics can't be 
hosted on-site, i.e. we would be dependent on (and sending internal data 
to) CloudBee's servers? Unfortunately, that's not possible for us.


Regards
Björn

Am 8/1/2019 um 5:23 PM schrieb Mark Waite:
I use CloudBees DevOptics to track utilization by label.  The label 
tracking portion ("Run Insights") is free.  I used it to confirm that 
my cluster is "unbalanced" and needs much more Windows execution 
capacity in order to be balanced. See more at 
https://www.cloudbees.com/products/cloudbees-devoptics


Here is a screenshot of a portion of the report showing my lack of 
Windows executors.  The top nodes listed are individual windows 
agents, then the bottom of the image is the aggregate "windows" label 
(maintained by the platform labeler plugin).


devoptics-labels.png

On Thu, Aug 1, 2019 at 9:13 AM Bjoern Hinrichs 
mailto:bjoern.hinri...@btc-es.de>> wrote:


Hi,

we use Matrix jobs for our build and test pipeline.
All of our test agents carry labels describing OS version, OS
language,
whether other tools are installed (and in which version), ...

To identify bottlenecks we need a plugin that allows us to see
statistics like utilization of our nodes and (especially) wait
times in
queue per label.

We tried several plugins, but couldn't find one that fits our
needs; the
last one we tried was Cluster Statistics Plugin, which gives an
average
wait time in queue, but unfortunately not per label.

Are there any plugins out there that might work for us?

Thanks in advance
Björn Hinrichs

-- 
You received this message because you are subscribed to the Google

Groups "Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to jenkinsci-users+unsubscr...@googlegroups.com
<mailto:jenkinsci-users%2bunsubscr...@googlegroups.com>.
To view this discussion on the web visit

https://groups.google.com/d/msgid/jenkinsci-users/39a95742-f7f7-bbf5-8244-7d9ec8b21237%40btc-es.de.



--
Thanks!
Mark Waite
--
You received this message because you are subscribed to the Google 
Groups "Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send 
an email to jenkinsci-users+unsubscr...@googlegroups.com 
<mailto:jenkinsci-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/CAO49JtEWbku0qFsYAXcwQuMpczGmubFmxC0sR_U5ZTOFxwrKGw%40mail.gmail.com 
<https://groups.google.com/d/msgid/jenkinsci-users/CAO49JtEWbku0qFsYAXcwQuMpczGmubFmxC0sR_U5ZTOFxwrKGw%40mail.gmail.com?utm_medium=email&utm_source=footer>.


--
You received this message because you are subscribed to the Google Groups "Jenkins 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/dc5909a4-1df5-aee3-28aa-119bfa849d04%40btc-es.de.

Statistics plugin to help identify bottlenecks

2019-08-01 Thread Bjoern Hinrichs 

Hi,

we use Matrix jobs for our build and test pipeline.
All of our test agents carry labels describing OS version, OS language, 
whether other tools are installed (and in which version), ...


To identify bottlenecks we need a plugin that allows us to see 
statistics like utilization of our nodes and (especially) wait times in 
queue per label.


We tried several plugins, but couldn't find one that fits our needs; the 
last one we tried was Cluster Statistics Plugin, which gives an average 
wait time in queue, but unfortunately not per label.


Are there any plugins out there that might work for us?

Thanks in advance
Björn Hinrichs

--
You received this message because you are subscribed to the Google Groups "Jenkins 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/39a95742-f7f7-bbf5-8244-7d9ec8b21237%40btc-es.de.