If you use CloudBees Jenkins and their Folder Plus plugin, there is a
feature on Folders that allows exactly this. There is a Controlled
Slave/Folder feature where you can set a mapping between one or more
folders to a slave. Such that only jobs in the mapped folder can execute
on the mapped
Hello,
I am putting the vSphere Cloud Plugin up for adoption - would any one like
ownership of it? If so please contact me directly via email.
Wiki page: https://wiki.jenkins-ci.org/display/JENKINS/vSphere+Cloud+Plugin
Source: https://github.com/jenkinsci/vsphere-cloud-plugin
Jason Swager
This should
help:
https://github.com/jenkinsci/pipeline-examples/blob/master/docs/BEST_PRACTICES.md#groovy-gotchas
On Friday, September 30, 2016 at 10:26:40 PM UTC-7, Sam K wrote:
>
> Anyone?
>
> On Friday, September 30, 2016 at 11:54:32 AM UTC-7, Sam K wrote:
>>
>> I have 4 sets of IP
Pipelines are very nice for their durable steps; when you've got persistent
slaves, it's rather nice to have a build pickup after an unexpected Jenkins
shutdown.
But if you're using temporary slaves - like from the Docker plugin - those
slaves often won't exist after the Jenkins restart. Is
We use the Docker Plugin
(https://wiki.jenkins-ci.org/display/JENKINS/Docker+Plugin) with Docker
Swarm. Since a Swarm Master presents itself just like a normal Docker
host, Jenkins doesn't see the difference and and uses it exactly the same.
On Wednesday, March 16, 2016 at 12:07:16 PM UTC-7,
Take a look at the NodeLabel Parameter plugin
(https://wiki.jenkins-ci.org/display/JENKINS/NodeLabel+Parameter+Plugin).
This offers a way to start a job with a parameter that is a label and the
job can run on all matching slaves.
On Tuesday, January 19, 2016 at 8:13:43 AM UTC-8, Julianus
I'm using Jenkins v1.642, running on Windows 2012 Server as a service (not
behind a web server). There are HTML and CSS files in the userContent that
provided a custom dashboard. That dashboard no longer renders properly. A
check of the browser console showed "Blocked script execution in
My apologies. After re-reading through the doc, I saw my mistake. Thank
you!
On Wednesday, December 23, 2015 at 7:07:30 AM UTC-8, Daniel Beck wrote:
>
>
> > On 23.12.2015, at 15:40, Jason Swager <jsw...@alohaoi.com >
> wrote:
> >
> > tried th
(sort of like rebooting your
slave mid-build), the build will fail.
To work around this, split the job across two jobs. First job reverts the
targeted slave VM, second job runs on the targeted slave VM.
Jason Swager
On Monday, November 23, 2015 at 7:09:29 AM UTC-8, Andreas Tscharner wrote
If you're using the Docker plugin
(https://wiki.jenkins-ci.org/display/JENKINS/Docker+Plugin), I found the
env vars worked fine when they were setup to be available to incoming .ssh
connections. That sometimes meant using a different config file within the
container to prep the variables.
A
Have you considered using the Docker plugin
(https://wiki.jenkins-ci.org/display/JENKINS/Docker+Plugin)? That plugin
seems to be a closer fit for what you might want. The entire slave is in a
container and consequently, all the PreBuild plugins and Publisher plugins
work like they would on a
It most likely means that you are using a version of esxi that doesn't have
all the feature enabled, such as the free version of the esxi or possibly a
time limited or trial version.
The plugin is designed to work with vSphere - talking to a vCenter. ESXi
standalone exposes the same API, but
Do you have a common file with the same fingerprint between jobs A, B, and
C? I have used this type of promotion before.
One thing you may want to check; do the fingerprints on job A show that the
fingerprint originated in job A? And are Job B and Job C listed with the
same fingerprint?
On
, Jason Swager wrote:
Do you have a common file with the same fingerprint between jobs A, B,
and C? I have used this type of promotion before.
One thing you may want to check; do the fingerprints on job A show that
the fingerprint originated in job A? And are Job B and Job C listed
The workflow plugin(s) have been very nice. They have made it far easier
to handle the build and test parts of our pipelines.
But where the workflow has not done as well is the publishers - only a
couple seem supported at this time. In particular, we use CheckStyle,
FindBugs, compiler
I'm having the same problem. I'm trying to install Jenkins 1.579 into
Tomcat 7 which is running on a Synology NAS server. The only thing I can
find is the exception that's listed above.
Anyone know how to solve this problem?
Jason
On Saturday, July 5, 2014 11:50:25 AM UTC-7, Swarup Rakshit
, at 23:42, Jason Swager jsw...@alohaoi.com javascript:
wrote:
I'm having the same problem. I'm trying to install Jenkins 1.579 into
Tomcat 7 which is running on a Synology NAS server. The only thing I can
find is the exception that's listed above.
Anyone know how to solve this problem
I don't believe that this is specifically a problem with the vSphere Cloud
plugin. I believe that this is a problem with your installed Java
attempting to contact the vSphere server using SSL. Recent versions of
Java have been deprecating and disabling (be default) certain low-security
Try restarting your slaves. There was an issue with EnvInject where global
variable changes wouldn't propagate to the slave unless the slave was fully
disconnected and reconnected.
On Friday, January 17, 2014 8:19:34 AM UTC-8, Allen Cronce wrote:
Hi all,
We've been trying to use Jenkins
.
I guess that we can reboot as a work around. We don't change back and
forth between the trunk and branches that often.
But is there a more permanent solution? Maybe the problem you mentioned
has been fixed in an update?
On Friday, January 17, 2014 9:47:52 AM UTC-8, Jason Swager wrote
Have you considered using the Custom Tool plugin (
https://wiki.jenkins-ci.org/display/JENKINS/Custom+Tools+Plugin)? We also
used to use the approach of copying files from /userContent to the slaves
on a per-build basis.
But Custom Tools made it rather easier. First, the tool would only be
See the Rebuild
pluginhttps://wiki.jenkins-ci.org/display/JENKINS/Rebuild+Plugin.
That plugin should do exactly what you want.
On Tuesday, November 19, 2013 11:39:00 AM UTC-8, Vishal wrote:
Hi All -
I am using Paramerterized build plugin. I have almost 10-12 parameters
that i need to
I've been looking for something like this for a while. I have tests that
require two (or more) slaves to be executed properly. But there doesn't
seem to be any plugin that currently does something like that.
We have a workaround, but it's not very robust at all. We have one job
that starts
Jenkins installation
and it works better than relying on the other trigger types. We can
control the logic very tightly and with all the quirks that we want, but
still use Jenkins to drive the overall process.
Jason Swager
On Tuesday, May 28, 2013 12:03:25 PM UTC-7, Jim Zajkowski wrote:
Hi all
If your slaves are VMs in a VMware vSphere/vCenter installation, you can
use the vSphere Cloud Plugin to revert the VM to a snapshot before starting
it up. Other virtualization cloud plugins will do a complete deploy of a
new VM.
Jason Swager
On Tuesday, January 29, 2013 11:06:18 PM UTC-8
Yes. Using the Promotions plugin, setup a promotion that automatically
triggers when all the downstream jobs have passed. The Promotions plugin
will expose that build as a permalink.
Use the Copy Artifacts plugin to pull artifacts from the permalink
associated with the promotion.
Jason
at the
particular slave that was running as a particular user.
I don't know of any way to run as multiple users on the Jenkins master
itself - but that was fine for us.
Jason Swager
On Friday, January 4, 2013 1:14:29 PM UTC-8, Adam Tong wrote:
Hi,
If it is complicated to make it personalized (each
I've been looking for a component - or a technique - in which two slaves
can be used for the same job. Since having one job run on two slaves isn't
feasible, it would probably be two (or more) jobs that each start on a
different slave at the same time. Having those jobs start at the same time
You can place arbitrary files in Jenkins userContent subdirectory, which
exposes them as retrievable via URL, via wget or curl. Or you can use the
Copy to Slave plugin and pull the files directly from Jenkins to a
workspace at the start the job.
A bit more odd, but you can have a job that
We do something similar with our build/test jobs. We do it by passing the
build number of build-project (and possibly the job name if multiple
builds feed to the same test jobs) to test-project. In turn,
test-project would pass that same build number to valgrind-project.
That way,
publishing this component like other
Jenkins components. It would be a great component to have if it works like
it seems to work.
On Sunday, November 25, 2012 2:08:21 PM UTC-8, Ivan Kalinin wrote:
Sure, here it is:
http://db.tt/reV8CziX
On Sunday, November 25, 2012 8:07:15 PM UTC+4, Jason
Having troubles getting a build environment built up. Could you publish
the jpi so that it could be loaded directly into Jenkins?
On Sunday, November 25, 2012 6:19:09 AM UTC-8, Ivan Kalinin wrote:
Hi there!
So have u tried that thing?
If its broken or something is missing I can try 2 fix
Throttle plugin with a category.
On Sunday, November 25, 2012 7:46:04 PM UTC-8, cliffano wrote:
Hi all,
I'm wondering what do you use nowadays to prevent multiple jobs from
building at the same time.
- lock and latches plugin
Its lock mechanism does not work when multiple jobs sharing
I would LOVE to see a plugin like this. If you could send me github
information after you make it available, I would be very grateful!
Thanks,
Jason
On Wednesday, November 21, 2012 10:40:36 PM UTC-8, Ivan Kalinin wrote:
Hi!
Actually, I've managed to develop a plugin for my own needs that
As far as I know, there is no plugin that lets you run individual build
steps on a different slave than the one that job originally started on.
But there ways to mimic this behavior.
We do something like this in our own tests. One job starts up; it has a
parameter that indicates another
This is what Slave Labels and the job option Restrict where this project
can be run are meant for. Set a specific label(s) to the slaves meant to
run the job, then restrict the job to just that label(s). Jenkins will
handle the rest. Our own Jenkins strategy relies heavily on this feature;
I agree that if you had a job with a conditional build step that had no
steps within it, then you could sort of expect to see this error. Easy
enough to workaround.
But the problem I'm seeing is happening when Jenkins starts up. We see the
typical starting up page, then we get this error.
Likewise. Is there a JIRA issue opened for this that we can upvote?
On Monday, November 19, 2012 5:32:43 AM UTC-8, alok kumar wrote:
Hi,
I am facing the following problem while trying to use conditional build
step plugin.
As a result of this, Jenkins fails to start.
If I
See
http://stackoverflow.com/questions/10177708/jenkins-on-windows-and-gui-tests-without-rdc
On Wednesday, October 31, 2012 3:26:35 AM UTC-7, Jeff MAURY wrote:
I'm using Jenkins to build an Eclipse based product and I have SWTBot
based tests that performs validation of the UI layer.
As
When marking a job as able to run concurrently on the same node, Jenkins
will append a @X (where X is the number of the concurrent build) to the
normal workspace directory to get a new, unique workspace directory.
Normally, this is all fine and good, but we've encountered a couple tools
that
As part of that batch command, add a SET to show you all the environment
variables. Then compare to a SET from the command prompt. I suspect that
you'll see several differences which account for msdev not working as you
would expect.
Jason
On Friday, October 19, 2012 2:29:47 PM UTC-7, Engler
On possibility that we've used for similar situations:
- Create a lock (Lock and Latches or Throttle plugins) for each slave
that you're going to test on.
- Set the slave up to accept only targeted jobs - that way other jobs
won't wander onto your slave. Also, the slave must be
The wiki page has the following text:
If you evaluate our value-add plugins in this manner and decide not to
purchase a license, you need to remove all the value-add plugins to keep
using it as a stock Jenkins installation.
This seems to imply that an evaluation license is only to evaluate,
of executors.
HTH
-Stephen
[/CloudBees hat]
On 25 September 2012 19:26, Jason Swager j.a.s...@gmail.com javascript:
wrote:
The wiki page has the following text:
If you evaluate our value-add plugins in this manner and decide not to
purchase a license, you need to remove all the value-add
We do something similar to this with the vSphere Cloud Plugin But instead
of cloning a new VM (which the plugin does not do), we revert a VM to a
previously clean state. The clean state already has the Jenkins slave
agent setup and automatically connecting when the VM starts. So when
235 9621
Sent from my T-Mobile G2
Ezt a levelet telefonról adták fel, ékezethibákat tartalmazhat.
2012.09.05. 15:53, Jason Swager j.a.s...@gmail.com javascript:
ezt írta:
Just to confirm, we are talking about the vSphere Cloud plugin?
https://wiki.jenkins-ci.org/display
The jobs do die when the server goes down.
On Thursday, September 6, 2012 5:59:40 AM UTC-7, Mandeville, Rob wrote:
Just to be sure: they don’t abort their jobs in this case?
--Rob
*From:* jenkins...@googlegroups.com javascript: [mailto:
jenkins...@googlegroups.com javascript:]
? And, this is not really a vSphere cloud but a single ESXi
server...
Garami Gábor
E-mail: gabor@hron.me javascript:
Tel: +36 20 235 9621
MSN: hr...@vipmail.hu javascript:
Skype: hron84
On Fri, Sep 7, 2012 at 3:31 PM, Jason Swager
j.a.s...@gmail.comjavascript:
wrote:
Your configurations
9621
MSN: hr...@vipmail.hu javascript:
Skype: hron84
On Wed, Sep 5, 2012 at 4:32 AM, Jason Swager
j.a.s...@gmail.comjavascript:
wrote:
How is the slave agent being started? Does the VM connect via Java Web
Start? Is Jenkins supposed to start the slave agent via SSH?
If via
https://issues.jenkins-ci.org/browse/JENKINS-14197 submitted
We've been running Jenkins 1.471 for a few days, and just updated a couple
plugins (XTrigger ones). After the restart, we found that for no job can
set the Restrict where this project can be run. Regardless of what value
that is enter and Saved, the field is blank when reloaded.
Has anyone
Take a look at
https://wiki.jenkins-ci.org/display/JENKINS/Throttle+Concurrent+Builds+Plugin
The Throttle Concurrent Builds plugin has a feature to specifically control
how many concurrent builds can run on one slave at a time. It should
resolve your problem if you only allow one current
I know that you can use the Jenkins CLI to offline a slave with the
disconnect-node command.
53 matches
Mail list logo
