Hi,
Oh, email-ext, one of my current favorite topics
I am calling it a little differently and that's working (for me it's jelly
instead of groovy but groovy worked before also):
def mailbody = '${JELLY_SCRIPT, template="template.jelly"}'
def mailsubject = 'Jenkins Build ' + currentBuild.ful
Hi Mark,
thanks for your support!
The URL is https:// the credentials stored are Username/password (with ID
and key of an API-key of an internal gitlab-site).
As mentioned: it is working in the configuration of my shared library and
from the checkout scm: scmGit(...)-step.
Ah, but in the library
Hi,
we are currently transforming all our SCMs to git. That worked for the jobs
that rely on a global library and just have a function-call in the pipeline
that's directly set in the job configuration.
However I have a couple of auxiliary jobs that are configured as 'Pipeline
script from git'
thanks for the comment.
Usually I am restarting without any jobs running. This time I am sure no
one ran.
ice...@googlemail.com schrieb am Freitag, 21. April 2023 um 10:38:31 UTC+2:
> From the backtrace and the description is seems a problem with the
> authorize-project plugin and restarts ( I
It didnÄt appear for a while so I hoped the problem was solved but after
the last update it happened again.
Even the scheduled builds don't run when this happens...
christop...@googlemail.com schrieb am Dienstag, 31. Januar 2023 um 10:56:25
UTC+1:
> This morning it happened again just after the
Hi,
TLDR: while running a job on a (windows) agent, is there a way to refresh
the environment from the agent to a shell- (actually bat-) step?
Whole story:
I want to test an application installer on a VM (I am resetting to a
snapshot after the test).
After the installation I want to make sure
I'd add something like a sleep to the loop
BR,
Christoph
ian.bo...@interceptgames.com schrieb am Donnerstag, 2. März 2023 um
01:27:09 UTC+1:
> For anyone else that stumbles across this question, I figured it out. I
> used the following code snippet to ensure the Jenkins pipeline waits unti
This morning it happened again just after the update of two plugins.
I found a lot of this in the log:
Caught exception evaluating: h.getJobPropertyDescriptors(it) in
/view/XXX/job/YYY/configure. Reason: java.lang.IllegalStateException: The
class jenkins.security.QueueItemAuthenticatorConfigura
Wouldn't a proper entry in the known-hosts file help here?
zil...@gmail.com schrieb am Montag, 30. Januar 2023 um 17:25:49 UTC+1:
> Bummer, thanks!
>
> On Monday, January 30, 2023 at 10:50:58 AM UTC-5 slide wrote:
>
>> I don't think the PO SSH plugin supports different methods of host key
>> ch
Ah, and when I'm talking about random effects after an upgrade combined
with a restart: I observed lost labels of lockable resources since a
similar period of time. This happened also right now after the last upgrade.
christop...@googlemail.com schrieb am Donnerstag, 19. Januar 2023 um
11:06:41
Hi,
since a couple of months I'm observing a weird behavior of my jenkins
installation after plugin upgrades.
I am running LTS (currently 2.375.2) on a windows server and updating very
close to each kind of update releases (jenkins itself as well as plugins).
After updates using plugin manager
Jochen,
don't use ${WORKSPACE} from outside the pipeline, it will be available as
soon as you are inside the pipeline with an agent.
That means: no need to have it as argument of your isccrPipeline
BR
jochen@gmail.com schrieb am Mittwoch, 18. Januar 2023 um 15:22:52 UTC+1:
>
> Okay, is
Hi Jochen,
one of my jobs generates a html-report as output from source code
documentation (doxygen). Instead of archiving I hand the html over to
publishHTML. Therefore displaying from the job works perfectly. I did not
care for downloading the html though so you might need to do both.
BR,
Ch
Hi Ullrich,
could you please help me with a question some posts older:
xunit seems to support a number of testDataPublishers where attachments()
is one of them. Do you know any documentation how this works? Is it the
same way it works for junit? I don't think so, otherwhise it should work
for m
Hi,
I am using xunit to collect googletest result files in my pipelines.
I recently started grouping the logs written by in internal logger and
results files of the tests to archive them and make them accessible to
debug the failed tests.
During this work I found the testDataPublishers attachem
>> agents, I guess.
>>>
>>> Thanks for the support!
>>> Christoph
>>> db...@cloudbees.com schrieb am Mittwoch, 23. November 2022 um 20:43:48
>>> UTC+1:
>>>
>>>> On Wed, Nov 23, 2022 at 2:26 PM 'christop...@googlemail.com&#
e of those two must be really depending on WMI windows
>> agents, I guess.
>>
>> Thanks for the support!
>> Christoph
>> db...@cloudbees.com schrieb am Mittwoch, 23. November 2022 um 20:43:48
>> UTC+1:
>>
>>> On Wed, Nov 23, 2022 at 2:26 PM &#x
peared and I still can
> start jobs. One of those two must be really depending on WMI windows
> agents, I guess.
>
> Thanks for the support!
> Christoph
> db...@cloudbees.com schrieb am Mittwoch, 23. November 2022 um 20:43:48
> UTC+1:
>
>> On Wed, Nov 23, 2022 at 2:26 P
Mittwoch, 23. November 2022 um 20:43:48
UTC+1:
> On Wed, Nov 23, 2022 at 2:26 PM 'christop...@googlemail.com' via Jenkins
> Users wrote:
>
>> And the fact that I have trouble when disabling WMI window agents tells
>> me there must be a real dependency somewhere, no
But - wait a second. As soon as I found out one or more culprits depending
on WMI windows agents these should immediately be declared deprecated,
right?
And you can declare the dependency.
Otherwise you'll run in trouble when there are still active installations
and WMI won't be shipped any lon
Hi Mark,
Hm, amongst these plugins are things like ViewVC (not working with CVS
currently but I'd try to fix it - was searching on CVS side), email ext
recipients column, doxygen, metrics disc usage.
And the fact that I have trouble when disabling WMI window agents tells me
there must be a real
Hi,
I have what we call a "hen-egg-problem" to solve a deprecation warning in
latest LTS:
Jenkins warns about deprecation of the WMI Windows agent plugin
When I want to disable that plugin for a test I get a warning, it might be
unsafe to disable the plugin because other plugins might depend on
Hi,
I got a similar issue with calling tests in debug and release variant. For
me results don't even show up twice, instead it's a matter of coincidence,
which results show up at all (individually for each single test case). In
my case it's about nunit and gtests which write their result in xml
Hi!
I am using a naked Windows vsphere agent for automatically testing an
application installer and some acceptance test of my software.
After every run I'd like to reset the machine to have a repeatable start
point for the tests.
I'm doing this like so:
lock('TestRunner') {
node('Installe
I don't get it:
when I execute my actual test in a try-block and move the xunit-step in the
finally-block everything works.
Is there any explanation for that behaviour?
christop...@googlemail.com schrieb am Donnerstag, 24. Februar 2022 um
16:25:31 UTC+1:
> It starts getting weird:
> I found an
It starts getting weird:
I found an xml-file in a totally different format as test output at the
same location. I whished, the parser would diplay the file name. I modified
the pattern to remove that from the parser input.
Now xunit runs witout any Error in the output but the following steps are
For the records:
I found a solution there:
https://stackoverflow.com/questions/53103996/where-does-the-jnlp-secret-come-from
But I regard the missing display of the proper startup command as a bug.
Thanks for listening!
christop...@googlemail.com schrieb am Mittwoch, 23. Februar 2022 um
08:54:5
Hi!
I am working with vsphere-cloud-plugin to dynamically set up vms as build
node from a template on demand. Therefore I made use of a template
.cmd-file provided by the plugin that loads slave.jar from the server and
reads the parameters from GuestInfo parameters.
I am also using the same scr
Oh, there's something else, I missed an output line to copy (this time with
debug- and release-tests in one run:
*12:24:17* INFO: Processing GoogleTest-1.8
*12:24:17* INFO: [GoogleTest-1.8] - 30 test report file(s) were found with
the pattern 'test*\*.gxml' relative to
'D:\workdir\workspace\My
This is what my pipeline looks like:
@Library('MyBuild') _
myBuild( module: MyModule', platform: 'win32vc19' )
Maybe its similar for the extended choice-plugin?
Thomas Markus schrieb am Montag, 21. Februar 2022 um 15:22:01 UTC+1:
> Hi,
>
> I'm using the extended choice parameter plugin for a se
Hi,
I have unit-test results in two different formats (junit as well as
googletest) and used to collect these using the xunit-plugin. Now I found
out accidentally that I parsed a junit file with the googletest format
which delivered a console output like "error parsing blablabla" however the
b
For me it does upgrade itself perfectly well (Server 2016, Zulu Java,
currently jenkins 2.319.2 running as service).
Only issue: I better manually restart the service after that because
sometimes my builds start getting awfully slow.
dheinric schrieb am Freitag, 4. Februar 2022 um 08:22:23 UTC+
ail.com schrieb am Mittwoch, 2. Februar 2022 um 13:58:46 UTC+1:
>
>
> Am 01.02.2022 um 22:27 schrieb 'christop...@googlemail.com' via Jenkins
> Users :
>
> [...]
>
>
> Any recommendations how I can improve that situation? Is there a way to
> realize some of
Hi,
I am building and testing a C/C++ application on Windows with jenkins using
declarative pipeline and a bit of scripts in a library.
I am using these tools:
cppcheck
googletest
robotframework
ranorex
I am using warnings-ng for the compiler warnings, taskscanner and cppcheck
and I am pretty ha
Hi,
Hm, no idea if this is related but today I suddenly had login issues to the
jenkins start page with a similar message. A restart of the jenkins service
helped.
Regards,
Christoph
ctr...@gmail.com schrieb am Dienstag, 5. Oktober 2021 um 16:21:18 UTC+2:
> Hi ,
>
> If any one has come across
In my opinion "call" is not the right statement to call an exe. Can you try
it with a start with the proper arguments?
Regards,
Christoph
slide schrieb am Donnerstag, 19. August 2021 um 17:28:44 UTC+2:
> Does the application try and interact with the desktop? If you are running
> the agent/jenk
Is there anybody out there still using CVS?
My IT from time to time changes passwords of the user used by jenkins to
access CVS and at the moment I have to modify a lot of jobs to make that
work.
Isn't there a way to combine the use of jenkins' own password storage with
CVS even though I can't
g, 27. Juli 2021 um 18:52:04
> UTC+2:
>
>> Can you find out what is going wrong? Add a call to GetLastError maybe?
>>
>> On Tue, Jul 27, 2021 at 2:22 AM 'christop...@googlemail.com' via Jenkins
>> Users wrote:
>>
>>>
>>> Hi eve
I still have a problem with access rights.
jonathan@gmail.com schrieb am Dienstag, 27. Juli 2021 um 18:52:04 UTC+2:
> Can you find out what is going wrong? Add a call to GetLastError maybe?
>
> On Tue, Jul 27, 2021 at 2:22 AM 'christop...@googlemail.com' via Jenkins
&g
Hi everybody,
I am building and testing windows applications on windows machines.
I have a number of tests which start several processes in the background.
After each of the tests I call a script to end the processes in case
something went wrong end they didn't end in a proper way during the te
Hi!
cppcheck has a command line argument to generate the proper xml-file, you
don't have to pipe anything to a file.
This works for me:
cppcheck.exe --xml --output-file=cppcheck-result-all.xml --enable=all .
I collect results like this:
recordIssues sourceCodeEncoding: 'ISO-8859-1', tools:
[c
Sorry for the late answer, I am only reading this after my vacation.
Just an idea: under windows the concept of unattended installation is used
when providing software updates and installation from a centralized server
in a domain environment without user interaction and a lot of common
installe
I don't know an answer to your question but I have other ideas.
Instead of running on the same slave you could find a way to access the
data created previously.
Either by accessing the same workspace or by transferring data from the
previous build by CopyArtifacts or stash/unstash or any other so
27; in the matrix-part by figuring out which
> nodes have this label set...
>
> Can I add a simple 1-, maybe 2-line script block in my pipeline to
> retrieve those values and use them?
>
>
> On 19.10.2020 9:45, 'christop...@googlemail.com' via Jenkins Users wrote:
> >
Hi,
I don't exactly understand your problem. Please have a look here:
https://www.jenkins.io/blog/2019/11/22/welcome-to-the-matrix/
you very easily can integrate matrix in a declarative pipeline. Insert your
nodes as an axis of the matrix and add an 'agent' in the stage doing your
job.
But may
Hello,
I am experiencing a strange difference with catching a timeout within a
script in a step or a script called from a step.
The problem I want to work around: in some situations my unit-test exe
(GoogleTest) is blocked with a popup from inside the C-runtime. To make the
build complete in
Maybe Framework != SDK
Have a look:
https://www.microsoft.com/de-de/download/details.aspx?id=6523
Regards,
Christoph
dana...@gmail.com schrieb am Donnerstag, 27. August 2020 um 16:21:56 UTC+2:
> Hi All,
>
> Here is a fuzzy bug (maybe) for ya.
>
> - Just installed a fresh Win 2016 Server.
> - Ins
n I click the link I see the browser asking
> for a forwarding for a short time, then I am back on the page of the build
> (http://%SERVER%:8080/job/%OTHER-JOBNAME%/ 305/ )
>
> Maybe this helped to solve the issue, thanks so far!
>
> Regards,
> Christoph
>
>
>
>
> u
age of the build
(http://%SERVER%:8080/job/%OTHER-JOBNAME%/ 305/ )
Maybe this helped to solve the issue, thanks so far!
Regards,
Christoph
ullrich...@gmail.com schrieb am Freitag, 21. August 2020 um 13:42:54 UTC+2:
>
> Am 21.08.2020 um 12:02 schrieb 'christop...@googlemail.com&#
I have two different kinds of jobs:
one is installing all the requirements for a job from a storage, building
one package and collecting warnings and cppcheck results with warnings ng.
Here everything works fine, display of warnings and cppcheck-analysis works.
In another jobs I am building all
Hi,
I am trying to call ranorex for execution of GUI tests.
To integrate Ranorex I followed this instructions:
https://www.ranorex.com/blog/integrating-ranorex-automation-in-jenkins-continuous-integration-process/
The things I am doing:
- build my own application with my own build environment ->
You mean postprocessing the result files to fake different class names to
workaround the laking separation of the testsuite by jenkins?
jeremy@riftio.com schrieb am Mittwoch, 5. August 2020 um 18:37:56 UTC+2:
> I believe this construct in the junit xml files can be safely modified to
> forc
Hi,
I found a strange phenomenon in test result display of jenkins. I had two
tests with very instable test results, good, bad, bad, good, good, good and
every time I opened the display, the sequence was different. But when
opening the tests I found out all of them either failed or passed.
Wha
Hi,
With emailext I know about two options: you can make use of a template and
add what you want from the template directly to the mail body.
Aditionally there's that 'attachLog'-option that attaches the complete log
to the mail.
Regards,
Christoph
fcenedese schrieb am Montag, 20. Juli 2020 um
54 matches
Mail list logo
