If you use the GUI, the scan interval for new branches is under Scan
Multibranch Pipeline Triggers and the orphan item setting is right after it.
On Thursday, April 14, 2022 at 5:02:50 PM UTC-4 nhoj.p...@gmail.com wrote:
> So I've changed mine to 7 days, which I've done using the groovy script
println("ignoring release change")
} else {
print("ignoring supermod change: ${pth}")
}
}
}
}
}
print("no interesting commits found.")
return false
}
Jeremy Mordkoff
Director, Engine
We use git and submodules. I only want certain builds to kick off if there
are commits in certain submodules. I have this as part of the pipeline now,
but this means my build history is filling up with jobs that did nothing
except figure out that they didn't need to run and that's pushing the
what are the perms on /export/build and /export/build/remoting.jar? One of
them is not writable by user jenkins
On Thursday, December 23, 2021 at 3:28:15 AM UTC-5 Harri wrote:
> Hi folks,
>
> using Jenkins 2.326 (the Debian package) the agent cannot be redeployed. It
> fails with
>
>
I'm assuming then that there's no network activity at the end of the 30 minute
window. What is the master doing? Perhaps cleaning up old builds? How many do
you keep? FYI...I normally keep about 100 per pipeline & branch.
Jeremy Mordkoff
Director, Engineering Services
[cid:d4bea858-87ac-
30 minutes sounds like a TCP timeout. Are there any firewalls or NAT
devices in the path? Sometimes they close the connection when the first FIN
is sent but before the last FIN-ACK and that can cause SSH to hang up.
A second possibility is asymetric routes where the replies come in on a
We are migrating from a single build pipeline to a pair of pipelines. The
first will build our "platform" and then the second will build the
application. This is because we will soon have multiple applications using
the same platform.
The applications should build if the platform changes OR
Also interested in the answer.
At the moment we scan the change log from the scm object and abort the
pipeline (cleanly) if there is nothing of interest to that pipeline to
build. I do not like this solution for multiple reasons.
Our plan is to split our code into 3 repos .. one for each app
What permission(s) do I need to grant to a new user in order for them to
see the pipeline syntax option?
I am a full admin and I see it.
They are authenticated but they only have rights to start and stop jobs
right now. I created a job for them that invokes a JenkinsFile and they are
busy
gt;
> Le mar. 1 déc. 2020 à 22:11, jeremy mordkoff a
> écrit :
>
>> We have too many tests, so we have broken them up into three groups. One
>> group that runs after each checkin and one group that runs once a day on
>> any branch with changes and one group that I wo
We have too many tests, so we have broken them up into three groups. One
group that runs after each checkin and one group that runs once a day on
any branch with changes and one group that I would like to limit to one
instance at a time but still test any active branch.
What is the best way
simple answer is no. an href cannot include a headers directive or anything
else)
It can be done using ajax
--
https://stackoverflow.com/questions/374885/can-i-change-the-headers-of-the-http-request-sent-by-the-browser
or you could build a proxy that did nothing but add that header. But if
Hi Joe,
Missing an option...I use the UI a lot but not exclusively.
On Thursday, October 29, 2020 at 10:11:55 AM UTC-4 Joseph Brueggen wrote:
> Hello everyone,
>
> My name is Joe—I'm a designer at CloudBees. I'm gathering some feedback
> about the Jenkins user interface.
>
> This new survey
okay, take this with a grain of salt because this is purely a guess
The ssh call is trying to serialize all of the local variables, but it
can't serialize the simpleTemplate. So move all the code that touches
simpletemplate into a function that's wrapped with the @NONCPS decorator.
something
I think it is failing on this line:
echo("MAL is ${MAL}")
On Monday, October 26, 2020 at 11:51:19 AM UTC-4 cw wrote:
> Also I'm confused why I'm getting a NotSerializableException related to
> SimpleTemplateEngine$SimpleTemplate when I'm in a try/except block which is
> running an
Are you running this in the main thread of the enkinsFile itself or in a
function? I suspect that you are in a function and you may need to use the
@NONCPS decorator
On Friday, October 23, 2020 at 4:31:55 PM UTC-4 cw wrote:
> That's great, thank you very much. I was able to take your code,
I stumbled
on
https://stackoverflow.com/questions/55423036/groovy-string-interpolation-when-string-is-defined-before-the-interpolated-varia
and I found this snippet worked just fine after I approved the signature in
jenkins
import groovy.text.SimpleTemplateEngine
node("docker") {
I think you need to do an "eval" or the equivalent to filepath in order to
expand the inner variable. I believe the ssh command is executing "ls
/srv/jboss/server/${MAL}/log/" but MAL on the far end is empty. So you need
to complete the expansion before calling ssh.
This might get you on the
We have a similar issue that only seems to occur during long running jobs
(over 5 hours). The traceback is different but we also see the EOF
exception. My client is Ubuntu linux
I tried to trace the issue by running tcpdumps at both ends on the ssh
session from the master to the slave but I
node {
stage("checkout") {
checkout scm
result = sh (script: "git log -1 | grep bump", returnStatus: true)
if (result == 0) {
println("CI bump commit...skipping build")
currentBuild.result = 'NOT_BUILT'
# can I just return here ? exit 0?
}
}
stage("build") {
I use scripted pipelines
I would like certain builds to abort early. (1) if it is just a branch
creation event, (2) if the commit message is "bump" and (3) if the only
change was in certain submodules.
I think I know all the pieces to make this work -- I can write a tool that
could determine
You could leave the Automatic triggering suppressed and write a tool to
scan your GIT repo looking for new branches and trigger the build via an
API call when a new branch is found. Sounds like you might need such a tool
anyway so that developers could schedule builds.
On Thursday, September
if you don't wait, then propagate is meaningless .. it can't be done.
I find scripted is more flexible, but any time a system gives you more rope,
you have to be careful not to hang yourself.
---
Jeremy Mordkoff
Director of Engineering Services
[cid:b0edc09c-0e90-4f2f-a1f2-715660acd801]<ht
sounds like the user is valid but not a member of any groups. Is that
possible?
On Thursday, August 20, 2020 at 1:28:21 PM UTC-4 Mk wrote:
> Hi Team,
>
> I am trying to configure LDAP(AD) Authentication in our Jenkins, Below is
> my configuration settings, But test LDAP connection is
But if JobB fails, it marks JobA also as failed even though JobA ran
> successfully. I want JobA to trigger JobB on Post success condition and
> then not fail JobA, if JobB fails, preferably without try/catch blocks.
>
> Regards,
> Venkatesh
>
>
> On Fri, Aug 21, 2020 at 9:31
I use scripted pipelines with a library. The master and the slaves are
running in ubuntu 16.04 docker containers.
About a week ago we started getting occasional failures from this call:
junit testResults: ".build/ub18_debug/artifacts/systemtest/*xml"
There are about 60 junit xml files in
Are you using scripted or declarative? I will assume scripted since that's
all I know :)
There's a don't wait option. Or you could wrap the call in a try/catch. Or
you could save currentBuild.result before calling job B and restore it after
On Friday, August 21, 2020 at 5:09:57 AM UTC-4
. But the expansion in a script. Like I said,
the build doesn't have to use the workspace created by jenkins.
On Wednesday, August 19, 2020 at 6:47:23 PM UTC-4 Anton Shepelev wrote:
> jeremy mordkoff:
>
> > I'm having a little trouble understanding exactly what you
> > are aski
I'm having a little trouble understanding exactly what you are asking. I
think you are asking if it is possible to trigger multiple builds in a
single workspace. If you use a single pipeline, this is trivial.
So perhaps you should consider writing a single pipeline that iterates over
all
How do you maintain and verify backwards compatibility with older releases
if you keep your devops code in a separate repo? I keep my devops code in
the same repo as the product code so that I know I can always go back and
rebuild an older release and get exactly the same results.
The only
I try to make the calls in my top level jenkinsfile atomic and complete,
i.e. each one performs a single function. By using long, descriptive names
I can avoid the need for lots of comments. It also makes building new
pipelines easy and encourages reuse across files, stages and steps. If I
see
yes, exactly. I post process mine to remove the hostname from the class
because my tooling generates the class based on the path.
On Thursday, August 6, 2020 at 9:27:32 AM UTC-4, christop...@googlemail.com
wrote:
>
> You mean postprocessing the result files to fake different class names to
>
I believe this construct in the junit xml files can be safely modified to
force jenkins to see the test results as two separate tests
> 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
4321.key
> crt filename is test2-harbor54321.crt
> core: test2-harbor.tank.local
> notary: test2-harbor.tank.local
> secretName: test2-harbor54321
> values.yaml is updated
> secret/test2-harbor54321 created
>
> NAME: harbor
> LAST DEPLOYED: Tue Aug 4 14:58:39 2020
> NAMESPACE:
nkins console i am still getting cant connect to the server :
>>
>> [image: image.png]
>>
>>
>> [image: image.png]
>>
>> and still from my local system to host i cant ssh directly:
>>
>> sakshi_rathore@MCN234 MINGW64 ~/.ssh (master)
>> $ ss
@Haibinh is that a bash answer?
@chencho -- are you looking for a solution in bash, or groovy? Declarative
or scripted?
On Friday, July 31, 2020 at 9:16:52 AM UTC-4, Haibinh Nguyen wrote:
>
> Hi,
> How about this:
>
> env.FIELD1 = params.FIELD1
> switch(FIELD1) {
> case "ABCD":
>
rver
>>> i also tried simple ssh to the server in build step but it is not
>>> working. Please advise where I am wrong.
>>>
>>> On Wed, Jul 29, 2020 at 8:35 AM Sakshi Rathore
>>> wrote:
>>>
>> I tried a lot of things but my ssh connection is n
Normally a SSH keypair is created on the client (in your case the jenkins
server) and then the public key is copied to the server. This is more
secure because the private key is created on the client and is never copied
anywhere else.
This level of security is rarely needed or enforced in a
.
>>
>> One more thing my Jenkins and ssh server is hosting on same IP address is
>> there anyway i can resolve this issue?
>>
>> On Mon, 27 Jul, 2020, 9:39 AM Sakshi Rathore, > > wrote:
>>
>>> thanks , i tried this as well but while getting connectio
I think the issue is that the client does not trust the server's host key.
These keys are stored by the client in ~/.ssh/known_hosts Try
copying/appending your .ssh/known_hosts file to jenkins'
On Monday, July 20, 2020 at 5:43:39 PM UTC-4, Sakshi Rathore wrote:
>
> 0I have a bash script
How do you normally handle branches and merging? I use the same tools and
branching and merging rules as the rest of the team, i.e. development and
QA.
In our case, the teams are required to merge from master to their project
branches on a regular basis, so they will eventually get my changes
ouch. I always keep one window open ssh'd in as root when I play with
sshd_config. Hopefully you have console access and a username/password with
sudo rights.
On Monday, July 20, 2020 at 5:43:39 PM UTC-4, Sakshi Rathore wrote:
>
> 0I have a bash script which connects and exexutes programs from
CI and passphrases never mix :) You're going to need some password-less
keys even if they have limited access.
On Monday, July 20, 2020 at 4:48:23 PM UTC-4, Randall Becker wrote:
>
> Turns out, you cannot use a key-pair with a passphrase in this situation.
> SSH key-pairs without a passphrase
Where is that file referenced? In your .ssh/config file? It sounds like a
simple path issue -- the config file is referencing a location that is not
available to jenkins. You can override this on the command line, e.g.
something like -i /c/Users/jenkins/ssh/id_rsa
On Monday, July 20, 2020 at
You are positive the key works and that jenkns is ssh'ing as the right
user? If so, can you check that the key is the only one offered? Finally,
enable debug logs on the target's SSH server and look for the cause.
On Tuesday, July 21, 2020 at 3:03:10 AM UTC-4, sakshira...@gmail.com wrote:
>
>
Hi,
I am heading down this path. My plan is build a web app that can accept
test results in various formats, and can aggregate them by build and
identify them by type.
It will track sources and link back to them
It will have the ability to normalize the xml paths, e.g.
We use git and SCM polling to trigger builds. There are two situations
where I do not want a commit to cause a build. One is when we bump the
release number and the other is when I initially create a branch.
Does anyone have a groovy snippet that returns all of the commit messages
since the
sending email always involves a forward lookup and many mail servers do a
reverse lookup on the sender's IP.
---
Jeremy Mordkoff
Director of Engineering Services
[cid:7231f95f-ee20-460d-bb39-189076c048ad]<https://www.riftio.com>
RIFT, inc
900 Chelmsford Street, 4th floor, tower 3
Lowe
check that DNS is working. I've seen huge delays on basic ops when DNS
lookups are timing out. Check forward and reverse lookups.
On Sunday, April 26, 2020 at 12:30:59 PM UTC-4, slide wrote:
>
> Can you add timestamps? It's hard to tell from what you posted when things
> are occurring.
>
>
I did stumble on the setting showing up in the UI when I 'view
configuration' for the job. But I would prefer to see it in a column if
possible.
On Thursday, April 23, 2020 at 10:17:26 AM UTC-4, jeremy mordkoff wrote:
>
> If I use this syntax
>
> properties([
> pipelineTr
If I use this syntax
properties([
pipelineTriggers([cron('0 21 * * *')])
])
Then the cron spec shows up under "periodic build trigger' in my list
views. But it does not show up if I use this syntax
properties([
pipelineTriggers([
[$class: "SCMTrigger", scmpoll_spec: 'H 22 *
Check the mail queue on the jenkins master node (usually "sudo mailq"). Or
try "echo test | mail -s test "jafargmail.com" on the master node.
gmail is very picky about whom they will accept mail from. Where are you
sending from? Does it have SPF and/or DKIM ?
On Friday, April 17, 2020 at
I have a similar issue, except my test results some from multiple jobs
my basic build pipeline runs unit tests then invokes the systemtest
pipeline. These run 10 to 30 times a day.
I also have a nightly pipeline that runs a lot of additional tests, like
coverage, interop, stress, load, upgrade.
I was thinking about this the other day. My plan is to create a simple
jenkins job that manages all of my other jenkins jobs. Some of the other
suggestions might feed into this.
My goal is to able to bootstrap a new jenkins server in minutes for
disaster recovery and for bringing up new
I went back and forth on this. In the end, I put everything in code. I
decided I wanted the minimum amount of stuff configured in jenkins.
I started out with a wrapper pipeline that loaded the "guts" of the job.
This was extremely restrictive in that each job had to conform at some
level to
I use a templating tool (m4) so I can build up a groovy script using
macros. In the first stage of my Jenklinsfiie, I call a bash script that
generates a groovy script, which I then load and execute. The loaded script
has N stages. Note that if the ordering of your stages changes, jenkins
will
yes. You can generate groovy code using any tool you want and then use load
to import it, e.g.
node('docker') {
checkout scm
// this step generates Jenkinsfile.do_merge
sh script: './bin/git/needs_merging -j'
pipeline = load 'Jenkinsfile.do_merge'
pipeline.execute()
}
and
My personal advice is to pick declarative or scripted pipelines and stick
to that everywhere. Switching back and forth always leads to mistakes. And
so keeping the Jenkinsfiles in source control is huge plus and I would
never go back to putting any logic inside my jenkins job config. Why are
I had noticed this and I had assumed that it was writing out the layers to
disk. I will also check for background tasks per Rob's comment
On Friday, June 21, 2019 at 5:01:17 PM UTC-4, Manny DaSilva wrote:
>
> Actually, it looks like the "exit 0" has nothing to do with it, but the
> long pause
I have the same question. The conversion of pathnames in the mailed out
logs makes them unreadable on smartphones. I would like to mail out the log
as is, with no conversion.
Does anyone know where (approximately) the code that makes this change
would live? I don't mind a little hacking
JLM
60 matches
Mail list logo
