Quantcast

Writing a Jenkins test case including different JDKs?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Writing a Jenkins test case including different JDKs?

Christoph Kutzinski
Hi,

while fixing https://issues.jenkins-ci.org/browse/JENKINS-10030 (the fix
itself is quite trivial),
I wondered if it's possible to write a testcase based on HudsonTestCase
to check for regression in that area.
I.e. firing off a Maven build in a different VM than the one running the
testcase.
I thought I could maybe use JDKInstaller for that, but it seems only to
support Oracle VMs while I would need a IBM Jdk to test.

Any ideas?

thanks
Christoph
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

Jerome Lacoste-3
what about making sure all Serializable classes in Jenkins have a serialVersionUID instead ?
jerome
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

Christoph Kutzinski
Yes that would be fixing it right now. But I wanted to have a regression test, so that it will never happen again.
So you're saying that we should maybe enable a FindBugs check for this which will fail the build if a Serializable class without a serialVersionUID is found?

Christoph

Am 15.08.2011 18:32, schrieb Jerome Lacoste:
what about making sure all Serializable classes in Jenkins have a serialVersionUID instead ?
jerome

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

Jerome Lacoste-3
> So you're saying that we should maybe enable a FindBugs check for this which will fail the build if a Serializable class without a serialVersionUID is found?

Something like that yes. Not sure if using findbugs is the best solution though. Will you get enough flexibility if you need to tune it further.

J
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

kohsuke Kawaguchi (CB)

+1 for FindBugs. I think we can have a Maven profile in our POM that
activates some checks, and then we can run them on J-on-J and as a part
of the release process. Requiring serialVersionUID for all serializable
classes seem like a reasonable check.

FindBugs would also likely to have other useful checks, too.

On 08/18/2011 11:44 PM, Jerome Lacoste wrote:
>>  So you're saying that we should maybe enable a FindBugs check for this
> which will fail the build if a Serializable class without a
> serialVersionUID is found?
>
> Something like that yes. Not sure if using findbugs is the best solution
> though. Will you get enough flexibility if you need to tune it further.
> http://findbugs.sourceforge.net/bugDescriptions.html#SE_NO_SERIALVERSIONID
>
> J


--
Kohsuke Kawaguchi | CloudBees, Inc. | http://cloudbees.com/
Try Nectar, our professional version of Jenkins
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

jyoung
Would having a Sonar server be overkill?  Sonar provides a nice combination
of warnings and analysis from various packages such as Findbugs, PMD, and
Checkstyle, plus makes it pretty easy to go through and tweak the various
rules to be on/ off or various levels of severity.

Plus it wouldn't hurt my feelings if Sonar and Jenkins were closer
friends....


Jeremy D. Young
Web Architect
O'Reilly Auto Parts

[hidden email] wrote on 08/23/2011 05:26:24 PM:

> From: Kohsuke Kawaguchi <[hidden email]>
> To: [hidden email]
> Cc: Jerome Lacoste <[hidden email]>
> Date: 08/23/2011 05:26 PM
> Subject: Re: Writing a Jenkins test case including different JDKs?
> Sent by: [hidden email]
>
>
> +1 for FindBugs. I think we can have a Maven profile in our POM that
> activates some checks, and then we can run them on J-on-J and as a part
> of the release process. Requiring serialVersionUID for all serializable
> classes seem like a reasonable check.
>
> FindBugs would also likely to have other useful checks, too.
>
> On 08/18/2011 11:44 PM, Jerome Lacoste wrote:
> >>  So you're saying that we should maybe enable a FindBugs check for
this
> > which will fail the build if a Serializable class without a
> > serialVersionUID is found?
> >
> > Something like that yes. Not sure if using findbugs is the best
solution
> > though. Will you get enough flexibility if you need to tune it further.
> >
http://findbugs.sourceforge.net/bugDescriptions.html#SE_NO_SERIALVERSIONID

> >
> > J
>
>
> --
> Kohsuke Kawaguchi | CloudBees, Inc. | http://cloudbees.com/
> Try Nectar, our professional version of Jenkins
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>

This communication and any attachments are confidential, protected by Communications Privacy Act 18 USCS § 2510, solely for the use of the intended recipient, and may contain legally privileged material. If you are not the intended recipient, please return or destroy it immediately. Thank you.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

domi@fortysix.ch
+1

On 24.08.2011, at 17:30, [hidden email] wrote:

> Would having a Sonar server be overkill?  Sonar provides a nice combination
> of warnings and analysis from various packages such as Findbugs, PMD, and
> Checkstyle, plus makes it pretty easy to go through and tweak the various
> rules to be on/ off or various levels of severity.
>
> Plus it wouldn't hurt my feelings if Sonar and Jenkins were closer
> friends....
>
>
> Jeremy D. Young
> Web Architect
> O'Reilly Auto Parts
>
> [hidden email] wrote on 08/23/2011 05:26:24 PM:
>
>> From: Kohsuke Kawaguchi <[hidden email]>
>> To: [hidden email]
>> Cc: Jerome Lacoste <[hidden email]>
>> Date: 08/23/2011 05:26 PM
>> Subject: Re: Writing a Jenkins test case including different JDKs?
>> Sent by: [hidden email]
>>
>>
>> +1 for FindBugs. I think we can have a Maven profile in our POM that
>> activates some checks, and then we can run them on J-on-J and as a part
>> of the release process. Requiring serialVersionUID for all serializable
>> classes seem like a reasonable check.
>>
>> FindBugs would also likely to have other useful checks, too.
>>
>> On 08/18/2011 11:44 PM, Jerome Lacoste wrote:
>>>> So you're saying that we should maybe enable a FindBugs check for
> this
>>> which will fail the build if a Serializable class without a
>>> serialVersionUID is found?
>>>
>>> Something like that yes. Not sure if using findbugs is the best
> solution
>>> though. Will you get enough flexibility if you need to tune it further.
>>>
> http://findbugs.sourceforge.net/bugDescriptions.html#SE_NO_SERIALVERSIONID
>>>
>>> J
>>
>>
>> --
>> Kohsuke Kawaguchi | CloudBees, Inc. | http://cloudbees.com/
>> Try Nectar, our professional version of Jenkins
>>
>> --
>> This message has been scanned for viruses and
>> dangerous content by MailScanner, and is
>> believed to be clean.
>>
>
> This communication and any attachments are confidential, protected by Communications Privacy Act 18 USCS § 2510, solely for the use of the intended recipient, and may contain legally privileged material. If you are not the intended recipient, please return or destroy it immediately. Thank you.

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Writing a Jenkins test case including different JDKs?

Richard Bywater
In reply to this post by jyoung
+1

On Thu, Aug 25, 2011 at 3:30 AM, <[hidden email]> wrote:
Would having a Sonar server be overkill?  Sonar provides a nice combination
of warnings and analysis from various packages such as Findbugs, PMD, and
Checkstyle, plus makes it pretty easy to go through and tweak the various
rules to be on/ off or various levels of severity.

Plus it wouldn't hurt my feelings if Sonar and Jenkins were closer
friends....


Jeremy D. Young
Web Architect
O'Reilly Auto Parts

[hidden email] wrote on 08/23/2011 05:26:24 PM:

> From: Kohsuke Kawaguchi <[hidden email]>
> To: [hidden email]
> Cc: Jerome Lacoste <[hidden email]>
> Date: 08/23/2011 05:26 PM
> Subject: Re: Writing a Jenkins test case including different JDKs?
> Sent by: [hidden email]
>
>
> +1 for FindBugs. I think we can have a Maven profile in our POM that
> activates some checks, and then we can run them on J-on-J and as a part
> of the release process. Requiring serialVersionUID for all serializable
> classes seem like a reasonable check.
>
> FindBugs would also likely to have other useful checks, too.
>
> On 08/18/2011 11:44 PM, Jerome Lacoste wrote:
> >>  So you're saying that we should maybe enable a FindBugs check for
this
> > which will fail the build if a Serializable class without a
> > serialVersionUID is found?
> >
> > Something like that yes. Not sure if using findbugs is the best
solution
> > though. Will you get enough flexibility if you need to tune it further.
> >
http://findbugs.sourceforge.net/bugDescriptions.html#SE_NO_SERIALVERSIONID
> >
> > J
>
>
> --
> Kohsuke Kawaguchi | CloudBees, Inc. | http://cloudbees.com/
> Try Nectar, our professional version of Jenkins
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>

This communication and any attachments are confidential, protected by Communications Privacy Act 18 USCS § 2510, solely for the use of the intended recipient, and may contain legally privileged material. If you are not the intended recipient, please return or destroy it immediately. Thank you.

Loading...