Jira plugin released

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

Jira plugin released

Kohsuke Kawaguchi-2

I just posted Hudson 1.86 and also the first release of the plugin for
integrating Hudson with Atlassian JIRA. See more about this plugin at my
blog
<http://weblogs.java.net/blog/kohsuke/archive/2007/03/hudsonjira_inte.html>

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]

smime.p7s (4K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

N Daley
Thanks Kohsuke!  I've deployed Hudson 1.86 and the Jira plugin on
http://lucene.zones.apache.org:8080/hudson

We'll see how the back pointer works on tonights nightly Hadoop build.

You may also notice the Hadoop-Patch build.  This is triggered automatically by hadoop-dev mailing list email that contains "Patch Available".  My scripts apply the patch before building and testing it.  Once the build is complete, my scripts use Jira's SOAP interface to update the relevant Jira bug with a "+1" or "-1".

Cheers,
Nige

----- Original Message ----
From: Kohsuke Kawaguchi <[hidden email]>
To: [hidden email]
Sent: Monday, March 12, 2007 5:39:29 PM
Subject: Jira plugin released


I just posted Hudson 1.86 and also the first release of the plugin for
integrating Hudson with Atlassian JIRA. See more about this plugin at my
blog
<http://weblogs.java.net/blog/kohsuke/archive/2007/03/hudsonjira_inte.html>

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]






 
____________________________________________________________________________________
Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
http://answers.yahoo.com/dir/?link=list&sid=396546091

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

Kohsuke Kawaguchi-2
N Daley wrote:

> Thanks Kohsuke!  I've deployed Hudson 1.86 and the Jira plugin on
> http://lucene.zones.apache.org:8080/hudson
>
> We'll see how the back pointer works on tonights nightly Hadoop build.
>
> You may also notice the Hadoop-Patch build.  This is triggered
> automatically by hadoop-dev mailing list email that contains "Patch
> Available".  My scripts apply the patch before building and testing it.  
> Once the build is complete, my scripts use Jira's SOAP interface to
> update the relevant Jira bug with a "+1" or "-1".
Cool. I was talking with one of my colleagues and we thought
implementing something like this would be very useful.

Basically, the idea is that you submit your change to Hudson (instead of
committing it), then have Hudson runs a build and test, and only when
everything went OK, hudson commits that to SCM on behalf of you.

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]

smime.p7s (4K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

N Daley
In reply to this post by Kohsuke Kawaguchi-2
Hmm, after upgrading to 1.86, all the failed Nightly builds at
http://lucene.zones.apache.org:8080/hudson/
fail with
java.lang.IndexOutOfBoundsExceptiontrying to run Subversion.  Any ideas?

Thx,
Nige

----- Original Message ----
From: Kohsuke Kawaguchi <[hidden email]>
To: [hidden email]
Sent: Monday, March 12, 2007 9:53:42 PM
Subject: Re: Jira plugin released

N Daley wrote:

> Thanks Kohsuke!  I've deployed Hudson 1.86 and the Jira plugin on
> http://lucene.zones.apache.org:8080/hudson
>
> We'll see how the back pointer works on tonights nightly Hadoop build.
>
> You may also notice the Hadoop-Patch build.  This is triggered
> automatically by hadoop-dev mailing list email that contains "Patch
> Available".  My scripts apply the patch before building and testing it.  
> Once the build is complete, my scripts use Jira's SOAP interface to
> update the relevant Jira bug with a "+1" or "-1".

Cool. I was talking with one of my colleagues and we thought
implementing something like this would be very useful.

Basically, the idea is that you submit your change to Hudson (instead of
committing it), then have Hudson runs a build and test, and only when
everything went OK, hudson commits that to SCM on behalf of you.

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]





 
____________________________________________________________________________________
No need to miss a message. Get email on-the-go
with Yahoo! Mail for Mobile. Get started.
http://mobile.yahoo.com/mail

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

Kohsuke Kawaguchi-2
N Daley wrote:
> Hmm, after upgrading to 1.86, all the failed Nightly builds at
> http://lucene.zones.apache.org:8080/hudson/
> fail with
> java.lang.IndexOutOfBoundsExceptiontrying to run Subversion.  Any ideas?

The error is happening inside svnkit, and I haven't touched subversion
part at all, so this is rather mysterious.

I do have some subversion jobs here, so I just run one of them for
sanity checking and that went fine.

I'll set up a test job to build the same repository as you do and see if
I can reproduce the problem. Sorry for the inconvenience.

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]

smime.p7s (4K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

Kohsuke Kawaguchi-2

I don't understand what's going on here. I need someone else to look
into this, so here it goes.

The top of the stack trace is:

> java.lang.IndexOutOfBoundsException
> at java.nio.Buffer.checkBounds(Buffer.java:454)
> at java.nio.HeapByteBuffer.get(HeapByteBuffer.java:125)
> at java.nio.ByteBuffer.get(ByteBuffer.java:674)
> at org.tmatesoft.svn.core.internal.delta.SVNDeltaReader.deflate(SVNDeltaReader.java:159)
> at org.tmatesoft.svn.core.internal.delta.SVNDeltaReader.nextWindow(SVNDeltaReader.java:124)

SVNDeltaReader.java:159 is the "myBuffer.get" line below. Notice that
it's just passing an array.

>         byte[] uncompressedData = new byte[uncompressedLength];
>         if (uncompressedLength == (compressedLength - (myBuffer.position() - originalPosition))) {
>             myBuffer.get(uncompressedData);
>             out.write(uncompressedData);

This is the ByteBuffer:674. It just delegates to the three arg version.
Simple enough.

>     public ByteBuffer get(byte[] dst) {
>         return get(dst, 0, dst.length);
>     }

And this is HeapByteBuffer.java:125. So before it does anything else it
does a bound check. Again, pretty normal stuff.

>     public ByteBuffer get(byte[] dst, int offset, int length) {
>         checkBounds(offset, length, dst.length);

And now this is Buffer.java:454.

>     static void checkBounds(int off, int len, int size) { // package-private
>         if ((off | len | (off + len) | (size - (off + len))) < 0)
>             throw new IndexOutOfBoundsException();
>     }

This is an interesting way of testing it, but basically "x<0" test is
the same as "is the most significant bit of x 1?", so "(x|y)<0" is the
same as "x<0 || y<0".

Now, if we remember, in this call stack, the offset is 0, the length is
dst.length, and the size is also dst.length.

So assuming that dst.length>=0 (which it must be, right?), I don't see
how this bound check can fail.

I mean, how could this be?


Kohsuke Kawaguchi wrote:

> N Daley wrote:
>> Hmm, after upgrading to 1.86, all the failed Nightly builds at
>> http://lucene.zones.apache.org:8080/hudson/
>> fail with
>> java.lang.IndexOutOfBoundsExceptiontrying to run Subversion.  Any ideas?
>
> The error is happening inside svnkit, and I haven't touched subversion
> part at all, so this is rather mysterious.
>
> I do have some subversion jobs here, so I just run one of them for
> sanity checking and that went fine.
>
> I'll set up a test job to build the same repository as you do and see if
> I can reproduce the problem. Sorry for the inconvenience.
>

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]

smime.p7s (4K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

N Daley
In reply to this post by Kohsuke Kawaguchi-2
I stopped Tomcat, blew away the session info, and restarted Tomcat.  Now it seems to be working fine.  Strange.

Nige

----- Original Message ----
From: Kohsuke Kawaguchi <[hidden email]>
To: [hidden email]
Sent: Tuesday, March 13, 2007 6:06:43 PM
Subject: Re: Jira plugin released


I don't understand what's going on here. I need someone else to look
into this, so here it goes.

The top of the stack trace is:

> java.lang.IndexOutOfBoundsException
>     at java.nio.Buffer.checkBounds(Buffer.java:454)
>     at java.nio.HeapByteBuffer.get(HeapByteBuffer.java:125)
>     at java.nio.ByteBuffer.get(ByteBuffer.java:674)
>     at org.tmatesoft.svn.core.internal.delta.SVNDeltaReader.deflate(SVNDeltaReader.java:159)
>     at org.tmatesoft.svn.core.internal.delta.SVNDeltaReader.nextWindow(SVNDeltaReader.java:124)

SVNDeltaReader.java:159 is the "myBuffer.get" line below. Notice that
it's just passing an array.

>         byte[] uncompressedData = new byte[uncompressedLength];
>         if (uncompressedLength == (compressedLength - (myBuffer.position() - originalPosition))) {
>             myBuffer.get(uncompressedData);
>             out.write(uncompressedData);

This is the ByteBuffer:674. It just delegates to the three arg version.
Simple enough.

>     public ByteBuffer get(byte[] dst) {
>         return get(dst, 0, dst.length);
>     }

And this is HeapByteBuffer.java:125. So before it does anything else it
does a bound check. Again, pretty normal stuff.

>     public ByteBuffer get(byte[] dst, int offset, int length) {
>         checkBounds(offset, length, dst.length);

And now this is Buffer.java:454.

>     static void checkBounds(int off, int len, int size) { // package-private
>         if ((off | len | (off + len) | (size - (off + len))) < 0)
>             throw new IndexOutOfBoundsException();
>     }

This is an interesting way of testing it, but basically "x<0" test is
the same as "is the most significant bit of x 1?", so "(x|y)<0" is the
same as "x<0 || y<0".

Now, if we remember, in this call stack, the offset is 0, the length is
dst.length, and the size is also dst.length.

So assuming that dst.length>=0 (which it must be, right?), I don't see
how this bound check can fail.

I mean, how could this be?


Kohsuke Kawaguchi wrote:

> N Daley wrote:
>> Hmm, after upgrading to 1.86, all the failed Nightly builds at
>> http://lucene.zones.apache.org:8080/hudson/
>> fail with
>> java.lang.IndexOutOfBoundsExceptiontrying to run Subversion.  Any ideas?
>
> The error is happening inside svnkit, and I haven't touched subversion
> part at all, so this is rather mysterious.
>
> I do have some subversion jobs here, so I just run one of them for
> sanity checking and that went fine.
>
> I'll set up a test job to build the same repository as you do and see if
> I can reproduce the problem. Sorry for the inconvenience.
>


--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]





 
____________________________________________________________________________________
Now that's room service!  Choose from over 150,000 hotels
in 45,000 destinations on Yahoo! Travel to find your fit.
http://farechase.yahoo.com/promo-generic-14795097

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Jira plugin released

Kohsuke Kawaguchi-2
N Daley wrote:
> I stopped Tomcat, blew away the session info, and restarted Tomcat.  Now
> it seems to be working fine.  Strange.

Maybe it's one of those bit errors on memory, JVM bug, or what.

But as long as it's working, I'm fine.

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]

smime.p7s (4K) Download Attachment