You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@geronimo.apache.org by Matt Hogstrom <ma...@hogstrom.org> on 2005/11/14 22:04:55 UTC
ApahceCon is T-26 days (or there abouts) thoughts about branching?
We're getting fairly close to the time where Turkey's will dominate the week and
then soon thereafter folks will converge on San Diego California for ApacheCon.
Here is the working list of content and the status as best I can tell:
- More console work
Look and feel upgrades added. Aaaron and Joe are continuing this effort.
- Remote deployment
Aaron has been working on this
- Hot deploy directory
I think this is done (tested ?)
- Bundled JavaMail with SMTP support
Hmmm...Geir was working this one I think
- Separate Jetty and Tomcat ZIP/TAR distributions
Jencks the maestro was working this one
- A working installer distribution
Jencks and Erik were looking at this
- Bug fixes
Not sure of the exact list. I've been crawling through JIRAs and attacking the
low hanging fruit
- Unplanned but cool additions
Jeff has been working on the Tomcat Clustering Gbeans and David was doing some
initial testing.
Not sure about the timing. Looks like we'll make ApacheCon based on current
course and speed. What's the right time to branch so we stabilize 1.0. I
expect there are some other items continuing in development. My gut says we
should branch no later than sometime next week and start TCK testing with the
GBuild environment Blevins has been working on.
Thoughts?
Matt
Re: ApahceCon is T-26 days (or there abouts) thoughts about branching?
Posted by Sachin Patel <sp...@gmail.com>.
Yep, thats what I got.
sachin
Aaron Mulder wrote:
> Yup, good catch. I think the default configId is the module name
> without its file extension -- can you confirm that's what you got (by
> running deployer.jar list-modules, for example)?
>
> Thanks,
> Aaron
>
> On 11/14/05, Sachin Patel <sp...@gmail.com> wrote:
>
>> Aaron Mulder wrote:
>>
>>> Anyway, on your list, hot deploy directory is done, but it would be
>>> nice if people could test it and make sure there aren't any issues.
>>> And I haven't actually been working on remote deployment yet, though I
>>> still mean to. :)
>>>
>>>
>> Just did a quick smoke test of the hot deploy. Looks good. However
>> when I ran a scenario where if I dropped in a war without
>> an embedded plan, the deploy was successful, but when deleting the file
>> to trigger an undeploy the following
>> exception occurred....
>>
>> 17:40:06,506 ERROR [DirectoryHotDeployer] Unable to undeploy
>> java.lang.NullPointerException
>> at
>> org.apache.geronimo.deployment.cli.DeployUtils.identifyTargetModuleIDs(DeployUtils.java:113)
>> at
>> org.apache.geronimo.deployment.hot.DirectoryHotDeployer.fileUpdated(DirectoryHotDeployer.java:215)
>> at
>> org.apache.geronimo.deployment.hot.DirectoryMonitor.scanDirectory(DirectoryMonitor.java:200)
>> at
>> org.apache.geronimo.deployment.hot.DirectoryMonitor.run(DirectoryMonitor.java:140)
>> at java.lang.Thread.run(Thread.java:552)
>>
>> Can we fall back on the module name as the configID if there is no plan,
>> or whatever default is being used when no plan is provided.
>>
>> Sachin
>>
>>
>
>
Re: ApahceCon is T-26 days (or there abouts) thoughts about branching?
Posted by Aaron Mulder <am...@alumni.princeton.edu>.
Yup, good catch. I think the default configId is the module name
without its file extension -- can you confirm that's what you got (by
running deployer.jar list-modules, for example)?
Thanks,
Aaron
On 11/14/05, Sachin Patel <sp...@gmail.com> wrote:
> Aaron Mulder wrote:
> >
> > Anyway, on your list, hot deploy directory is done, but it would be
> > nice if people could test it and make sure there aren't any issues.
> > And I haven't actually been working on remote deployment yet, though I
> > still mean to. :)
> >
> Just did a quick smoke test of the hot deploy. Looks good. However
> when I ran a scenario where if I dropped in a war without
> an embedded plan, the deploy was successful, but when deleting the file
> to trigger an undeploy the following
> exception occurred....
>
> 17:40:06,506 ERROR [DirectoryHotDeployer] Unable to undeploy
> java.lang.NullPointerException
> at
> org.apache.geronimo.deployment.cli.DeployUtils.identifyTargetModuleIDs(DeployUtils.java:113)
> at
> org.apache.geronimo.deployment.hot.DirectoryHotDeployer.fileUpdated(DirectoryHotDeployer.java:215)
> at
> org.apache.geronimo.deployment.hot.DirectoryMonitor.scanDirectory(DirectoryMonitor.java:200)
> at
> org.apache.geronimo.deployment.hot.DirectoryMonitor.run(DirectoryMonitor.java:140)
> at java.lang.Thread.run(Thread.java:552)
>
> Can we fall back on the module name as the configID if there is no plan,
> or whatever default is being used when no plan is provided.
>
> Sachin
>
Re: ApahceCon is T-26 days (or there abouts) thoughts about branching?
Posted by Sachin Patel <sp...@gmail.com>.
Aaron Mulder wrote:
>
> Anyway, on your list, hot deploy directory is done, but it would be
> nice if people could test it and make sure there aren't any issues.
> And I haven't actually been working on remote deployment yet, though I
> still mean to. :)
>
Just did a quick smoke test of the hot deploy. Looks good. However
when I ran a scenario where if I dropped in a war without
an embedded plan, the deploy was successful, but when deleting the file
to trigger an undeploy the following
exception occurred....
17:40:06,506 ERROR [DirectoryHotDeployer] Unable to undeploy
java.lang.NullPointerException
at
org.apache.geronimo.deployment.cli.DeployUtils.identifyTargetModuleIDs(DeployUtils.java:113)
at
org.apache.geronimo.deployment.hot.DirectoryHotDeployer.fileUpdated(DirectoryHotDeployer.java:215)
at
org.apache.geronimo.deployment.hot.DirectoryMonitor.scanDirectory(DirectoryMonitor.java:200)
at
org.apache.geronimo.deployment.hot.DirectoryMonitor.run(DirectoryMonitor.java:140)
at java.lang.Thread.run(Thread.java:552)
Can we fall back on the module name as the configID if there is no plan,
or whatever default is being used when no plan is provided.
Sachin
Re: ApahceCon is T-26 days (or there abouts) thoughts about branching?
Posted by Aaron Mulder <am...@alumni.princeton.edu>.
I wouldn't mind TCK testing what we've got as soon as David J has the
Tomcat/Jetty distros ready and calling it 1.0-RC1 if we can finish the
TCK in the next week or so.
On a side note, it still feels like there are some significant bugs
out there, though I haven't reviewed the list. Can you send a list of
what looks like outstanding bugs for 1.0?
Could we get Alan to bless the state of security for 1.0? I'm not
thinking of anything in particular, I just want to make sure he
doesn't have anything left to do.
Also, it would be nice if the client container could run remotely. I
think David J said there wasn't a ton of work for this but it wasn't a
real high priority either. Maybe if someone's sitting around bored
they could look at it?
Anyway, on your list, hot deploy directory is done, but it would be
nice if people could test it and make sure there aren't any issues.
And I haven't actually been working on remote deployment yet, though I
still mean to. :)
Thanks,
Aaron
On 11/14/05, Matt Hogstrom <ma...@hogstrom.org> wrote:
> We're getting fairly close to the time where Turkey's will dominate the week and
> then soon thereafter folks will converge on San Diego California for ApacheCon.
>
> Here is the working list of content and the status as best I can tell:
>
> - More console work
> Look and feel upgrades added. Aaaron and Joe are continuing this effort.
> - Remote deployment
> Aaron has been working on this
> - Hot deploy directory
> I think this is done (tested ?)
> - Bundled JavaMail with SMTP support
> Hmmm...Geir was working this one I think
> - Separate Jetty and Tomcat ZIP/TAR distributions
> Jencks the maestro was working this one
> - A working installer distribution
> Jencks and Erik were looking at this
> - Bug fixes
> Not sure of the exact list. I've been crawling through JIRAs and attacking the
> low hanging fruit
> - Unplanned but cool additions
> Jeff has been working on the Tomcat Clustering Gbeans and David was doing some
> initial testing.
>
> Not sure about the timing. Looks like we'll make ApacheCon based on current
> course and speed. What's the right time to branch so we stabilize 1.0. I
> expect there are some other items continuing in development. My gut says we
> should branch no later than sometime next week and start TCK testing with the
> GBuild environment Blevins has been working on.
>
> Thoughts?
>
> Matt
>
>
>