To purge the local repo we have a jenkins plugin that can be used :
https://wiki.jenkins-ci.org/display/JENKINS/Maven+Repo+Cleaner+Plugin
(note that these days I have a doubt that it doesn't have a bug thus if we
are using it at Apache too it could help us to stabilize it :-) )
For real HTTP proxy
Note that it wasn't really called until r1179571, so I think maybe the
http-shared module should go back to not supporting it. I suppose that was
added for Aether, which in turn is no longer needed :)
While looking at that I noticed this code is probably being called for anything
using the new
purge local repo and any caching proxies in between that is
- Stephen
---
Sent from my Android phone, so random spelling mistakes, random nonsense
words and other nonsense are a direct result of using swype to type on the
screen
On 30 Nov 2011 05:16, "Stephen Connolly"
wrote:
> yep... now all w
yep... now all we need is a way to easily purge if the vote is cancelled
- Stephen
---
Sent from my Android phone, so random spelling mistakes, random nonsense
words and other nonsense are a direct result of using swype to type on the
screen
On 30 Nov 2011 00:24, "Barrie Treloar" wrote:
> On We
On Wed, Nov 30, 2011 at 9:33 AM, Benson Margulies wrote:
>> Are you able to give an overview or point me at the docs I haven't read yet.
>
> So, whenever we stage a release, Nexus creates a private repository
> just for that one release of that one item. That's the URL sent around
> in release vot
On Tue, Nov 29, 2011 at 5:37 PM, Barrie Treloar wrote:
> On Wed, Nov 30, 2011 at 3:34 AM, Benson Margulies
> wrote:
>> https://issues.apache.org/jira/browse/INFRA-4163
>
> I read that issue, but I don't understand how its meant to be used.
> And I suspect its my lack of understanding of Nexus.
>
On Wed, Nov 30, 2011 at 3:34 AM, Benson Margulies wrote:
> https://issues.apache.org/jira/browse/INFRA-4163
I read that issue, but I don't understand how its meant to be used.
And I suspect its my lack of understanding of Nexus.
Are you able to give an overview or point me at the docs I haven't
https://issues.apache.org/jira/browse/INFRA-4163
-
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
Hi,
We solved N issues:
https://jira.codehaus.org/secure/ReleaseNote.jspa?projectId=11126&version=13669
There are still a couple of issues left in JIRA:
https://jira.codehaus.org/secure/IssueNavigator.jspa?reset=true&jqlQuery=project+%3D+MASSEMBLY+AND+status+%3D+Open+ORDER+BY+priority+DESC&mode=h
Salut Arnaud!
of course, I need to check my designer skills but I'm sure we can do it :)
In that case I think that missing alignment is due to the fact that
the right logo is bigger than the feather one and the Bootstrap css
pushed on top images.
Thanks for the feedback!
Simo
http://people.apache.
Oh, good, I don't need to do anything.
On Tue, Nov 29, 2011 at 10:17 AM, Dennis Lundberg
wrote:
> Hi
>
> Yes, it's a bug in the site plugin. It's fixed in the next version.
>
> On Tuesday, November 29, 2011, Benson Margulies
> wrote:
>>
> http://maven.apache.org/shared/maven-common-artifact-filt
On Tue, Nov 29, 2011 at 10:25 AM, Stephen Connolly
wrote:
> Will need the jenkins jobs to clean their local repo periodically in case
> of re-votes
>
> On 29 November 2011 15:24, Benson Margulies wrote:
>
>> Indeed nexus has a feature that allows a single, persistent, group, to
>> map to whatever
Will need the jenkins jobs to clean their local repo periodically in case
of re-votes
On 29 November 2011 15:24, Benson Margulies wrote:
> Indeed nexus has a feature that allows a single, persistent, group, to
> map to whatever staging repos we have pending. I've opened an INFRA
> ticket to make
Indeed nexus has a feature that allows a single, persistent, group, to
map to whatever staging repos we have pending. I've opened an INFRA
ticket to make use of this, and I will proceed to improve the Jenkins
jobs to use it unless someone objects.
--
Hi
Yes, it's a bug in the site plugin. It's fixed in the next version.
On Tuesday, November 29, 2011, Benson Margulies
wrote:
>
http://maven.apache.org/shared/maven-common-artifact-filters-1.4/shared/maven-common-artifact-filters/
>
> Is this on purpose or by accident? There seem to be two extra
The site is in fact sitting at
http://maven.apache.org/shared/maven-common-artifact-filters-1.4/shared/maven-common-artifact-filters/.
On Tue, Nov 29, 2011 at 6:44 AM, Benson Margulies
wrote:
> Hi,
>
> We solved an issue:
> http://jira.codehaus.org/secure/ReleaseNote.jspa?atl_token=ACIO-CAVI-QX
http://maven.apache.org/shared/maven-common-artifact-filters-1.4/shared/maven-common-artifact-filters/
Is this on purpose or by accident? There seem to be two extra directories to me.
-
To unsubscribe, e-mail: dev-unsubscr...@mav
2011/11/29 Arnaud Héritier :
> What I'm doing at eXo to manage such cascading releases without having a
> huge manual process to keep everything up-to-date is that I have a profile
> that allow to activate a group with all staging repositories. Jenkins is
> allowed to take artifacts from this stagi
I think doing RCs like we have done in the past for all other 3.x releases
might be a good idea.
On Nov 29, 2011, at 4:00 AM, Olivier Lamy wrote:
> So due to issues when releasing project and missing md5/sha1 on some
> env whereas not on other.
> The vote is cancelled for more investigations.
>
What I'm doing at eXo to manage such cascading releases without having a
huge manual process to keep everything up-to-date is that I have a profile
that allow to activate a group with all staging repositories. Jenkins is
allowed to take artifacts from this staging group BUT it has to delete all
loc
I'd just vote on both at the same time... jenkins can fail if it wants...
or I can see about getting Apache a license for skippy (a.k.a. CloudBee's
Skip Builds plugin)
On 29 November 2011 13:22, Benson Margulies wrote:
> This morning, I considered the following:
>
> 1) Stage maven-common-filters
2011/11/29 Benjamin Bentmann :
> Tamás Cservenák wrote:
>
>> Out of curiosity, why is chunked transfer encoding used to transfer
>> the few byte long SHA1 string?
>
>
> https://issues.sonatype.org/browse/AETHER-128
>
>
>> many servers does not support it
>> completely...
>
>
> Which suggests to dep
Tamás Cservenák wrote:
Out of curiosity, why is chunked transfer encoding used to transfer
the few byte long SHA1 string?
https://issues.sonatype.org/browse/AETHER-128
many servers does not support it
completely...
Which suggests to deprecate the problematic method in the StreamingWagon
A
2011/11/29 Olivier Lamy :
> Hello,
>
> I'd like to release Apache Maven SCM 1.6
>
> We fixed 18 issues:
> https://jira.codehaus.org/secure/ReleaseNote.jspa?version=17296&styleName=Text&projectId=10527&Create=Create
>
> Staged repo: https://repository.apache.org/content/repositories/maven-265/
>
> S
Just a useless thing : Could it be possible to have the feather logo
vertically centered and with the same width of the Maven logo to have it
aligned ?
My 0,0001 cents
On Tue, Nov 29, 2011 at 2:26 PM, Olivier Lamy wrote:
> Hello,
>
> I'd like to release Apache Maven SCM 1.6
>
> We fixed
Hello,
I'd like to release Apache Maven SCM 1.6
We fixed 18 issues:
https://jira.codehaus.org/secure/ReleaseNote.jspa?version=17296&styleName=Text&projectId=10527&Create=Create
Staged repo: https://repository.apache.org/content/repositories/maven-265/
Staging site: http://maven.apache.org/scm-1
1.13.1 AFAIR on irc :-)
On Tue, Nov 29, 2011 at 2:17 PM, Olivier Lamy wrote:
> btw for who interested benjamin (thanks !) did a quick change on
> aether to have content length passed to wagon which fix the issue (a
> core it test has been changed to failed in case chunked is used)
>
> I'm waitin
This morning, I considered the following:
1) Stage maven-common-filters 1.4 and call a vote
2) Make a settings.xml for myself that included the staging repo.
3) Update maven-assembly-plugin to use maven-common-filters 1.4,
building using the settings.xml.
4) stage maven-assembly-plugin and call
CXF has a nice CachedOutputStream that fits into this space.
Let me know of I can help.
On Tue, Nov 29, 2011 at 6:35 AM, Olivier Lamy wrote:
> Agree I will cancel the vote to investigate more.
>
> I wonder why it works when deploying maven scm to
> repository.apache.org and tested on a locally i
btw for who interested benjamin (thanks !) did a quick change on
aether to have content length passed to wagon which fix the issue (a
core it test has been changed to failed in case chunked is used)
I'm waiting now the release 1.14 available in central to start again
the release process.
2011/11/
because for those files the streaming api is used.
something like putFromStream( InputStream stream, String destination )
so to know the size I will have to dump the stream to a tmp file to
calculate the size. (crazy)
2011/11/29 Tamás Cservenák :
> Out of curiosity, why is chunked transfer encod
Out of curiosity, why is chunked transfer encoding used to transfer
the few byte long SHA1 string?
Personally, I'd avoid the use of chunked whenever possible, since --
just like this example shows -- many servers does not support it
completely...
Thanks,
~t~
On Tue, Nov 29, 2011 at 1:21 PM, Oliv
Thanks.
That's what I have seen.
Content-Lenght is null because chunked is used.
I will enhance core it to ensure Content-Length header is correct and
update wagon http to fix that.
BTW canceling a release because a http server does not support chunck
is a pain
2011/11/29 Tamás Cservenák :
Hi,
Nexus does not respond with HTTP 411.
It has to be some nexus-fronting app (httpd or nginx?)
And I am 99% confident that according to RFC[1], this was a case of
incomplete request or such.
As I inspected, nexus@codehaus is fronted by nginx 0.8.24
By googling, I found this:
http://www.lamnk.
So due to issues when releasing project and missing md5/sha1 on some
env whereas not on other.
The vote is cancelled for more investigations.
2011/11/25 Olivier Lamy :
> Hello,
>
> I'd like to release Apache Maven 3.0.4.
>
> We fixed 31 issues.
> See release notes:
> http://jira.codehaus.org/secur
2011/11/29 Benjamin Bentmann :
> Olivier Lamy wrote:
>
>> I will add a core it test which check deploy of sha1 and md5.
>
>
> You might want to check
> MavenITmng4235HttpAuthDeploymentChecksumsTest
> before.
Thanks.
And btw this it pass well :-)
>
>
>
> Benjamin
>
> ---
Hi,
We solved an issue:
http://jira.codehaus.org/secure/ReleaseNote.jspa?atl_token=ACIO-CAVI-QX7G-9IAS%7C083f4f243731b9ee702e76ba2ccc07788ab70ac3%7Clin&version=16798&styleName=Text&projectId=11761&Create=Create
Release Notes - Maven Shared Components - Version
maven-common-artifact-filters-
Agree I will cancel the vote to investigate more.
I wonder why it works when deploying maven scm to
repository.apache.org and tested on a locally installed nexus instance
?
Can nexus folks who are listening here explains to me in which case
this happen http Return code: 411, ReasonPhrase:Length R
On 29 November 2011 11:17, Benson Margulies wrote:
> Folks,
>
> A couple of thoughts in response to Stephen:
>
> The convention seems pretty clear to send a 'yoo-hoo' before starting
> a release in case anyone is dying to add something. Since I had to do
> that anyway, I thought I might as well s
Hello,
shouldnt fluido have v1.1-SNAPSHOP in trunk? Patch attached.
Cheers
--
http://www.grobmeier.de
https://www.timeandbill.de
-
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-
Folks,
A couple of thoughts in response to Stephen:
The convention seems pretty clear to send a 'yoo-hoo' before starting
a release in case anyone is dying to add something. Since I had to do
that anyway, I thought I might as well solicit input on the version
number.
I did get poked once about t
Olivier Lamy wrote:
I will add a core it test which check deploy of sha1 and md5.
You might want to check
MavenITmng4235HttpAuthDeploymentChecksumsTest
before.
Benjamin
-
To unsubscribe, e-mail: dev-unsubscr...@maven.apac
I think delivering artifacts without md5 and sha1 is pretty much a blocker as
this might introduce security issues along the line.
So I'd say we investigate further before we go ahead.
Thus I change my vote to a
-1 (binding) also.
We now have a few cases which work well and a few cases which ca
my fault:
http://maven.apache.org/skins/maven-fluido-skin/source-repository.html
this shows the tag which is the version i have created the patch against
On Tue, Nov 29, 2011 at 10:57 AM, Christian Grobmeier
wrote:
> Hello,
>
> shouldnt fluido have v1.1-SNAPSHOP in trunk? Patch attached.
>
> Chee
I see the same checksum issue when trying to deploy a snapshot over at
Codehaus mojo:
https://nexus.codehaus.org/content/repositories/snapshots/org/codehaus/mojo/jboss-packaging-maven-plugin/2.3-SNAPSHOT/
I'm changing my vote to -1 (non-binding). IMHO we do not want end user
issues due to somethin
That stacktrace was with
org.apache.maven.wagon
wagon-http
2.1
in the pom...
if I try
org.apache.maven.wagon
wagon-http
1.0
I get
[DEBUG] Using connector WagonRepositoryConnector with priority 0 fo
ML didn't let me sent hte patches, here is the issue:
https://jira.codehaus.org/browse/MSKINS-13
On Tue, Nov 29, 2011 at 10:58 AM, Christian Grobmeier
wrote:
> Current search button in fludio is not using bootstrapped css. Here is
> patch which make it look more nicely.
>
> Cheers
> Christian
>
Even with a clean local repo, .md5's are still not being deployed for me...
If this is an issue with the embedded WAGON (and it is looking like it
could be) then I have to change my vote back negative
-0.9 (binding)
[DEBUG] Using connector WagonRepositoryConnector with priority 0 for
https://nex
Current search button in fludio is not using bootstrapped css. Here is
patch which make it look more nicely.
Cheers
Christian
--
http://www.grobmeier.de
https://www.timeandbill.de
-
To unsubscribe, e-mail: dev-unsubscr...@mave
@Stephen the issue was with wagon-http 2.0
see http://jira.codehaus.org/browse/WAGON-353
I will add a core it test which check deploy of sha1 and md5.
2011/11/29 Stephen Connolly :
> OK, just tried a simple deploy using a file: based wagon with the staged
> artifacts...
>
>
>> [stephenc@stephenc
for got to mention
On 29 November 2011 09:44, Stephen Connolly wrote:
> I have side-by side staging repos with 3.0.3 and 3.0.4, the only change
> between them was MAVEN_HOME
and PATH
>
> 3.0.3 has the md5's 3.0.4 doesn't for the exact same screen, and
> oss.sonatype.org refused to close the 3.
Attached are screenshots of a more recent example
On 29 November 2011 09:44, Stephen Connolly wrote:
> I have side-by side staging repos with 3.0.3 and 3.0.4, the only change
> between them was MAVEN_HOME
>
> 3.0.3 has the md5's 3.0.4 doesn't for the exact same screen, and
> oss.sonatype.org ref
I have side-by side staging repos with 3.0.3 and 3.0.4, the only change
between them was MAVEN_HOME
3.0.3 has the md5's 3.0.4 doesn't for the exact same screen, and
oss.sonatype.org refused to close the 3.0.4 deployed release last night
because its validation failed due to missing md5's
On 29 Nov
hi Stephen!
I did a local install of Apache MyFaces-2.1.6 to a private Nexus and I got the
md5 and sha1
Just to be sure: are you really looking under 'browse storage', because 'browse
index' doesn't show md5 and sha1 at all?
txs and LieGrue,
strub
- Original Message -
> From: Steph
OK, just tried a simple deploy using a file: based wagon with the staged
artifacts...
> [stephenc@stephenc ~]$ md5 ~/Downloads/apache-maven-3.0.4-bin.tar.gz
> MD5 (/Users/stephenc/Downloads/apache-maven-3.0.4-bin.tar.gz) =
> c1e67c7f32929b428266c88f7f62bee4
> [stephenc@stephenc ~]$ tar -xzvf
> ~/
Just tried a deploy @ mojo and having the same issue
I'll try for a minimal pom on local file system, see how that goes
On 29 November 2011 08:44, Olivier Lamy wrote:
> Just did a release of maven scm and md5/sha1 are there :
> https://repository.apache.org/content/repositories/maven-265/org
+1
2011/11/29 Stephane Nicoll :
> +1
>
> On Tue, Nov 29, 2011 at 9:32 AM, Kristian Rosenvold <
> kristian.rosenv...@zenior.no> wrote:
>
>> Judging by the current set of changes, I'd go for 2.2.2
>>
>> Kristian
>>
>> Den 29.11.2011 09:27, skrev Stephen Connolly:
>>
>> if you are stepping up to mak
Just did a release of maven scm and md5/sha1 are there :
https://repository.apache.org/content/repositories/maven-265/org/apache/maven/scm/maven-scm-api/1.6/
My env:
mbp-olamy:scm olamy$ which mvn
/Users/olamy/softs/maven/apache-maven-3.0.4/bin/mvn
mbp-olamy:scm olamy$ env | grep MAVEN_OPTS
MAVE
+1
On Tue, Nov 29, 2011 at 9:32 AM, Kristian Rosenvold <
kristian.rosenv...@zenior.no> wrote:
> Judging by the current set of changes, I'd go for 2.2.2
>
> Kristian
>
> Den 29.11.2011 09:27, skrev Stephen Connolly:
>
> if you are stepping up to make the release, you decide the version number,
>>
Judging by the current set of changes, I'd go for 2.2.2
Kristian
Den 29.11.2011 09:27, skrev Stephen Connolly:
if you are stepping up to make the release, you decide the version number,
and you decide if there is enough of a change to merrit your effort.
sounds like you feel it is worth your t
if you are stepping up to make the release, you decide the version number,
and you decide if there is enough of a change to merrit your effort.
sounds like you feel it is worth your time and you are adding a feature, so
it is not a bug fix => 2.3 or 3.0 depending obey how big that feature feels
to
61 matches
Mail list logo
