Preventing merge commits
by Jason T. Greene
If you want to prevent merge commits from ever entering a particular
branch of yours, say master, just do this:
git config branch.master.mergeoptions --ff-only
--
Jason T. Greene
JBoss AS Lead / EAP Platform Architect
JBoss, a division of Red Hat
13 years, 7 months
Limiting Server Thread Usage
by Jason T. Greene
Moving to a new thread.
The big problem we run into with this is that almost every application
of a thread pool that we have needs to be highly tailored to its usage
to get the most optimal performance. So we end up with quite a few
different pools and it becomes difficult to impose a server wide limit.
There however some potential strategies we could take. Although I am
unsure as to how the overall effectiveness would be:
1. Sharing idle threads between pools
2. Force everything to go through a special blocking thread factory via
instrumentation of java.lang.Thread. Any attempt to allocate over the
max would lead to thread reclamation attempts and finally blocking until
a timeout is reached.
3. Some kind of auto-tuning weighting model. If the max total threads is
N, force all thread pools to use a percentage of N, potentially based on
establishing current config value divided by combined total.
One thing I wonder though is if cloud providers are "barking up the
wrong tree"? It seems a better limitation of an application is raw CPU
clock time and max memory usage. How they split that time into threads
doesn't really affect the scalability of the physical server, it's all
virtual process performance (who cares if someone wastes time context
switching?).
On 9/14/11 10:39 AM, Scott Stark wrote:
> The other big cross cutting concern is controlling the total number of
> threads in use by the application server. When running under a
> constrained environment that uses something like pam_limits module to
> control how many process(==java threads) a user can have, it is
> difficult to know what the server max thread usage is right now.
>
--
Jason T. Greene
JBoss AS Lead / EAP Platform Architect
JBoss, a division of Red Hat
13 years, 9 months
Testsuite update
by Ondřej Žižka
Hi everyone,
as was discussed on this list, the current testsuite harness had some
problems fulfilling even basic requirements for various use cases,
including but not limited to:
* Using from IDE through pure maven
* -DallTests vs. individial modules vs. distinctive groups vs. even
overlaying groups
* Configuring AS using kind of hierarchical templates
* Imposing certain test groups order
* Invoking only some steps of the testsuite
* Running same tests with different AS configuration
etc etc.
I've been working on an updated concept, which is currently at
https://github.com/OndraZizka/jboss-as/commits/TS-modules-tmp .
I would like you to review it, run it, and send me some comments.
cd testsuite/integration
mvn clean install -f pom2.xml -DallTests -fae
mvn clean install -f pom2.xml -Dts.iiop
(There are 2 pom.xml's so I can compare two harnesses easily.)
The scripts are not updated yet so they do not work.
CLI and OSGi tests fail. To be sorted out.
The big news are:
* Execution split to modules
* Code and resources remained in the same location
* Added few system properties which provide paths to important
directories - see https://docs.jboss.org/author/display/AS71/AS+7
+Testsuite+Test+Developer+Guide
* Work directory is now in target/workdir
These cahnges exposed some issues in test code which need to be fixed -
hardcoded relative paths.
Since there were no system properties to use, there was no other way,
but from now on please use those properties, and fix your existing
tests.
(This would happen anyway because some tests clutter workdir (which
happens to be module's root dir) with various temp files.)
Thanks,
Ondra
14 years
Re: [jboss-as7-dev] Problem met on as7 clustering with mod_cluster
by Weinan Li
On Jan 20, 2012, at 1:26 AM, jean-frederic clere wrote:
> That isn't an error in httpd configuration, but probably in the connector of the web subsystem in AS7.
>
> Mail me all the files I will have a look later.
>
I've tried to reconfigure all things again and the problem get solved! I guess the problem is caused by different 'port-offset' in 'socket-bindings' settings in host.xml between master and slave.
There was a typo in port-offset of master, so I guess mod_cluster is try to use the offset setting in master, and when it's trying to use the same offset to connect to slave, it throws error. But it's really weird it say cannot connect to 'localhost' instead of the ip address of slave.
Anyway, the cluster works now! Thank you for help Jean-Frederic :-)
Success Log:
[Server:server-three] 02:21:43,139 INFO [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (Incoming-4,null) ISPN000094: Received new cluster view: [server-three/web|1] [server-three/web, server-three-slave/web]
Now there's another question: How can I configure mod_cluster to use round-robin to distribute requests?
I've tried to access cluster with different computers/browsers, but it seems all the requests are direct to master. Can I configure this in domain.xml?
> Cheers
>
> Jean-Frederic
Cheers,
- Weinan Li
JBoss, Redhat
14 years
Parallel initialization of more web applications in EAR
by Marek Novotny
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi AS developers,
I am exploring the behavior of parallel initialization of web
application in EAR. More concretely more than one web application in EAR.
I know that spec doesn't define the initialization order of EAR
submodules, but the parallel initialization of web application is new in
AS7 and this breaks the backward compatibility for existing enterprise
applications with multiple wars inside. New Java Enterprise application
6 configuration setting <initialize-in-order>true</initialize-in-order>
set up in application.xml can enable proper initialization flow, but
without it is not possible to deploy.
Could somebody shed light on parallel web applicaiton init? Original
forum post https://community.jboss.org/message/635123
And details could be seen also at
https://issues.jboss.org/browse/JBSEAM-4856?focusedCommentId=12662448&pag...
Cheers,
- --
Marek Novotny
- --
JBoss Seam and WFK Productization Lead
Red Hat Czech s.r.o.
Purkynova 99
612 45 Brno
Email: mnovotny(a)redhat.com
Office phone: +420 532 294 287, ext. 82-62 087
mobile: +420 608 509 230
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iEYEARECAAYFAk8nEsgACgkQU4HO8G8hNxV0jgCfXmgIy5a5mUctcy35Xg8SIky+
ubkAoOrmfoF7ByaQjUX4Wbfu+3w5sAH1
=Uv1K
-----END PGP SIGNATURE-----
14 years, 1 month
master/trunk snapshot issues for CLI and -ds.xml
by Max Rydahl Andersen
Hi,
I wanted to try out AS7 master tonight to test out the new -ds.xml features but I ran into a bunch of issues;
and since it seem to be such basic issues I didn't want to pollute jira upfront.
Is any of this expected currently:
A) jboss-cli --connect just freezes
B) jboss-cli --gui loads UI but NPE's and clicking causes more errors
C) deploying -ds.xml with xadatasource fails after logging a lot of errors/warnings about private api
"[org.jboss.as.dependency.private] (MSC service thread 1-1) JBAS018567: Deployment "deployment.wonka.war" is using a private module ("<x.y.z>") which may be changed or removed in future versions without notice"
D) deploying -ds.xml without xadatasource seem to work but once app is trying to access logs gets filled with more:
[org.jboss.as.dependency.private] (MSC service thread 1-1) JBAS018567: Deployment "deployment.wonka.war" is using a private module ("org.jboss.iiop-client:main") which may be changed or removed in future versions without notice
E) I can't figure out how to reference it in persistence.xml or resource lookups.
Using the datasource name, i.e. java:jboss/datasources/DeployedDS does not work in persistence.xml
Tried both with -ds.xml inside and outside war.
note, I did nothing special - just a servlet and a CDI bean + persistence.xml; trying to keep it as simple as possible.
should I open lira's or is master considered non-functional in these areas at the moment ? :)
/max
http://about.me/maxandersen
14 years, 1 month
Not Backward Compatible JBossWS subsystem DMR model change
by Richard Opalka
Hello everybody,
this is ask for confirmation that nobody's affected.
Considering the fact DMR model for <endpoint-config>
WS subsystem element didn't work at all in AS 7.1.0.CR1, I believe that
nobody should be affected, but to be 100% sure, here I go.
This is related to this pull request:
https://github.com/jbossas/jboss-as/pull/1263
There are many non backward compatible DMR model changes
for <endpoint-config> element of WS subsystem introduced.
The pull request will be accepted when there's nobody affected by this
change.
Is anybody out there using DMR model of <endpoint-config> WS subsystem
element?
I need confirmation namely from:
* Heiko Braun
* Heiko Rupp,
* Alexey Loubyansky
* anyone affected
Thanks in advance,
Rio
--
Richard Opalka
JBoss, by Red Hat
Cell: +420 731 186 942
14 years, 1 month
