from:"asookazian2"

I'm able to reproduce the following exception (at least the first one which
happens on startup of Karaf).  Karaf 3.0.2 and we've replaced BoneCP with
HikariCP-java6 2.2.2 recently as BoneCP has been deprecated.

Please note that in my Spring applicationConfig.xml for this WAR
mega-bundle, I'm using the following to lookup the JDBC datasource (which
was working previously without these exceptions with BoneCP):

  jee:jndi-lookup id=dataSource
jndi-name=osgi:service/jdbc/FooDataSource/

What is the recommended way of doing the JNDI lookup here?  jee:jndi-lookup
is a Spring specific way but I'm not sure is recommended for OSGi.

The interesting thing to note is that there are some inserts happening into
the DB in question after startup when I execute some test cases from the web
app front end.

20141203 15:43:47.320 [ERROR] Thread-66 | 273:my-project-server |
org.activiti.engine.impl.jobexecutor.AcquireJobsRunnable | exception during
job acquisition: Could not open JDBC Connection for transaction; nested
exception is org.osgi.framework.ServiceException
org.springframework.transaction.CannotCreateTransactionException: Could not
open JDBC Connection for transaction; nested exception is
org.osgi.framework.ServiceException
at
org.springframework.jdbc.datasource.DataSourceTransactionManager.doBegin(DataSourceTransactionManager.java:240)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.springframework.transaction.support.AbstractPlatformTransactionManager.getTransaction(AbstractPlatformTransactionManager.java:371)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.springframework.transaction.support.TransactionTemplate.execute(TransactionTemplate.java:127)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.activiti.spring.SpringTransactionInterceptor.execute(SpringTransactionInterceptor.java:40)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.activiti.engine.impl.interceptor.LogInterceptor.execute(LogInterceptor.java:33)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.activiti.engine.impl.jobexecutor.AcquireJobsRunnable.run(AcquireJobsRunnable.java:57)[273:my-project-server:1.0.0.SNAPSHOT]
at java.lang.Thread.run(Thread.java:745)[:1.7.0_71]
Caused by: org.osgi.framework.ServiceException
at
org.apache.aries.jndi.services.ServiceHelper$JNDIServiceDamper.call(ServiceHelper.java:197)[98:org.apache.aries.jndi.url:1.0.0]
at Proxy3e789f04_5f82_464a_ab5b_df52583e7c46.getConnection(Unknown
Source)[:]
at
org.springframework.jdbc.datasource.DataSourceTransactionManager.doBegin(DataSourceTransactionManager.java:202)[273:my-project-server:1.0.0.SNAPSHOT]
... 6 more




20141204 08:40:37.490 [ERROR] Thread-67 | 273:my-project-server |
org.activiti.engine.impl.jobexecutor.AcquireJobsRunnable | exception during
job acquisition: Could not open JDBC Connection for transaction; nested
exception is java.sql.SQLException: Timeout after 47143144ms of waiting for
a connection.
org.springframework.transaction.CannotCreateTransactionException: Could not
open JDBC Connection for transaction; nested exception is
java.sql.SQLException: Timeout after 47143144ms of waiting for a connection.
at
org.springframework.jdbc.datasource.DataSourceTransactionManager.doBegin(DataSourceTransactionManager.java:240)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.springframework.transaction.support.AbstractPlatformTransactionManager.getTransaction(AbstractPlatformTransactionManager.java:371)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.springframework.transaction.support.TransactionTemplate.execute(TransactionTemplate.java:127)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.activiti.spring.SpringTransactionInterceptor.execute(SpringTransactionInterceptor.java:40)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.activiti.engine.impl.interceptor.LogInterceptor.execute(LogInterceptor.java:33)[273:my-project-server:1.0.0.SNAPSHOT]
at
org.activiti.engine.impl.jobexecutor.AcquireJobsRunnable.run(AcquireJobsRunnable.java:57)[273:my-project-server:1.0.0.SNAPSHOT]
at java.lang.Thread.run(Thread.java:745)[:1.7.0_71]
Caused by: java.sql.SQLException: Timeout after 47143144ms of waiting for a
connection.
at
com.zaxxer.hikari.pool.HikariPool.getConnection(HikariPool.java:205)[231:com.zaxxer.HikariCP-java6:2.2.2]
at
com.zaxxer.hikari.HikariDataSource.getConnection(HikariDataSource.java:108)[231:com.zaxxer.HikariCP-java6:2.2.2]
at Proxydb8a93e5_8d4e_43a2_b487_005e667ffa99.getConnection(Unknown
Source)[:]
at Proxy479ae98e_177d_4ad4_8509_6f6eb9d96b88.getConnection(Unknown
Source)[:]
at
org.springframework.jdbc.datasource.DataSourceTransactionManager.doBegin(DataSourceTransactionManager.java:202)[273:my-project-server:1.0.0.SNAPSHOT]
... 6 more



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Could-not-open-JDBC-connection-for-transaction-tp4036968.html
Sent from the Karaf - User mailing

Re: Could not open JDBC connection for transaction

I installed the BoneCP bundle and reverted back to the BoneCP version of the
datasource xml and it still reproduces...

that's interesting as the applicationConfig.xml and datasources have not
been modified recently...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Could-not-open-JDBC-connection-for-transaction-tp4036968p4036970.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Could not open JDBC connection for transaction

I have enabled DEBUG level logging on a Spring and Activiti class and I've
determined that the first attempt fails with the exception but the 2nd
attempt to acquire jobs (30 secs later) is successful...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Could-not-open-JDBC-connection-for-transaction-tp4036968p4036971.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Using Spring with Karaf 3.x

We have some legacy J2EE apps that we've ported (finally) to Karaf 3.0.x
which use Spring.  Mainly for dependency injection and possibly some
templates (e.g. JdbcTemplate, etc.) and AOP.

Anyways, looks like there are no Spring (or Spring dm?) examples here:
https://github.com/cschneider/Karaf-Tutorial and there are no Spring
examples in the Apache Karaf Cookbook.

Is Spring usage with OSGi/Karaf recommended or not?  If yes, where can I
find some examples?

I stopped using spring on OSGi a long time ago.

http://stackoverflow.com/questions/24595900/invalid-bundle-when-starting-apache-karaf



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Using-Spring-with-Karaf-3-x-tp4036977.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Karaf skipping feature install

We have a feature install as follows:

feature name=myFeature version=1.0.0-SNAPSHOT resolver=(obr)
start-level=90 description=description

feature version=1.0.0-SNAPSHOTfeature1/feature
feature version=1.0.0-SNAPSHOTfeature2/feature

bundlemvn:com.foo.bar/component1/1.0.0-SNAPSHOT/bundle
/feature

feature name=feature2 version=1.0.0-SNAPSHOT resolver=(obr)
start-level=90 description=description

feature version=1.0.0-SNAPSHOTfeature1/feature

bundlemvn:com.foo.bar/component2/1.0.0-SNAPSHOT/bundle
/feature

In a particular Karaf 3.0.2 server (with older versions of our software),
component1 is being installed successfully but component2 is not.

In a different Karaf 3.0.2 server (with latest versions of our software),
component1 and component2 is being installed successfully.

In the former case, I did not see any exceptions in the karaf log so it
seems as if karaf is skipping the compononent2 install.  I verified that the
JAR does indeed exist in the karaf/data/repo directory structure.

Has anybody else seen this and can you provide an explanation?

I verified in the Karaf cmd line console using 'list | grep -i
componentName' that the bundle was not already installed...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Karaf-skipping-feature-install-tp4036927.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Unable to uninstall multiple bundles

4 bundles suddenly and simultaneously go into installed state after karaf
restart (previously were active).  I can't uninstall the bundles (bundle
is invalid is the error at cmd line).  karaf 3.0.2.  one of them is simply a
datasource.xml in deploy dir.

i had to delete from karaf/data/cache dir to get rid of them.

wtf



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Karaf skipping feature install

This envmt is using only one karaf.

I was able to reproduce this behavior with a new Karaf install and latest of
our software. 

However, it behaved normally on the 1st install attempt (i.e. component1 and
component2 were both installed and active).  However, on the 2nd install
attempt, the behavior reproduced.  Maybe it's possible I am not uninstalling
both features using feature:uninstall foo before the re-install and that
is causing this problem/behavior?


jbonofre wrote
 Do you have the obr feature installed on one Karaf ?
 
 Regards
 JB
 
 On 12/03/2014 09:25 PM, asookazian2 wrote:
 We have a feature install as follows:

 
 feature name=myFeature version=1.0.0-SNAPSHOT resolver=(obr)

  start-level=90 description=description

  
 feature version=1.0.0-SNAPSHOT
 feature1
 /feature
  
 feature version=1.0.0-SNAPSHOT
 feature2
 /feature

  
 bundle
 mvn:com.foo.bar/component1/1.0.0-SNAPSHOT
 /bundle
  
 /feature

 
 feature name=feature2 version=1.0.0-SNAPSHOT resolver=(obr)

  start-level=90 description=description

  
 feature version=1.0.0-SNAPSHOT
 feature1
 /feature

  
 bundle
 mvn:com.foo.bar/component2/1.0.0-SNAPSHOT
 /bundle
  
 /feature

 In a particular Karaf 3.0.2 server (with older versions of our software),
 component1 is being installed successfully but component2 is not.

 In a different Karaf 3.0.2 server (with latest versions of our software),
 component1 and component2 is being installed successfully.

 In the former case, I did not see any exceptions in the karaf log so it
 seems as if karaf is skipping the compononent2 install.  I verified that
 the
 JAR does indeed exist in the karaf/data/repo directory structure.

 Has anybody else seen this and can you provide an explanation?

 I verified in the Karaf cmd line console using 'list | grep -i
 
 componentName
 ' that the bundle was not already installed...



 --
 View this message in context:
 http://karaf.922171.n3.nabble.com/Karaf-skipping-feature-install-tp4036927.html
 Sent from the Karaf - User mailing list archive at Nabble.com.

 
 -- 
 Jean-Baptiste Onofré

 jbonofre@

 http://blog.nanthrax.net
 Talend - http://www.talend.com





--
View this message in context: 
http://karaf.922171.n3.nabble.com/Karaf-skipping-feature-install-tp4036927p4036931.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Unable to uninstall multiple bundles

I don't have the log available that would pertain to this problem.

I did try to bundle:stop and bundle:start and both of those failed (I
think with invalid bundle error).

from etc/config.properties:

#
# Framework selection properties
#
karaf.framework=felix


jbonofre wrote
 What's in the log (when the bundles went in installed state) ?
 Did you try to start by hand ?
 
 I bet that you use the Felix framework (default one in Karaf): it's 
 probably the bug on the framework when the cache is corrupted.
 
 Are you able to reproduce the scenario ?
 
 Regards
 JB
 
 On 12/03/2014 10:22 PM, asookazian2 wrote:
 4 bundles suddenly and simultaneously go into installed state after
 karaf
 restart (previously were active).  I can't uninstall the bundles
 (bundle
 is invalid is the error at cmd line).  karaf 3.0.2.  one of them is
 simply a
 datasource.xml in deploy dir.

 i had to delete from karaf/data/cache dir to get rid of them.

 wtf



 --
 View this message in context:
 http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928.html
 Sent from the Karaf - User mailing list archive at Nabble.com.

 
 -- 
 Jean-Baptiste Onofré

 jbonofre@

 http://blog.nanthrax.net
 Talend - http://www.talend.com





--
View this message in context: 
http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928p4036933.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Unable to uninstall multiple bundles

how and why does the cache get corrupted?

we are under the impression that we need to halt/start karaf after doing our
feature installs in order to avoid any corruption issues but not sure if
this is really required or not.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928p4036934.html
Sent from the Karaf - User mailing list archive at Nabble.com.

cascading feature uninstall option

Is there a cascading feature:uninstall option?  I checked --help and didn't
see it.

I mean if you have bundles and features in a single feature.xml, currently
if you feature:uninstall that feature it will only uninstall the bundles
embedded in it, not necessarily the other embedded features.

I'd like an option that will uninstall everything...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/cascading-feature-uninstall-option-tp4036938.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Unable to uninstall multiple bundles

reproduced exact same 4 bundles are in installed state after restart and
error states Bundle x is invalid when I try to stop the bundle.  This
occurred after I killed the JVM as the halt cmd in karaf was hanging.

stack trace before JVM stopped:

20141203 14:49:35.376 [ERROR] Thread-116 | : |
org.activiti.engine.impl.jobexecutor.AcquireJobsRunnable | exception during
job acquisition: Could not open JDBC Connection for transaction; nested
exception is java.lang.IllegalStateException: Invalid BundleContext.
org.springframework.transaction.CannotCreateTransactionException: Could not
open JDBC Connection for transaction; nested exception is
java.lang.IllegalStateException: Invalid BundleContext.
at
org.springframework.jdbc.datasource.DataSourceTransactionManager.doBegin(DataSourceTransactionManager.java:240)[284:workflow-ws-server:1.0.0.SNAPSHOT]
at
org.springframework.transaction.support.AbstractPlatformTransactionManager.getTransaction(AbstractPlatformTransactionManager.java:371)[284:workflow-ws-server:1.0.0.SNAPSHOT]
at
org.springframework.transaction.support.TransactionTemplate.execute(TransactionTemplate.java:127)[284:workflow-ws-server:1.0.0.SNAPSHOT]
at
org.activiti.spring.SpringTransactionInterceptor.execute(SpringTransactionInterceptor.java:40)[284:workflow-ws-server:1.0.0.SNAPSHOT]
at
org.activiti.engine.impl.interceptor.LogInterceptor.execute(LogInterceptor.java:33)[284:workflow-ws-server:1.0.0.SNAPSHOT]
at
org.activiti.engine.impl.jobexecutor.AcquireJobsRunnable.run(AcquireJobsRunnable.java:57)[284:workflow-ws-server:1.0.0.SNAPSHOT]
at java.lang.Thread.run(Thread.java:745)[:1.7.0_71]
Caused by: java.lang.IllegalStateException: Invalid BundleContext.
at
org.apache.felix.framework.BundleContextImpl.checkValidity(BundleContextImpl.java:514)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleContextImpl.getServiceReferences(BundleContextImpl.java:425)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.aries.jndi.services.ServiceHelper.findService(ServiceHelper.java:375)[98:org.apache.aries.jndi.url:1.0.0]
at
org.apache.aries.jndi.services.ServiceHelper.access$500(ServiceHelper.java:66)[98:org.apache.aries.jndi.url:1.0.0]
at
org.apache.aries.jndi.services.ServiceHelper$JNDIServiceDamper.call(ServiceHelper.java:180)[98:org.apache.aries.jndi.url:1.0.0]
at Proxy386d92f9_189d_4945_b1ed_57c9fecb6138.getConnection(Unknown
Source)[:]
at
org.springframework.jdbc.datasource.DataSourceTransactionManager.doBegin(DataSourceTransactionManager.java:202)[284:workflow-ws-server:1.0.0.SNAPSHOT]
... 6 more



jbonofre wrote
 No it's not. Without details about your bundles, what they do and how they
 work, it's not easy.
 
 We have a bunch of users on Karaf and that use the features without
 problem. So I suspect something related to your distribution or
 application.
 
 Regards
 JB
 
 
 Sent from my Samsung Galaxy smartphone.
 
 
  Original message 
 From: asookazian2 lt;

 asookazian@

 gt; 
 Date:03/12/2014  22:41  (GMT+01:00) 
 To: 

 user@.apache

  
 Cc:  
 Subject: Re: Unable to uninstall multiple bundles 
 
 how and why does the cache get corrupted?
 
 we are under the impression that we need to halt/start karaf after doing
 our
 feature installs in order to avoid any corruption issues but not sure if
 this is really required or not.
 
 
 
 --
 View this message in context:
 http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928p4036934.html
 Sent from the Karaf - User mailing list archive at Nabble.com.





--
View this message in context: 
http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928p4036941.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Unable to uninstall multiple bundles

also i don't understand why those specific 4 (last installed) bundles have
presumably become corrupted.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Unable-to-uninstall-multiple-bundles-tp4036928p4036943.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: akhon r we can not kotha bolte parbo na ata kokhono somvob na............

what the hell is this random spam going on all the time here?  



--
View this message in context: 
http://karaf.922171.n3.nabble.com/akhon-r-we-can-not-kotha-bolte-parbo-na-ata-kokhono-somvob-na-tp4036950p4036951.html
Sent from the Karaf - User mailing list archive at Nabble.com.

trouble uninstalling feature

2014-11-24 Thread asookazian2

I'm trying to uninstall a feature prior to installing it.

I see in the karaf cmd line (feature:repo-list) that there is only one repo
added for this feature name.

However, when I issue 'feature:uninstall foo' it fails and states that I
must specify a specific version b/c there are three versions for the same
feature name.

How does karaf find this list of features/versions?  I have grep'd my repo
and system folders for the feature name and only find the 1.0.0-SNAPSHOT
version for this feature.

Is it possible to do something like 'feature:uninstall foo *' and have karaf
uninstall all versions of the same feature?

Also, prior to any installations I delete the related folders for this
feature in repo and system.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/trouble-uninstalling-feature-tp4036646.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: trouble uninstalling feature

2014-11-24 Thread asookazian2

karaf 3.0.2



--
View this message in context: 
http://karaf.922171.n3.nabble.com/trouble-uninstalling-feature-tp4036646p4036647.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: trouble uninstalling feature

2014-11-24 Thread asookazian2

Thx for reply.  So I did 'feature:list -i' and see only 1.0.0-SNAPSHOT
version, the other versions it claims are there are not there for this
feature name.  Something seems wrong...


jbonofre wrote
 feature:list -i
 
 feature:uninstall foo/version
 
 You can do a script to uninstall all features with the same name (and 
 different version).
 
 Regards
 JB
 
 On 11/24/2014 11:24 PM, asookazian2 wrote:
 I'm trying to uninstall a feature prior to installing it.

 I see in the karaf cmd line (feature:repo-list) that there is only one
 repo
 added for this feature name.

 However, when I issue 'feature:uninstall foo' it fails and states that I
 must specify a specific version b/c there are three versions for the same
 feature name.

 How does karaf find this list of features/versions?  I have grep'd my
 repo
 and system folders for the feature name and only find the 1.0.0-SNAPSHOT
 version for this feature.

 Is it possible to do something like 'feature:uninstall foo *' and have
 karaf
 uninstall all versions of the same feature?

 Also, prior to any installations I delete the related folders for this
 feature in repo and system.



 --
 View this message in context:
 http://karaf.922171.n3.nabble.com/trouble-uninstalling-feature-tp4036646.html
 Sent from the Karaf - User mailing list archive at Nabble.com.

 
 -- 
 Jean-Baptiste Onofré

 jbonofre@

 http://blog.nanthrax.net
 Talend - http://www.talend.com





--
View this message in context: 
http://karaf.922171.n3.nabble.com/trouble-uninstalling-feature-tp4036646p4036650.html
Sent from the Karaf - User mailing list archive at Nabble.com.

restarting Karaf after uninstalling bundles or features

2014-11-21 Thread asookazian2

Karaf 3.0.1

is it necessary to restart Karaf after uninstalling bundles or features to
avoid a missing bundles problem we've been seeing?  Sometimes and suddenly,
some bundles disappear when you type 'list'.  Has anybody else seen this and
what is the best practice when uninstalling then re-installing the same
feature?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/restarting-Karaf-after-uninstalling-bundles-or-features-tp4036577.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: restarting Karaf after uninstalling bundles or features

2014-11-21 Thread asookazian2

I've only seen this happen with our custom bundles...


jbonofre wrote
 Hi,
 
 no, normally it's not necessary to restart Karaf (it depends of the 
 bundles, but generally speaking), and I didn't see such kind of issue.
 Is it only with ouyr own bundles, or with Karaf bundles as well ?
 
 Regards
 JB
 
 On 11/21/2014 08:07 PM, asookazian2 wrote:
 Karaf 3.0.1

 is it necessary to restart Karaf after uninstalling bundles or features
 to
 avoid a missing bundles problem we've been seeing?  Sometimes and
 suddenly,
 some bundles disappear when you type 'list'.  Has anybody else seen this
 and
 what is the best practice when uninstalling then re-installing the same
 feature?



 --
 View this message in context:
 http://karaf.922171.n3.nabble.com/restarting-Karaf-after-uninstalling-bundles-or-features-tp4036577.html
 Sent from the Karaf - User mailing list archive at Nabble.com.

 
 -- 
 Jean-Baptiste Onofré

 jbonofre@

 http://blog.nanthrax.net
 Talend - http://www.talend.com





--
View this message in context: 
http://karaf.922171.n3.nabble.com/restarting-Karaf-after-uninstalling-bundles-or-features-tp4036577p4036588.html
Sent from the Karaf - User mailing list archive at Nabble.com.

RE: non-deterministic installation order with ds files in deploy dir

2014-11-17 Thread asookazian2

Thx for responses.  How would the user modify the foo-ds.xml post-install
using these methods?  It is easy to do if it's in the deploy dir.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/non-deterministic-installation-order-with-ds-files-in-deploy-dir-tp4036433p4036479.html
Sent from the Karaf - User mailing list archive at Nabble.com.

non-deterministic installation order with ds files in deploy dir

2014-11-14 Thread asookazian2

I've been testing a feature install which requires a foo-ds.xml (datasource)
to be copied into the karaf/deploy dir first and then the 'feature:install
bar' executed next.  It seems that the order of installation of the foo-ds
bundle may be before or after the feature bundle installs.  If the foo-ds
bundle is activated after the depending bundle is activated, then I see an
exception in the log (something like a JNDI lookup error and the datasource
can't be found).  If I restart karaf, all bundles end up in Active state due
to the start-level configs (the foo-ds.xml bundle start-level is 80 and the
depending bundle is 85).

What is the best way to handle this so all bundles are guaranteed to be
Active state after feature install?  We've discussed adding a sleep
seconds=5/ after the foo-ds.xml copy to deploy dir and prior to the
feature install.  Is there any better way to handle this?  Another idea was
to have a loop in the ant install script which checks the foo-ds.xml
bundle's state and once it is Active then install the feature.  thx.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/non-deterministic-installation-order-with-ds-files-in-deploy-dir-tp4036433.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: non-deterministic installation order with ds files in deploy dir

2014-11-14 Thread asookazian2

we are using Karaf 3.0.1



--
View this message in context: 
http://karaf.922171.n3.nabble.com/non-deterministic-installation-order-with-ds-files-in-deploy-dir-tp4036433p4036434.html
Sent from the Karaf - User mailing list archive at Nabble.com.

restart cmd

2014-11-08 Thread asookazian2

Is there a Karaf cmd in 3.0.1 that will restart the server from the console? 
equivalent to 'halt' and then 'bin/karaf'?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/restart-cmd-tp4036293.html
Sent from the Karaf - User mailing list archive at Nabble.com.

how to determine if a bundle has already been installed?

2014-11-07 Thread asookazian2

Let's say I have a feature set which requires that another feature already be
installed (i.e. a set of bundles should already be installed in Karaf
3.0.1).

What is the recommended method of determining if that bundle or feature set
is already installed?

Currently I am using the following in my ant target:

bundle:list | grep 'bundleFoo'



--
View this message in context: 
http://karaf.922171.n3.nabble.com/how-to-determine-if-a-bundle-has-already-been-installed-tp4036281.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: [ANN] Apache Karaf 4.0.0.M1 Released!

2014-10-15 Thread asookazian2

Please supply a link to list of new features in Karaf 4.0.0.  thx.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/ANN-Apache-Karaf-4-0-0-M1-Released-tp4035927p4035969.html
Sent from the Karaf - User mailing list archive at Nabble.com.

install on an existing bundle with same bundle-version fails

2014-10-10 Thread asookazian2

Please consider giving message/warning to Karaf cmd line/console when user
tries to install a bundle that already exists.  Currently, it seems to just
ignore the command and the bundle.jar in the data/cache/xxx dir is not
updated.

Please use update bundle_id instead of install on a pre-existing bundle,
something like that would be helpful.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/install-on-an-existing-bundle-with-same-bundle-version-fails-tp4035779.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Schema for custom shell commands

2014-09-29 Thread asookazian2

I am running into the same problem with Karaf 3.0.1.

Caused by:
org.osgi.service.blueprint.container.ComponentDefinitionException: Unable to
find a matching factory method getScope on class
org.apache.karaf.shell.console.commands.NamespaceHandler for arguments
[com.nextgate.accessmanager.user.setup.ProvisionUserCommand] when
instanciating bean #recipe-25
at
org.apache.aries.blueprint.container.BeanRecipe.getInstance(BeanRecipe.java:318)
at
org.apache.aries.blueprint.container.BeanRecipe.internalCreate2(BeanRecipe.java:806)
at
org.apache.aries.blueprint.container.BeanRecipe.internalCreate(BeanRecipe.java:787)
at
org.apache.aries.blueprint.di.AbstractRecipe.create(AbstractRecipe.java:106)
at
org.apache.aries.blueprint.di.MapRecipe.internalCreate(MapRecipe.java:111)

I am getting this exception with v1.1.0 and v1.0.0.  Please advise as I
cannot continue testing at this point.

This is my blueprint.xml:

?xml version=1.0 encoding=UTF-8? 
blueprint xmlns=http://www.osgi.org/xmlns/blueprint/v1.0.0;
  
command-bundle xmlns=http://karaf.apache.org/xmlns/shell/v1.1.0;
command
action
class=com.nextgate.accessmanager.user.setup.ProvisionUserCommand/
/command
/command-bundle
/blueprint

Interestingly, a different class which extends OsgiCommandSupport with a
different blueprint.xml in the same Karaf instance is NOT having this
problem (it's using v1.1.0).  I'm not sure what the difference is, the
blueprint.xml files are very similar.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Schema-for-custom-shell-commands-tp4029047p4035579.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Class com/sun/org/apache/xerces/internal/dom/ElementImpl illegally accessing package private member of class com/sun/org/apache/xerces/internal/dom/CoreDocumentImpl

2014-08-26 Thread asookazian2

We ran into the following error with our rebranded version of Karaf 3.0.1 on
AIX (does not reproduce on Windows or Linux):

body
HTTP ERROR 500

pProblem accessing /ws/wf/WorkflowManagerWS. Reason:
preServer Error/pre/p
Caused by:
prejava.lang.IllegalAccessError: Class
com/sun/org/apache/xerces/internal/dom/ElementImpl illegally accessing
package private member of class
com/sun/org/apache/xerces/internal/dom/CoreDocumentImpl

As suggested in the following thread:

http://comments.gmane.org/gmane.comp.apache.cxf.user/30677

we removed the following file:
apache-karaf-3.0.1/lib/endorsed/org.apache.servicemix.specs.saaj-api-1.3-2.4.0.jar,
restarted karaf, re-tested and the error went away on AIX.  We are running
JDK 1.7.

Has anybody else run into this problem and why is that particular JAR and
the others included in the endorsed dir?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Class-com-sun-org-apache-xerces-internal-dom-ElementImpl-illegally-accessing-package-private-member-l-tp4034932.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: karaf 3.0.1 start error

2014-08-14 Thread asookazian2

I believe the machine needs to be online so the maven repos can be hit...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/karaf-3-0-1-start-error-tp4034645p4034781.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Data source xml change not taking effect in first karaf start post modification

2014-08-14 Thread asookazian2

This is an issue. We have users who will not know this unless it is
explicitly documented.

You should be able to modify the datasource-xxx.xml file when karaf is not
running and karaf should always check for any changes to the xml files in
deploy dir vs. the already deployed xml bundles. Then this wouldn't be a
problem. So a bug/enhancement request is in order I believe...

Achim Nierbeck wrote
No Issue, just not the right usage?
If Karaf is running, the change is detected.
It is even detected when the Karaf Instance is shut down.
So it works as designed, AFAIC.

regards, Achim

2014-08-11 11:25 GMT+02:00 Christian Schneider lt;

chris@

gt;:

I also think this is the issue.

Do you think this should be filed as a bug in felix fileinstall?

Christian

Am 11.08.2014 10:29, schrieb Achim Nierbeck:

I think the main problem seems to be, the xml file was altered while
karaf wasn't able to realize it's been altered.
After the startup it still used the old one as there has already been a
bundle generated for it.
My best guess right now, this is due to the fact that
karaf/FileInstaller
wasn't seeing the change it wasn't able to re-load the change,
make sure you alter it while Karaf is running then it will be able to
get
it faster.

regards, Achim

--
Christian Schneider
http://www.liquid-reality.de

Open Source Architect
Talend Application Integration Division http://www.talend.com

Apache Member
Apache Karaf lt;http://karaf.apache.org/gt; Committer PMC
OPS4J Pax Web lt;http://wiki.ops4j.org/display/paxweb/Pax+Web/gt;
Committer
Project Lead
blog lt;http://notizblog.nierbeck.de/gt;

Software Architect / Project Manager / Scrum Master

--
View this message in context:
http://karaf.922171.n3.nabble.com/Data-source-xml-change-not-taking-effect-in-first-karaf-start-post-modification-tp4034670p4034782.html
Sent from the Karaf - User mailing list archive at Nabble.com.

No appenders could be found for logger (log4j related)

We are seeing the following message when deploying a WAR bundle and starting
karaf 3.0.1:

admin@NextGate () log4j:WARN No appenders could be found for logger
(org.springframework.web.servlet.DispatcherServlet).
log4j:WARN Please initialize the log4j system properly.

I was told to remove log4j and use 'mvn dependency:tree' to determine how
the transitive dependency is happening:

i executed the following cmd: mvn dependency:tree -Dverbose -Dincludes=log4j
10:28 [INFO] --- maven-dependency-plugin:2.8:tree (default-cli) @
RelationManager-OSGi ---
[INFO]
com.nextgate.mm.relation.deploys:RelationManager-OSGi:war:1.0.0-SNAPSHOT
[INFO] +- org.slf4j:slf4j-log4j12:jar:1.7.2:compile
[INFO] | \- (log4j:log4j:jar:1.2.16:compile - version managed from 1.2.17;
omitted for duplicate)
[INFO] \- log4j:log4j:jar:1.2.16:compile

I tried to remove the log4j reference in the bundle-classpath for
maven-bundle-plugin config but starting seeing other exceptions
(ClassNotFoundExceptions for log4j classes).

This is what is relevant in the bundle-classpath for the maven-bundle-plugin
section in pom.xml for this WAR project. I have tried removing
log4j-1.2.16.jar and slf4j-api-1.7.2.jar and see various exceptions in
karaf.log.

Bundle-ClassPath.
...
,WEB-INF/lib/log4j-1.2.16.jar
,WEB-INF/lib/slf4j-api-1.7.2.jar
,WEB-INF/lib/slf4j-log4j12-1.7.2.jar
...
/Bundle-ClassPath

Now I have finally resorted to including a log4j.xml file in the root of the
WAR so that log4j will find it in the classpath. Now the WARN logging above
to the karaf console is gone but the logging for the file appender has a
different format than what is in the karaf.log.

Here is my current log4j.xml config. I was thinking about using
./data/log/karaf.log and appending to that log instead of a test.log.

appender name=FILE class=org.apache.log4j.RollingFileAppender

layout class=org.apache.log4j.PatternLayout

/layout
/appender

Am I missing something simple here (i.e. making this too complicated as a
solution)? Any assistance is appreciated. thx.

--
View this message in context:
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

This works:

   appender name=FILE class=org.apache.log4j.RollingFileAppender
 
 
 
 

 layout class=org.apache.log4j.PatternLayout
   
 /layout  
   /appender

This does *not* work (i.e. no logging of the same above contents in test.log
are visible in the karaf.log; why?)

   appender name=FILE class=org.apache.log4j.RollingFileAppender
 
 
 
 

 layout class=org.apache.log4j.PatternLayout
   
 /layout  
   /appender



--
View this message in context: 
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034742.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

Achim, thx for reply!  So I need to use Import-Package only instead of
Bundle-Classpath?

And if yes, which packages do I need to import specifically?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034743.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

Ok so I removed the three JARs (one log4j and two slf4j) from the
Bundle-ClassPath section and added the following to the package-import
section:

Import-Package
...
,org.apache.log4j
,org.slf4j
...
/Import-Package

Now I am not seeing the WARN in the console during karaf startup, but I am
seeing the following exception in the log.  Please advise how to resolve. 
Do I need to add log4j back to the bundle-classpath?  thx.

20140813 10:01:03.662 [WARN ] FelixStartLevel |
138:org.ops4j.pax.web.pax-web-extender-war |
org.ops4j.pax.web.extender.war.internal.Activator | Error while destroying
extension for bundle RelationManager-OSGi [242]
java.util.concurrent.ExecutionException: java.lang.NoClassDefFoundError:
org/apache/log4j/LogManager
at 
java.util.concurrent.FutureTask.report(FutureTask.java:122)[:1.7.0_51]
at java.util.concurrent.FutureTask.get(FutureTask.java:188)[:1.7.0_51]
at
org.ops4j.pax.web.extender.war.internal.extender.AbstractExtender.destroyExtension(AbstractExtender.java:309)[138:org.ops4j.pax.web.pax-web-extender-war:3.1.0]
at
org.ops4j.pax.web.extender.war.internal.extender.AbstractExtender.bundleChanged(AbstractExtender.java:188)[138:org.ops4j.pax.web.pax-web-extender-war:3.1.0]
at
org.apache.felix.framework.util.EventDispatcher.invokeBundleListenerCallback(EventDispatcher.java:868)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.util.EventDispatcher.fireEventImmediately(EventDispatcher.java:789)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.util.EventDispatcher.fireBundleEvent(EventDispatcher.java:514)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.Felix.fireBundleEvent(Felix.java:4403)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.Felix.stopBundle(Felix.java:2520)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.Felix.setActiveStartLevel(Felix.java:1309)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.FrameworkStartLevelImpl.run(FrameworkStartLevelImpl.java:304)[org.apache.felix.framework-4.2.1.jar:]
at java.lang.Thread.run(Thread.java:744)[:1.7.0_51]
Caused by: java.lang.NoClassDefFoundError: org/apache/log4j/LogManager
at
org.springframework.util.Log4jConfigurer.shutdownLogging(Log4jConfigurer.java:117)
at
org.springframework.web.util.Log4jWebConfigurer.shutdownLogging(Log4jWebConfigurer.java:170)
at
org.springframework.web.util.Log4jConfigListener.contextDestroyed(Log4jConfigListener.java:51)
at
org.eclipse.jetty.server.handler.ContextHandler.doStop(ContextHandler.java:823)
at
org.eclipse.jetty.servlet.ServletContextHandler.doStop(ServletContextHandler.java:160)
at
org.ops4j.pax.web.service.jetty.internal.HttpServiceContext.doStop(HttpServiceContext.java:229)
at
org.eclipse.jetty.util.component.AbstractLifeCycle.stop(AbstractLifeCycle.java:89)
at
org.ops4j.pax.web.service.jetty.internal.JettyServerImpl$1.stop(JettyServerImpl.java:202)
at
org.ops4j.pax.web.service.internal.HttpServiceStarted.begin(HttpServiceStarted.java:1018)
at
org.ops4j.pax.web.service.internal.HttpServiceProxy.begin(HttpServiceProxy.java:417)
at
org.ops4j.pax.web.extender.war.internal.UnregisterWebAppVisitorWC.visit(UnregisterWebAppVisitorWC.java:82)
at
org.ops4j.pax.web.extender.war.internal.model.WebApp.accept(WebApp.java:641)
at
org.ops4j.pax.web.extender.war.internal.WebAppPublisher$WebAppDependencyListener.unregister(WebAppPublisher.java:264)
at
org.ops4j.pax.web.extender.war.internal.WebAppPublisher$WebAppDependencyListener.removedService(WebAppPublisher.java:224)
at
org.ops4j.pax.web.extender.war.internal.WebAppPublisher$WebAppDependencyListener.removedService(WebAppPublisher.java:135)
at
org.osgi.util.tracker.ServiceTracker$Tracked.customizerRemoved(ServiceTracker.java:956)[karaf-org.osgi.core.jar:]
at
org.osgi.util.tracker.ServiceTracker$Tracked.customizerRemoved(ServiceTracker.java:864)[karaf-org.osgi.core.jar:]
at
org.osgi.util.tracker.AbstractTracked.untrack(AbstractTracked.java:341)[karaf-org.osgi.core.jar:]
at
org.osgi.util.tracker.ServiceTracker.close(ServiceTracker.java:375)[karaf-org.osgi.core.jar:]
at
org.ops4j.pax.web.extender.war.internal.WebAppPublisher.unpublish(WebAppPublisher.java:127)
at
org.ops4j.pax.web.extender.war.internal.WebObserver.undeploy(WebObserver.java:247)
at
org.ops4j.pax.web.extender.war.internal.WebObserver$1.doDestroy(WebObserver.java:185)
at
org.ops4j.pax.web.extender.war.internal.extender.SimpleExtension.destroy(SimpleExtension.java:70)
at
org.ops4j.pax.web.extender.war.internal.extender.AbstractExtender$2.run(AbstractExtender.java:288)
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)[:1.7.0_51]
at

Re: No appenders could be found for logger (log4j related)

So I added back the following to the Bundle-ClassPath section:

,WEB-INF/lib/log4j-1.2.16.jar

and still getting this exception on karaf startup.  I still have the import
for org.apache.log4j in the Import-Package section so not sure what else
to try at this point

Caused by: java.lang.ClassNotFoundException: org.apache.log4j.LogManager not
found by org.ops4j.pax.logging.pax-logging-api [7]
at
org.apache.felix.framework.BundleWiringImpl.findClassOrResourceByDelegation(BundleWiringImpl.java:1532)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleWiringImpl.access$400(BundleWiringImpl.java:75)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleWiringImpl$BundleClassLoader.loadClass(BundleWiringImpl.java:1955)[org.apache.felix.framework-4.2.1.jar:]
at java.lang.ClassLoader.loadClass(ClassLoader.java:358)[:1.7.0_51]
at
org.apache.felix.framework.BundleWiringImpl.getClassByDelegation(BundleWiringImpl.java:1374)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleWiringImpl.searchImports(BundleWiringImpl.java:1553)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleWiringImpl.findClassOrResourceByDelegation(BundleWiringImpl.java:1484)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleWiringImpl.access$400(BundleWiringImpl.java:75)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.BundleWiringImpl$BundleClassLoader.loadClass(BundleWiringImpl.java:1955)[org.apache.felix.framework-4.2.1.jar:]
at java.lang.ClassLoader.loadClass(ClassLoader.java:358)[:1.7.0_51]
... 36 more



--
View this message in context: 
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034747.html
Sent from the Karaf - User mailing list archive at Nabble.com.

delete datasource-foo.xml from deploy and re-copy

delete datasource-foo.xml from deploy and re-copy to deploy dir (with a
subsequent feature:install).

then do 'list' in karaf cmd line.  I do not see the bundle id in the list
until i halt and restart karaf 3.0.1.  is this intentional or a bug?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/delete-datasource-foo-xml-from-deploy-and-re-copy-tp4034749.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

Thanks for the tip. I tried this and it did not work.

Still seeing exception in the karaf.log:

Caused by: java.lang.ClassNotFoundException: org.apache.log4j.LogManager not
found by org.ops4j.pax.logging.pax-logging-api [7]

I will try to research the response that Achim gave.  thx guys.


bane73 wrote
 Not really the most correct answer, but try modifying your export section
 to export everything.
 That SHOULD work, I believe, because your project is using that JAR (thus,
 it's contained inside of it) but you are not telling OSGI that you want to
 use it and so it is blocking it from your JAR's classpath.  
 
 IOW, in your POM right below the 
 Bundle-ClassPath
  section you should have a 
 Export-Package
  section.  Delete everything and replace with a wildcard:
 
 ie: 
 Export-Package
 *
 /Export-Package
 That should do the trick.  If it does, don't settle for that though.  The
 correct answer is, I think, to install org.apache.log4j either as it's own
 bundle or as a part of a dependency-bundle.  I'm still pretty new to
 OSGI, though, so I could be wrong.





--
View this message in context: 
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034752.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

In the web.xml:

listener
   
listener-classorg.springframework.web.util.Log4jConfigListener/listener-class
/listener

Should we be using a different listener?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034755.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

Rugby, man that's tough! hope you're ok. ;)

http://grepcode.com/file/repo1.maven.org/maven2/org.springframework/spring-core/2.5.4/org/springframework/util/Log4jConfigurer.java

This class is using LogManager...

jbonofre wrote
Hi guys,

sorry I was out for a rugby game ;)

Pax Logging wraps log4j, it doesn't import the whole log4j api.
Some part of the log4j API is not included because all doesn't make
sense in OSGi.
It's the case for org.apache.log4j.LogManager: it's part of log4j but
it's not included in pax-logging (api or service).

In the latest pax-logging release, I included a couple of Log4j and
SLf4j new classes, but not this one as it's not really usefull in OSGi
(due to the Logging service).

So, what's your usage of LogManager ? I'm pretty sure you can avoid to
use it.

Regards
JB

On 08/13/2014 10:21 PM, asookazian2 wrote:
Thanks for the tip. I tried this and it did not work.

Still seeing exception in the karaf.log:

Caused by: java.lang.ClassNotFoundException: org.apache.log4j.LogManager
not
found by org.ops4j.pax.logging.pax-logging-api [7]

I will try to research the response that Achim gave. thx guys.

bane73 wrote
Not really the most correct answer, but try modifying your export
section
to export everything.
That SHOULD work, I believe, because your project is using that JAR
(thus,
it's contained inside of it) but you are not telling OSGI that you want
to
use it and so it is blocking it from your JAR's classpath.

IOW, in your POM right below the

Bundle-ClassPath
section you should have a

Export-Package
section. Delete everything and replace with a wildcard:

ie:

Export-Package
*

/Export-Package
That should do the trick. If it does, don't settle for that though.
The
correct answer is, I think, to install org.apache.log4j either as it's
own
bundle or as a part of a dependency-bundle. I'm still pretty new to
OSGI, though, so I could be wrong.

--
View this message in context:
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034752.html
Sent from the Karaf - User mailing list archive at Nabble.com.

--
Jean-Baptiste Onofré

jbonofre@

http://blog.nanthrax.net
Talend - http://www.talend.com

--
View this message in context:
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034756.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: No appenders could be found for logger (log4j related)

I commented out the below listener block in the web.xml.  Now it seems to be
behaving better (no exceptions) but I'm not sure what side-effect it may
have to remove that...


asookazian2 wrote
 In the web.xml:
 listener
 
 listener-class
 org.springframework.web.util.Log4jConfigListener
 /listener-class
 
 /listener
 Should we be using a different listener?





--
View this message in context: 
http://karaf.922171.n3.nabble.com/No-appenders-could-be-found-for-logger-log4j-related-tp4034739p4034757.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Unable to find the InitialContextFactory org.eclipse.jetty.jndi.InitialContextFactory

2014-08-11 Thread asookazian2

This is the fix I just made today which resolved the problem somehow.  The
problem is we'd like to better understand the following:

1) why was this fix needed?
2) is this the best way to fix the problem?
3) why/how is the problem fixed?

before (unable to login to WAR app): 

cxt = new InitialContext();

after (able to login to WAR app; exception does not reproduce):

ClassLoader cl = Thread.currentThread().getContextClassLoader();
   
Thread.currentThread().setContextClassLoader(this.getClass().getClassLoader());
try {
cxt = new InitialContext();
}
finally {
Thread.currentThread().setContextClassLoader(cl);
}



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Unable-to-find-the-InitialContextFactory-org-eclipse-jetty-jndi-InitialContextFactory-tp4034652p4034697.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Unable to find the InitialContextFactory org.eclipse.jetty.jndi.InitialContextFactory

2014-08-08 Thread asookazian2

Any explanation as to how to resolve this exception?  We can access the URL
home page for the WAR bundle fine until we deploy another EAR bundle at
which point we see the following exception:

I google'd a bit and threads are pointing to jndi.properties but I haven't
actually touched this file before and not sure where it should be located
and not sure what exact root cause is and how to resolve.  Any help
appreciated.

20140808 13:16:45.973 [ERROR] qtp500171771-47 |
249:com.nextgate.mm.PersonDQM |
com.nextgate.dqm.presentation.amfendpoint.DqmMessageBrokerFilter |
MessageException
flex.messaging.MessageException:
com.sun.mdm.index.master.ProcessingException :
com.sun.mdm.index.master.ProcessingException: MDM-MI-SRC501: Failed to read
MIDM configuration: Unable to find the InitialContextFactory
org.eclipse.jetty.jndi.InitialContextFactory.
at
flex.messaging.services.remoting.adapters.JavaAdapter.invoke(JavaAdapter.java:447)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.services.RemotingService.serviceMessage(RemotingService.java:183)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.MessageBroker.routeMessageToService(MessageBroker.java:1503)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.endpoints.AbstractEndpoint.serviceMessage(AbstractEndpoint.java:884)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
com.nextgate.dqm.presentation.amfendpoint.DqmAMFEndpoint.serviceMessage(DqmAMFEndpoint.java:92)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
com.nextgate.dqm.presentation.amfendpoint.DqmMessageBrokerFilter.invoke(DqmMessageBrokerFilter.java:169)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.endpoints.amf.LegacyFilter.invoke(LegacyFilter.java:158)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.endpoints.amf.SessionFilter.invoke(SessionFilter.java:44)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.endpoints.amf.BatchProcessFilter.invoke(BatchProcessFilter.java:67)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.endpoints.amf.SerializationFilter.invoke(SerializationFilter.java:146)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
flex.messaging.endpoints.BaseHTTPEndpoint.service(BaseHTTPEndpoint.java:278)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
org.springframework.flex.servlet.MessageBrokerHandlerAdapter.handle(MessageBrokerHandlerAdapter.java:101)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
org.springframework.web.servlet.DispatcherServlet.doDispatch(DispatcherServlet.java:875)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
org.springframework.web.servlet.DispatcherServlet.doService(DispatcherServlet.java:807)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
org.springframework.web.servlet.FrameworkServlet.processRequest(FrameworkServlet.java:571)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
org.springframework.web.servlet.FrameworkServlet.doPost(FrameworkServlet.java:511)[249:com.nextgate.mm.PersonDQM:1.0.0.SNAPSHOT]
at
javax.servlet.http.HttpServlet.service(HttpServlet.java:595)[64:org.apache.geronimo.specs.geronimo-servlet_3.0_spec:1.0]
at
javax.servlet.http.HttpServlet.service(HttpServlet.java:668)[64:org.apache.geronimo.specs.geronimo-servlet_3.0_spec:1.0]
at
org.eclipse.jetty.servlet.ServletHolder.handle(ServletHolder.java:684)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.eclipse.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1496)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.ops4j.pax.web.service.internal.WelcomeFilesFilter.doFilter(WelcomeFilesFilter.java:185)[78:org.ops4j.pax.web.pax-web-runtime:3.1.0]
at
org.eclipse.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1467)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.eclipse.jetty.servlet.ServletHandler.doHandle(ServletHandler.java:499)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.ops4j.pax.web.service.jetty.internal.HttpServiceServletHandler.doHandle(HttpServiceServletHandler.java:69)[79:org.ops4j.pax.web.pax-web-jetty:3.1.0]
at
org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:137)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.eclipse.jetty.security.SecurityHandler.handle(SecurityHandler.java:533)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.eclipse.jetty.server.session.SessionHandler.doHandle(SessionHandler.java:231)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at
org.eclipse.jetty.server.handler.ContextHandler.doHandle(ContextHandler.java:1086)[69:org.eclipse.jetty.aggregate.jetty-all-server:8.1.14.v20131031]
at

rolling upgrades to Karaf/Cellar

2014-07-31 Thread asookazian2

what is the best practice for rolling upgrades for a cluster of Karaf
servers?  I mean to say that we have 2 nodes in the cluster and we want to
update/upgrade software on server 1 but keep server 2 on the previous
version temporarily until everything is stable with the upgrade on server 1
then upgrade server 2. 

If we are using Cellar, the config, features and bundles will be always
replicated as long as both servers are running.

Also, what exactly is handling the failover for karaf instances in a
cluster?  Is it Karaf itself, cellar or hazelcast, or the software/hardware
load balancer?  thx.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/rolling-upgrades-to-Karaf-Cellar-tp4034531.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: rolling upgrades to Karaf/Cellar

2014-07-31 Thread asookazian2

Please add a link on this page to the Cellar clustering user guide:

http://karaf.apache.org/index/subprojects/cellar.html
http://karaf.apache.org/index/subprojects/cellar.html  



--
View this message in context: 
http://karaf.922171.n3.nabble.com/rolling-upgrades-to-Karaf-Cellar-tp4034531p4034532.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

2014-07-22 Thread asookazian2

Thanks for replying.  There is only one reference to configadmin in the
cellar docs:

http://karaf.apache.org/manual/cellar/latest/user-guide/index.html

How would I modify the etc/config.properties to be able to sync
etc/a/b/c/foo.properties?

Also, it would be nice if the Cellar docs was in one HTML page if possible.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034377.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

2014-07-22 Thread asookazian2

Where is the documentation on how the ConfigAdmin works???



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034378.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

I just tried to setup a brand new 2 node horz cluster on 2 windows VMs.  Both
nodes are joining the default group/cluster.

However, when I run our installer script with both karaf nodes (3.0.1)
running, the bundles install successfully on node A but not on node B. 
These are the actual bundles for our custom software, not the previous web
service example I had deployed.  

Here are the exceptions for node B.  Please note that we have
directories/files in the system directory (such as features XML which have
the bundle install info) that are not being sync'd with the node B.

This is how I installed cellar on both nodes:

karaf@root() feature:repo-add
mvn:org.apache.karaf.cellar/apache-karaf-cellar/3.0.0/xml/features
karaf@root() feature:install cellar

20140721 12:30:56.860 [INFO ]
hz._hzInstance_1_cellar.global-operation.thread-1 | 241:com.hazelcast |
com.hazelcast.cluster.ClusterService | [192.168.172.219]:5701 [cellar]
[3.2.3]

Members [2] {
Member [192.a.b.c]:5701
Member [192.a.b.d]:5701 this
}

20140721 12:30:58.389 [INFO ] hz._hzInstance_1_cellar.cached.thread-18 |
241:com.hazelcast | com.hazelcast.core.LifecycleService |
[192.168.172.219]:5701 [cellar] [3.2.3] Address[192.168.172.219]:5701 is
MERGED
20140721 12:39:57.560 [ERROR] pool-14-thread-71 |
248:org.apache.karaf.cellar.features |
org.apache.karaf.cellar.features.RepositoryEventHandler | CELLAR FEATURES:
failed to add/remove repository URL
mvn:com.nextgate.am/ngam-features/9.0.0-SNAPSHOT/xml/features
java.lang.NullPointerException
at
org.apache.karaf.cellar.features.RepositoryEventHandler.handle(RepositoryEventHandler.java:80)[248:org.apache.karaf.cellar.features:3.0.0]
at
org.apache.karaf.cellar.features.RepositoryEventHandler.handle(RepositoryEventHandler.java:29)[248:org.apache.karaf.cellar.features:3.0.0]
at Proxy816e4134_52be_45a8_9a8d_f2b4b7585c2a.handle(Unknown
Source)[:]
at Proxyd66f100e_e050_4991_b1e8_898982523878.handle(Unknown
Source)[:]
at
org.apache.karaf.cellar.core.event.EventDispatchTask.run(EventDispatchTask.java:57)[242:org.apache.karaf.cellar.core:3.0.0]
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)[:1.7.0_60]
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)[:1.7.0_60]
at java.lang.Thread.run(Thread.java:745)[:1.7.0_60]
20140721 12:39:58.496 [WARN ] pool-14-thread-71 | 5:org.ops4j.pax.url.mvn |
org.ops4j.pax.url.maven.commons.MavenRepositoryURL | Repository spec
https://repository.apache.org/content/repositories/orgapachekaraf-1006/ does
not contain an identifier. This is deprecated  discouraged  just evil.
20140721 12:39:59.729 [WARN ] pool-14-thread-71 | 5:org.ops4j.pax.url.mvn |
org.ops4j.pax.url.mvn.internal.AetherBasedResolver | Error resolving
artifactcom.nextgate.am:ngam-webservice:jar:9.0.0-SNAPSHOT:Could not find
artifact com.nextgate.am:ngam-webservice:jar:9.0.0-SNAPSHOT in defaultlocal
(file:/C:/sw/ngs/current/ngs/data/repo/)
org.sonatype.aether.resolution.ArtifactResolutionException: Could not find
artifact com.nextgate.am:ngam-webservice:jar:9.0.0-SNAPSHOT in defaultlocal
(file:/C:/sw/ngs/current/ngs/data/repo/)
at
org.sonatype.aether.impl.internal.DefaultArtifactResolver.resolve(DefaultArtifactResolver.java:538)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.sonatype.aether.impl.internal.DefaultArtifactResolver.resolveArtifacts(DefaultArtifactResolver.java:216)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.sonatype.aether.impl.internal.DefaultArtifactResolver.resolveArtifact(DefaultArtifactResolver.java:193)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.sonatype.aether.impl.internal.DefaultRepositorySystem.resolveArtifact(DefaultRepositorySystem.java:286)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.ops4j.pax.url.mvn.internal.AetherBasedResolver.resolve(AetherBasedResolver.java:250)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.ops4j.pax.url.mvn.internal.AetherBasedResolver.resolveFile(AetherBasedResolver.java:239)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.ops4j.pax.url.mvn.internal.AetherBasedResolver.resolve(AetherBasedResolver.java:223)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.ops4j.pax.url.mvn.internal.Connection.getInputStream(Connection.java:122)[5:org.ops4j.pax.url.mvn:1.6.0]
at
org.apache.felix.framework.util.SecureAction.getURLConnectionInputStream(SecureAction.java:524)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.cache.JarRevision.initialize(JarRevision.java:165)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.cache.JarRevision.init(JarRevision.java:77)[org.apache.felix.framework-4.2.1.jar:]
at
org.apache.felix.framework.cache.BundleArchive.createRevisionFromLocation(BundleArchive.java:878)[org.apache.felix.framework-4.2.1.jar:]
at

Re: Cellar for karaf 3.0.1 active/active and failover clustering

No, they are using data/repo.

http://karaf.apache.org/index/subprojects/cave/download/karaf-cave-3.0.0-release.html

Do I need to use Karaf Cave?  We probably cannot setup/install/config
Artifactory at the client sites.  Please advise link to the Cave docs.  thx.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034349.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

Populate from an external repository
If you have a bunch of artifacts to upload, it’s not very efficient to use
the cave:upload-artifact command.
The cave:populate-repository command allows you to upload a set of artifacts
from an “external” repository:

karaf@root cave:populate-repository cave-repo
file:/home/jbonofre/.m2/repository
In this example, Cave will browse the file:/home/jbonofre/.m2/repository
location, looking for OSGi bundles, and will copy the artifacts in your Cave
Repository storage location.

http://blog.nanthrax.net/2011/08/apache-karaf-cave-preview/
http://blog.nanthrax.net/2011/08/apache-karaf-cave-preview/  

Do I need to install Cave on all the nodes?  and then do I need to do the
above (cave:populate-repository) for both nodes or just one?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034350.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

On Windows, I can't write this cmd properly for it to work:

cave:repository-create cave-repo -l C:/sw/ngs/current/ngs/data/repo data
Error executing command cave:repository-create: too many arguments specified

Example from blog: 

cave:create-repository -l /home/jbonofre/.m2/repository m2



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034352.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

I don't believe so, here is our config:

org.ops4j.pax.url.mvn.repositories =
http://repo1.maven.org/maven2@id=central,
http://repository.springsource.com/maven/bundles/release@id=spring.ebr.release,
http://repository.springsource.com/maven/bundles/external@id=spring.ebr.external,
file:C:\\sw\\ngs\\current\\ngs/system@id=system.repository,
file:C:\\sw\\ngs\\current\\ngs\\data/kar@id=kar.repository@multi,
https://repository.apache.org/content/repositories/orgapachekaraf-1006/

I was able to replicate bundle installs with the web service POC project I
was working on recently.  Round robin from httpd and failover worked as well
prior.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034353.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

Is it possible to replicate data/repo?  We are considering running
feature:repo-add and install bundles on each node which defeats the purpose
of Cellar which replicates bundles, features and configs...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034354.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

If I make a change to a file in /etc/a/b/c dir, will it replicate to the same
dir structure in the other nodes?  I just tried that with both nodes running
and it doesn't seem to do anything...



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034355.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: Cellar for karaf 3.0.1 active/active and failover clustering

In fact I made a change to etc/org.ops4j.pax.url.mvn.cfg and it did not
replicate either (only a comment I tried).



--
View this message in context: 
http://karaf.922171.n3.nabble.com/Cellar-for-karaf-3-0-1-active-active-and-failover-clustering-tp4033204p4034356.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: bundle in lifecycle infinite loop

2014-07-11 Thread asookazian2

The blacklisting is for the following datasource.xml:

blueprint:file:/C:/sw/ngs/deploy/datasource-person-ws.xml

I'm blacklisting on both nodes (etc/org.apache.karaf.cellar.groups.cfg):

default.bundle.whitelist.inbound = *
default.bundle.whitelist.outbound = *
default.bundle.blacklist.inbound =
blueprint:file:/C:/sw/ngs/deploy/datasource-person-ws.xml
default.bundle.blacklist.outbound =
blueprint:file:/C:/sw/ngs/deploy/datasource-person-ws.xml
default.bundle.sync = true



--
View this message in context: 
http://karaf.922171.n3.nabble.com/bundle-in-lifecycle-infinite-loop-tp4033960p4034174.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: bundle in lifecycle infinite loop

2014-07-11 Thread asookazian2

We are trying to achieve active/active for a web service (for example) as
well as failover...

I have tested this and it seems to work so far with cellar 3.0.0 and karaf
3.0.1



--
View this message in context: 
http://karaf.922171.n3.nabble.com/bundle-in-lifecycle-infinite-loop-tp4033960p4034175.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: writing a bug for Karaf/Cellar

2014-07-10 Thread asookazian2

thanks this seemed to work.

Cellar docs need lots more tutorials and/or examples.

Good one is how to setup a 2 node horz cluster using
mod_proxy_balancer/httpd.  thx.



--
View this message in context: 
http://karaf.922171.n3.nabble.com/writing-a-bug-for-Karaf-Cellar-tp4034078p4034145.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: bundle in lifecycle infinite loop

2014-07-10 Thread asookazian2

this seems to be a bug but the workaround JB suggested to blacklist the
bundle for the datasource xml (use la -l to find the location and then
update bundle blacklist inbound/outbound in
etc/org.apache.karaf.cellar.groups.cfg)



--
View this message in context: 
http://karaf.922171.n3.nabble.com/bundle-in-lifecycle-infinite-loop-tp4033960p4034146.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: writing a bug for Karaf/Cellar

How exactly do I exclude the datasource xml bundle?

If you copy the xml file to the deploy dir, Karaf creates a bundle for it
with a corresponding bundle.jar.

Are you saying I need to add it to the
etc/org.apache.karaf.cellar.groups.cfg file like this:

default.config.blacklist.inbound = org.apache.felix.fileinstall*, \
   org.apache.karaf.cellar*, \
   org.apache.karaf.management, \
   org.apache.karaf.shell, \
   org.ops4j.pax.logging, \
   org.ops4j.pax.web, \
   bundle.jar
default.config.blacklist.outbound = org.apache.felix.fileinstall*, \
org.apache.karaf.cellar*, \
org.apache.karaf.management, \
org.apache.karaf.shell, \
org.ops4j.pax.logging, \
org.ops4j.pax.web, \
bundle.jar

What exactly would the syntax be for referencing the jar or datasource xml
here??



--
View this message in context: 
http://karaf.922171.n3.nabble.com/writing-a-bug-for-Karaf-Cellar-tp4034078p4034107.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: writing a bug for Karaf/Cellar

sorry i meant this block same file:

default.bundle.whitelist.inbound = *
default.bundle.whitelist.outbound = *
default.bundle.blacklist.inbound = none
default.bundle.blacklist.outbound = none
default.bundle.sync = true



--
View this message in context: 
http://karaf.922171.n3.nabble.com/writing-a-bug-for-Karaf-Cellar-tp4034078p4034108.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: writing a bug for Karaf/Cellar

is it possible to blacklist a particular bundle id or range of bundle id's?



--
View this message in context: 
http://karaf.922171.n3.nabble.com/writing-a-bug-for-Karaf-Cellar-tp4034078p4034109.html
Sent from the Karaf - User mailing list archive at Nabble.com.

Re: writing a bug for Karaf/Cellar