headers
jira | 2 Nov 2008 07:21
Picon
Favicon

Subscription: open issues

Issue Subscription
Filter: open issues (279 issues)
Open Issues for Apache Jackrabbit
Subscriber: jackrabbitdev

Key         Summary
JCR-1845    clarify QValueFactory.create(File)
            https://issues.apache.org/jira/browse/JCR-1845
JCR-1844    Convenience method to Or multiple values with a single filter
            https://issues.apache.org/jira/browse/JCR-1844
JCR-1842    SPI: RepositoryService.obtain should allow to pass null workspaceName indicating the default workspace
            https://issues.apache.org/jira/browse/JCR-1842
JCR-1839    Introduce Event.getDate()
            https://issues.apache.org/jira/browse/JCR-1839
JCR-1838    Garbage collection deletes temporary files in FileDataStore
            https://issues.apache.org/jira/browse/JCR-1838
JCR-1836    Persistence: support property databaseType
            https://issues.apache.org/jira/browse/JCR-1836
JCR-1834    Create RepositoryFactory implementation
            https://issues.apache.org/jira/browse/JCR-1834
JCR-1827    SessionImpl should have  <at> PreDestroy annotation a method to shut down cleanly from within container
            https://issues.apache.org/jira/browse/JCR-1827
JCR-1823    Repository.login throws IllegalStateException
            https://issues.apache.org/jira/browse/JCR-1823
JCR-1822    Patches for previously incompleted patches
            https://issues.apache.org/jira/browse/JCR-1822
JCR-1818    Too many open files when merging large index segments
            https://issues.apache.org/jira/browse/JCR-1818
JCR-1810    Jcr-Test: SerializationTest logs exception during tearDown.
            https://issues.apache.org/jira/browse/JCR-1810
(Continue reading)

Permalink | Reply | 0 comments |
headers
Trang Nguyen (JIRA | 2 Nov 2008 22:26
Picon
Favicon

Updated: (JCR-1846) Jackrabbit stuck thread issue due to fat lock


     [
https://issues.apache.org/jira/browse/JCR-1846?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Trang Nguyen updated JCR-1846:
------------------------------

    Summary: Jackrabbit stuck thread issue due to fat lock  (was: Jackrabbit stuck thread issue to to fat lock)

> Jackrabbit stuck thread issue due to fat lock
> ---------------------------------------------
>
>                 Key: JCR-1846
>                 URL: https://issues.apache.org/jira/browse/JCR-1846
>             Project: Jackrabbit
>          Issue Type: Bug
>          Components: jackrabbit-core
>    Affects Versions: core 1.4.5
>         Environment: jrockit-R27.5.0-jdk1.5.0_14
> weblogic 9.2.3
> spring 2.5.5
> oracle-jdbc-10.2.0.2.0
>            Reporter: Trang Nguyen
>
> Hello,
> We are running jackrabbit 1.4.5 using a persistent file data store within a weblogic container and
encountering a variety of thread locking issues. To get around the problem, we are forced synchronize
thread access to the JCR repository or reduce thread worker count to 1 which has a heavy performance impact
on our application. I'm not exactly sure what the problem is and was wondering if there is a workaround or
fix planned for this?
(Continue reading)

Permalink | Reply | 0 comments |
headers
Trang Nguyen (JIRA | 2 Nov 2008 22:26
Picon
Favicon

Created: (JCR-1846) Jackrabbit stuck thread issue to to fat lock

Jackrabbit stuck thread issue to to fat lock
--------------------------------------------

                 Key: JCR-1846
                 URL: https://issues.apache.org/jira/browse/JCR-1846
             Project: Jackrabbit
          Issue Type: Bug
          Components: jackrabbit-core
    Affects Versions: core 1.4.5
         Environment: jrockit-R27.5.0-jdk1.5.0_14
weblogic 9.2.3
spring 2.5.5
oracle-jdbc-10.2.0.2.0
            Reporter: Trang Nguyen

Hello,

We are running jackrabbit 1.4.5 using a persistent file data store within a weblogic container and
encountering a variety of thread locking issues. To get around the problem, we are forced synchronize
thread access to the JCR repository or reduce thread worker count to 1 which has a heavy performance impact
on our application. I'm not exactly sure what the problem is and was wondering if there is a workaround or
fix planned for this?

<Oct 30, 2008 10:45:32 AM PDT> <Error> <WebLogicServer> <BEA-000337> <[STUCK] ExecuteThread: '5' for
queue: 'weblogic.kernel.Default (self-tuning)' has been busy for "1,863" seconds working on the
request "weblogic.work.j2ee.J2EEWorkManager$WorkWithListener <at> 2117cc9", which is more than the
configured time (StuckThreadMaxTime) of "1,800" seconds. Stack trace:
Thread-94 "[STUCK] ExecuteThread: '5' for queue: 'weblogic.kernel.Default (self-tuning)'" <alive,
in native, suspended, waiting, priority=1, DAEMON> {
    -- Waiting for notification on:
(Continue reading)

Permalink | Reply | 17 comments |
headers
Trang Nguyen (JIRA | 2 Nov 2008 22:28
Picon
Favicon

Updated: (JCR-1846) Jackrabbit thread contention issue due to fat lock


     [
https://issues.apache.org/jira/browse/JCR-1846?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Trang Nguyen updated JCR-1846:
------------------------------

    Description: 
Hello,

We are running jackrabbit 1.4.5 using a persistent file data store within a weblogic container and
encountering a variety of thread locking issues. To get around the problem, we are forced synchronize
thread access to the JCR repository or reduce thread worker count to 1 which has a heavy performance impact
on our application. I'm not exactly sure what the problem is and was wondering someone is looking into this
issue and if there is a workaround/fix planned?

<Oct 30, 2008 10:45:32 AM PDT> <Error> <WebLogicServer> <BEA-000337> <[STUCK] ExecuteThread: '5' for
queue: 'weblogic.kernel.Default (self-tuning)' has been busy for "1,863" seconds working on the
request "weblogic.work.j2ee.J2EEWorkManager$WorkWithListener <at> 2117cc9", which is more than the
configured time (StuckThreadMaxTime) of "1,800" seconds. Stack trace:
Thread-94 "[STUCK] ExecuteThread: '5' for queue: 'weblogic.kernel.Default (self-tuning)'" <alive,
in native, suspended, waiting, priority=1, DAEMON> {
    -- Waiting for notification on:
EDU.oswego.cs.dl.util.concurrent.WriterPreferenceReadWriteLock$WriterLock <at> 152c384[fat lock]
    java.lang.Object.wait(Object.java:???)
    java.lang.Object.wait(Object.java:474)
    EDU.oswego.cs.dl.util.concurrent.WriterPreferenceReadWriteLock$WriterLock.acquire(Unknown Source)
    org.apache.jackrabbit.core.journal.AbstractJournal.lockAndSync(AbstractJournal.java:235)
    org.apache.jackrabbit.core.journal.DefaultRecordProducer.append(DefaultRecordProducer.java:49)
(Continue reading)

Permalink | Reply | 0 comments |
headers
iapilgrim | 3 Nov 2008 04:37
Picon

Get Indentifier of a Node


Hi,
In JSR 283 spec, we can get a node  by the API Node.getIndentifer(). I don't
know which JR version implement this feature.
I'm using JR 1.4.6, and I see it partly implement JSR 283 such as new query.
If we can use Node.getIndentifer(), it will be a good new for us.
I've just downloaded JR from trunk but can't find any test case for
Node.getIndentifier.
Thanks and Regards,
Van
Permalink | Reply | 3 comments |
headers
Philip Cher | 3 Nov 2008 04:41
Favicon

WebDAV in Atlassian Confluence

Hi Everyone,

I am working on a WebDAV project that is utilizing Jackrabbit as the
front end of WebDAV that maps back to Atlassian Confluence hierarchy as
a repository. We're able to replace the default Jackrabbit repository
with Confluence as a repository and able to browse Confluence hierarchy
and write to Confluence with WebDAV on Windows XP, Vista and Mac. The
problem I am facing now is that I couldn't edit text file (double click
on text file) on Windows XP and Vista.

 From my findings I notice that mounting as Shared Folder/Network Drive,
editing text files is possible and when mounting as Web Folder, text
files are not editable. Below is the way how I mount and compare between
Apache WebDAV Server and Confluence.

First I tried to mount Apache WebDAV Server (without Basic
Authentication), they are below:-

    *Shared Folder/Network Drive*
    1. Start > My Computer > Tools > Map Network Drive
    2. In the 'type in folder' type in  '\\localhost\webdav' or
    'http://localhost/webdav' (Windows will convert this path to
    \\localhost\webdav)
    3. Click Next and a link will be created in 'My Network Places'

    *Web Folder*
    1. Start > My Network Places
    2. Add a network place
    3. In the 'Internet or network address' type in the address
    'http://localhost:8080/webdav' (Windows will not convert this path
(Continue reading)

Permalink | Reply | 3 comments |
headers
Julian Reschke | 3 Nov 2008 09:33
Picon
Picon

Re: WebDAV in Atlassian Confluence

Philip Cher wrote:
> ...
> Is there anything wrong somewhere that might cause the text file not
> editable? And is there anything else for me to do to make text files
> editable besides mounting differently as I described above?
> ...

That's a Windows issue, not a Jackrabbit issue. It applies to *any* 
WebDAV server.

The answer: the MS Webfolder client is just an explorer extension; you 
can browse folders, delete files, copy and move files around, that's it. 
Everything else beyond that depends on the Windows application 
supporting WebDAV directly -- for instance, MS Office does, most others 
don't.

BR, Julian
Permalink | Reply | 2 comments |
headers
Sébastien Launay | 3 Nov 2008 09:42
Favicon

Re: Get Indentifier of a Node

Hi,

Node.getIdentifier() in JSR 283 is just the new method name for
Node.getUUID() (which has been deprecated).
But, from what i understand you want to retrieve a node from his identifier
so you need to use Session.getNodeByUUID(String) (which is also deprecated
and replaced with Session.getNodeByIdentifier(String) in JSR 283).
Because these modifications are only method renaming, you already have this
feature in JCR 1.0.

Regards,

--
Sébastien Launay

iapilgrim a écrit :
> Hi,
> In JSR 283 spec, we can get a node  by the API Node.getIndentifer(). I don't
> know which JR version implement this feature.
> I'm using JR 1.4.6, and I see it partly implement JSR 283 such as new query.
> If we can use Node.getIndentifer(), it will be a good new for us.
> I've just downloaded JR from trunk but can't find any test case for
> Node.getIndentifier.
> Thanks and Regards,
> Van
>
Permalink | Reply | 2 comments |
headers
Philip Cher | 3 Nov 2008 10:06
Favicon

Re: WebDAV in Atlassian Confluence

Hi Julian,

Thanks alot! That helps clear some doubts that I had on this issue =)

Regards,
Philip

Julian Reschke wrote:
> Philip Cher wrote:
>
>> ...
>> Is there anything wrong somewhere that might cause the text file not
>> editable? And is there anything else for me to do to make text files
>> editable besides mounting differently as I described above?
>> ...
>>
>
> That's a Windows issue, not a Jackrabbit issue. It applies to *any*
> WebDAV server.
>
> The answer: the MS Webfolder client is just an explorer extension; you
> can browse folders, delete files, copy and move files around, that's it.
> Everything else beyond that depends on the Windows application
> supporting WebDAV directly -- for instance, MS Office does, most others
> don't.
>
> BR, Julian
> --
> Message  protected by MailGuard: e-mail anti-virus, anti-spam and content filtering.
> http://www.mailguard.com.au
(Continue reading)

Permalink | Reply | 1 comment |
headers
iapilgrim | 3 Nov 2008 11:16
Picon

Re: Get Indentifier of a Node


Sébastien Launay wrote:
> 
> Hi,
> 
> Node.getIdentifier() in JSR 283 is just the new method name for
> Node.getUUID() (which has been deprecated).
> But, from what i understand you want to retrieve a node from his
> identifier
> so you need to use Session.getNodeByUUID(String) (which is also deprecated
> and replaced with Session.getNodeByIdentifier(String) in JSR 283).
> Because these modifications are only method renaming, you already have
> this
> feature in JCR 1.0.
> 
> Regards,
> 
> --
> Sébastien Launay
> 
>  
> 
> 
Thanks Launay,
I want to use Session.getNodeByIndentifier.
As I know so far, when we use jcr 1.0, to get a Node, we can use
session.getItem or session.getNodeByUUID. That means an identifier of a node
can be path ( plus name) or its uuid ( if node is referenceable).
So what is the identifier of a node in JR?  Is it still a path ( and name)?
What are the new features of identifier in jcr 2.0 over one in jcr 1.0?
(Continue reading)

Permalink | Reply | 1 comment |

Gmane