/metric/StatisticsProviderFactory.java:[59,54]
error: illegal forward reference
Chetan Mehrotra
On Mon, Nov 23, 2015 at 5:31 PM, Travis CI wrote:
> Build Update for apache/jackrabbit-oak
> -
>
> Build: #6910
> Status: Broken
>
> Duration
. By default only TimeSeries based
data would be collected.
Kindly have a look at OAK-3654 and provide your inputs there!
Chetan Mehrotra
[1] http://metrics.dropwizard.io/
[2] http://metrics.dropwizard.io/3.1.0/getting-started/
Have a look at webapp example [1] for suggested setup. The repository
should be created once and then reused.
Chetan Mehrotra
[1] https://github.com/apache/jackrabbit-oak/tree/trunk/oak-examples/webapp
On Wed, Nov 18, 2015 at 4:02 AM, David Marginian
wrote:
> https://jackrabbit.apache.org/
Probably these are intermittent failures.
However bit unsure for B) - As there exception comes just in the
OakInitializer.initialize call which should be single threaded. So not
sure why there was a conflict
Chetan Mehrotra
[1] https://paste.apache.org/ngsL
On Tue, Nov 17, 2015 at 10:42 AM, Chetan Mehrot
Looks like running check-release.sh for 1.2.8 build (or 1.2 branch)
is not triggering testcase. It looks like Marcel raised this earlier
also at [1]. Would followup there
Chetan Mehrotra
[1] http://markmail.org/thread/f5uvng2b2yvqu7ix
On Tue, Nov 17, 2015 at 12:49 AM, Julian Reschke wrote
Was this change done? I just ran the check-release script for 1.2.8
and it did not executed any test
Chetan Mehrotra
On Mon, Sep 7, 2015 at 6:31 PM, Davide Giannella wrote:
> On 07/09/2015 11:08, Marcel Reutegger wrote:
>> ...
>> I would like to revert this change and run test
[X] +1 Release this package as Apache Jackrabbit Oak 1.0.24
Chetan Mehrotra
On Mon, Nov 9, 2015 at 6:25 AM, Jason Harrop wrote:
> My objective is to add a binary file via http. Various things I tried:
...
> 3. webdav put: creates node structure, but binary content seems to be awol
Can you elaborate on this? For adding binary files webdav should work fine.
Hi Davide,
I am working on OAK-3092 and hope to get it done by today. So can the
release be delayed a bit to include this change?
Chetan Mehrotra
On Thu, Nov 5, 2015 at 9:37 PM, Davide Giannella wrote:
> Hello team,
>
> I'm planning to cut Oak 1.3.10 next Monday morning GMT.
&g
for consumption
by oak-lucene
Chetan Mehrotra
be aware of drivers etc. So we can get rid of
that
@Julian - Thoughts?
Chetan Mehrotra
[1]
https://github.com/apache/jackrabbit-oak/commit/7f844d5cde52dc53c41cc01aad9079afb275438a
On Mon, Oct 26, 2015 at 4:20 PM, Francesco Mari
wrote:
> A friendly reminder of this issue. Is there a speci
if
we should remove it from implementation side
Also it would be good to have a testcase to validate the above logic
if we are removing dependency from a tested Guava utility method
Chetan Mehrotra
s not
get started otherwise it would trigger large traversal if no index was
present or previous revision of index is not usable (due to some
corruption)
Chetan Mehrotra
On Wed, Oct 21, 2015 at 2:24 PM, Thomas Mueller wrote:
> Hi,
>
> If an index provider is (temporarily) not available, th
rue(Assert.java:43)
at org.junit.Assert.assertNull(Assert.java:551)
at org.junit.Assert.assertNull(Assert.java:562)
at
org.apache.jackrabbit.oak.plugins.segment.CompactionMapTest.removeSome(CompactionMapTest.java:156)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
====
Ch
On Thu, Sep 24, 2015 at 1:56 PM, Thomas Mueller wrote:
> what about getIndexCostInfo
+1
Chetan Mehrotra
Hi Thomas,
On Wed, Sep 23, 2015 at 6:51 PM, wrote:
> /**
> + * Get the index cost. The query must already be prepared.
> + *
> + * @return the index cost
> + */
> +String getIndexCost();
Should this be returning string? May be we should name it better
Chetan Mehrotra
Hi Marcel,
A short description of what was the fix would be very helpful for
future reference!
Chetan Mehrotra
On Tue, Sep 22, 2015 at 9:00 PM, wrote:
> Author: mreutegg
> Date: Tue Sep 22 15:30:08 2015
> New Revision: 1704655
>
> URL: http://svn.apache.org/viewvc?rev=1704655
Hi Julian,
LucenePropertyIndex does support contains for properties, For an
example see LucenePropertyIndexTest#fulltextBooleanComplexOrQueries.
You can give your usecase a try
Chetan Mehrotra
On Fri, Sep 18, 2015 at 5:33 PM, Julian Sedding wrote:
> Hi there
>
> Digging through the
Opened OAK-3412 to track this issue.
Chetan Mehrotra
On Tue, Sep 15, 2015 at 3:24 PM, Julian Reschke wrote:
> On 2015-09-15 10:03, Chetan Mehrotra wrote:
>>
>> If I change the code in Namespace#isValidLocalName to use
>> Character#isWhitespace then NameValidator sta
?
Chetan Mehrotra
On Mon, Sep 14, 2015 at 4:13 PM, Chetan Mehrotra
wrote:
> Micheal Durig mentioned that this has been discussed earlier. So
> going back in time this was discussed in OAK-1891 and thread [1]. Oak
> used to prevent such nodes from getting created earlier but that logic
>
(see Namespace#isValidLocalName).
However when that change was done the check initially used
Character#isWhitespace and switched to using Character#isSpaceChar
which is limited to very few checks. Now looks like isSpaceChar is
returning false for '\n', '\r' etc not sure why.
cause problem other. Specially when JR2 does not allow creation
of such node names [1]
So the question is
Should Oak allow node names with non space whitespace chars like \t, \r etc
Chetan Mehrotra
[1]
https://github.com/apache/jackrabbit/blob/trunk/jackrabbit-spi-commons/src/main/java/org/apache
How did you performed the test? If you tested out with explain then
current code in 1.0/1.2 return misleading result and this got fixed in
trunk with OAK-2943. Technically Oak would convert the OR clause to a
union query and then each part of union should then be able to make
use of index.
Chetan
Wrapper. So would be better to move this logic there.
Or you want to customize the analyzer for that field only during
indexing.
Chetan Mehrotra
+1
Chetan Mehrotra
On Tue, Sep 1, 2015 at 1:10 PM, Davide Giannella wrote:
> Hello team,
>
> some of us noticed we lack the epics in our jira so I raised an issue
> asking whether that would be possible to have them.
>
> https://issues.apache.org/jira/browse/INFRA-10185
>
&
Welcome Francesco!
Chetan Mehrotra
On Fri, Aug 28, 2015 at 5:18 PM, Michael Dürig wrote:
> Hi,
>
> Please welcome Francesco as a new committer and PMC member of the Apache
> Jackrabbit project. The Jackrabbit PMC recently decided to offer Francesco
> committership based on hi
can think of some offheap solution. You
can also look into using MVStore which is being used in
DocumentNodeStore for persistent cache.
Chetan Mehrotra
[1]
https://issues.apache.org/jira/browse/OAK-2882?focusedCommentId=14550198&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabp
is not on svn but git as the git
mirrors might lag behind!
Chetan Mehrotra
[1]
https://github.com/apache/jackrabbit-oak/commit/63c4a4db95b0f39f9ab27f499416c99c918a4955
On Wed, Aug 19, 2015 at 9:35 PM, Apache Jenkins Server
wrote:
> The Apache Jenkins build system has built Apache Jackrabb
d repo. Fixed
that now with rev http://svn.apache.org/r1696522. Hopefully this
should fix the issue!
Chetan Mehrotra
===
Do share your feedback around above proposed approach. In particular
following aspect
Q - Should we make oak-pojosr based setup as one of the
recommended/supported approach for configuring Oak in non OSGi env
Chetan Mehrotra
[1] http://jackrabbit.apache.org/oak/docs/construct.html
[2] https://gi
.assertFailed(InvokerHelper.java:398)
at
org.codehaus.groovy.runtime.ScriptBytecodeAdapter.assertFailed(ScriptBytecodeAdapter.java:646)
at
org.apache.jackrabbit.oak.run.osgi.MBeanIntegrationTest.jmxIntegration(MBeanIntegrationTest.groovy:47)
--
Chetan Mehrotra
On Sat, Aug 15, 2015 at 10:39 AM, Apache Jenkins Server
wrote:
>
Now the compile passes but 55 test fail. Majority of them from Solr.
Opened OAK-3215 to track that
Chetan Mehrotra
On Wed, Aug 12, 2015 at 10:38 AM, Apache Jenkins Server
wrote:
> The Apache Jenkins build system has built Apache Jackrabbit Oak matrix (build
> #325)
>
> Status: S
the following articles:
[ERROR] [Help 1]
http://cwiki.apache.org/confluence/display/MAVEN/DependencyResolutionException
[ERROR]
[ERROR] After correcting the problems, you can resume the build with the command
[ERROR] mvn -rf :oak-blob-cloud
Chetan Mehrotra
On Wed, Aug 12, 2015 at 9:50 AM, Apa
Looks like some issue with my recent work. Would have a look.
Thanks for filing the issue and marking current one as ignored!
Chetan Mehrotra
On Mon, Aug 10, 2015 at 2:02 PM, Angela Schreiber wrote:
> hi
>
> i get the following failures in oak-pojosr. is it only me?
> anybody wor
uld be fine. So should the final location be /oak-example/webapp ?
I can then move the current copy to that place. I have some local
commit now in my git-svn. Once I am done I would commit then to
current place and then move them to final place.
Would that be OK?
Chetan Mehrotra
sure why it did not failed.
Chetan Mehrotra
On Thu, Aug 6, 2015 at 3:15 PM, Michael Dürig wrote:
>
> Hi Chetan,
>
> While I agree that this is a good show case I don't like the copy paste
> approach. While this is simple and cheap now, we keep underestimating the
> hidden c
. Then go to oak-doc/target/scmpublish-checkout
4. and directly commit the changed html files via svn commit
This is also documented under oak-doc/README.md [1] and has proved to
be a faster approach for me compared to building and deploying the
whole site
Chetan Mehrotra
[1] https://github.com
figures the webdav servlet and JCR Remoting which work with any
repository implementation and thus should work with Oak.
Chetan Mehrotra
to with most basic setup.
I would like to modify this module to use oak-pojorsr [2] to configure
complete Oak stack as we do it in Sling. For that I would like to copy
this module to oak under oak-webapp and then refactor it to run
complete Oak stack.
Thoughts?
Chetan Mehrotra
[1] https
Hi Davide,
It would be helpful if while moving the bugs to next version we do not
add any comment like 'Bulk Move to xxx'. This would reduce the
unnecessary noise in bug comment history.
Recently this was discussed on DL at [1]
Chetan Mehrotra
[1] http://markmail.org/thread/2jvphl
> Time elapsed: 0.182 sec <<< ERROR!
>> org.apache.jackrabbit.oak.api.CommitFailedException: OakLucene0003: Failed
>> to index the node /test
Thats a known issue on Windows OAK-3072. Not a blocker for this release
Chetan Mehrotra
+1
Chetan Mehrotra
On Fri, Jul 31, 2015 at 3:44 PM, Alex Parvulescu
wrote:
> [X] +1 Release this package as Apache Jackrabbit Oak 1.0.18
>
> On Fri, Jul 31, 2015 at 11:01 AM, Amit Jain wrote:
>
>> A candidate for the Jackrabbit Oak 1.0.18 release is available a
Hi Team,
Currently most of the issues scheduled for 1.3.x release have comments
like 'Bulk Move to xxx'. This creates unnecessary noise in the comment
log. Would it be possible to move the issues to next version silently
i.e. just get fix version changed and not add any comment
Chetan Mehrotra
to see if the problem affect more usecases and
then decide to invest in such a feature!
Chetan Mehrotra
gle node deployment when TarMK is used
but would still have a lag as based on frequency of AsyncIndexer which
we are seeing is having delays of upto 10-20 sec and may vary. For
cluster where indexing is happening on a single node it cannot be
used.
Chetan Mehrotra
Lucene
Query mapping logic getting reused as is.
But yes it would be an interesting approach where instead of doing
this as Lucene query index level QE can work with both indexes and
combine the results. So an aspect to keep in mind
Chetan Mehrotra
nsient index would compliment it.
Chetan Mehrotra
On Fri, Jul 24, 2015 at 1:01 PM, Ian Boston wrote:
> Hi Chetan,
>
> The overall approach looks ok.
>
> Some questions about indexing.
>
> How will you deal with JVM failure ?
> and related.
> How frequently will comm
Observor so as to not add cost to commit
call. Observor would listen only for local changes and would invoke
IndexUpdate on the diff
Chetan Mehrotra
support this hybrid model
for same index type - So something to be looked into
Thoughts?
Chetan Mehrotra
NRT - Near real Time is technically a Lucene term
https://wiki.apache.org/lucene-java/NearRealtimeSearch. However using
here as approach is bit similar!
[2] Such a merged cursor and
uccess callback. If we can get both then much better!
Chetan Mehrotra
On Thu, Jul 23, 2015 at 3:08 PM, wrote:
> OAK-260 Avoid the "Turkish Locale Problem"
So the fix version is ... ;)
Chetan Mehrotra
callback
to IndexEditor if they are implementing Closeable?
Thoughts?
Chetan Mehrotra
On Wed, Jul 22, 2015 at 8:01 PM, Manfred Baedke
wrote:
> verifying that the BlobStoreBlob in question comes from this very instance.
> It shouldn't use equals(), though.
Makes sense. Lets discuss this via patch on the bug itself then!
Chetan Mehrotra
care of that. So probably we use that
here
So if we plan to support such case then we cleanup this part first and
then fix it.
wdyt?
Chetan Mehrotra
Also note that its possible that later we may be wrapping the
BlobStore instance. For example add a wrapper for monitioring and in
such a case this equality condition might fail. A more stable fix
would be to check with registered BlobStore weather it knows the given
blobId or not.
Chetan Mehrotra
rement to support equality for BlobStore instance. So would like
to understand the usecase preferably with a testcase. May be the
problem affect SegmentNodeStore also (not sure though)
Chetan Mehrotra
essages);
> +}
If all that work is to be done for debug logging then probably the
whole block should be within isDebugEnabled check
+ LOG.debug("insert of " + ids + " failed", ex);
Instead of using concatenation it would be better to use placeholders like
LOG.debug("insert of {} failed",ids, ex);
Chetan Mehrotra
others common lang classes so copying would
be tricky
2. Guava Stopwatch private method [2] - Guava Stopwatch internally has
such a method but its not exposed. Probably we can copy that and
expose that in oak-commons.
Thoughts?
Chetan Mehrotra
[1]
https://github.com/apache/commons-lang/blob/master
If this looks like a bug (which it appears to be) then it would be
better if we create an issue for this and have that merged to branches
also
Chetan Mehrotra
On Tue, Jul 14, 2015 at 9:28 AM, wrote:
> Author: dbrosius
> Date: Tue Jul 14 03:58:28 2015
> New Revision: 1690861
>
e would be DB specific approaches for
creating sparse index. Currently RDBDocumentStore stores some default
value if no value is specified. It might be better to store null there
[1]
> Each *document*? Did you mean collection?
Yes I meant collection there.
Chetan Mehrotra
[1] http://stackoverflow.com
+1. All checks passed
Chetan Mehrotra
On Fri, Jul 10, 2015 at 1:28 PM, Davide Giannella wrote:
> [X] +1 Release this package as Apache Jackrabbit Oak 1.0.17
> Davide
>
>
dexing property 'foo'
you need to index it for every revision
Chetan Mehrotra
schema.
Chetan Mehrotra
For issues to be merged to 1.0 and 1.2 branches.
Then later we can query for such issues.
Thoughts?
Chetan Mehrotra
(AbstractLifeCycle.java:64)
at
org.apache.jackrabbit.oak.remote.http.handler.RemoteServer.start(RemoteServer.java:54)
at
org.apache.jackrabbit.oak.remote.http.handler.RemoteServerIT.setUp(RemoteServerIT.java:134)
Chetan Mehrotra
On Thu, Jul 2, 2015 at 9:51 AM, Apache Jenkins Server
wrote:
> The Apache Jenkins bu
seeing the error ... may be
introduce later
Chetan Mehrotra
On Tue, Jun 30, 2015 at 12:00 PM, Francesco Mari
wrote:
> I suggest to fix OAK-3022 maintaining exactly
> the same behaviour, and without changing the SegmentNodeStoreService
Makes sense. They are two different issue
Chetan Mehrotra
org.apache.jackrabbit.oak.remote.http.handler.RemoteServerIT.load(RemoteServerIT.java:119)
at
org.apache.jackrabbit.oak.remote.http.handler.RemoteServerIT.testPatchLastRevisionAddMultiPathProperty(RemoteServerIT.java:1199)
-
Chetan Mehrotra
On Tue, Jun 30, 2015 at 10:11 AM, Apache Jenkins Server
wrote:
> The Apa
a similar namespaceing. For e.g. if
the property name is 'cache' then when reading from fwk then
'oak.documentstore.cache' would be used
oak.mongo.uri and oak.mongo.db are spacial cased though and not follow
this rule.
Chetan Mehrotra
On Tue, Jun 30, 2015 at 2:55 AM, Francesco Mari
-Doak.mongo.uri) which would supercede the one
packaged by default. This simplifies the testing.
Chetan Mehrotra
On Mon, Jun 29, 2015 at 7:01 PM, Davide Giannella wrote:
> On 29/06/2015 10:22, Francesco Mari wrote:
>> ...
>>
>> Is it possible - or does it make sense - to make this behavi
creation and a
> deletion) the events are never dropped, right? The line in [1] even if it
> skips the current NodeState it will include it in a later revision, right?
Yes
Chetan Mehrotra
ntinue
to work. Though there is no gurantee that they see *each* local
change.
Chetan Mehrotra
to external.
There are two solutions
1. Throttle the commits - CommitRateLimiter
2. Have a non limiting queue - Then you end up with OOM if gap in
processing rate is large
Chetan Mehrotra
On Thu, Jun 25, 2015 at 8:47 PM, Marius Petria wrote:
> Hi,
>
> I understand that under high l
break
Chetan Mehrotra
[1]
https://github.com/apache/jackrabbit-oak/blob/trunk/oak-lucene/src/main/java/org/apache/jackrabbit/oak/plugins/index/lucene/LuceneIndexEditor.java#L822
Just ensure that your Observer is fast as its invoked the critical path.
This would probably end up with a design similar to Background
Observer. May be better option would be to allow BO have non bounded
queue.
Chetan Mehrotra
On Wed, Jun 17, 2015 at 2:05 PM, Carsten Ziegeler wrote:
>
t at *any* cost that you can used a different impl which uses
say LinkedBlockingQueue and does not enforce any limit. That would be
similar to how JcrResourceListener works which uses an unbound in
memory queue
Chetan Mehrotra
ollection into separate
database on same server then we can look further there
Chetan Mehrotra
s like
$query: { _id: { $gt: "2:/a/", $lt: "2:/a0" }
Chetan Mehrotra
f 1:/a would be become tricky
Chetan Mehrotra
On Fri, Jun 12, 2015 at 1:56 PM, Amit Jain wrote:
> [ ] +1 Release this package as Apache Jackrabbit Oak 1.0.15
All checks ok
Chetan Mehrotra
4.2 sec <<< FAILURE!
> junit.framework.AssertionFailedError: Wrong size of NodeIterator in result
> expected:<51> but was:<-1>
> at junit.framework.Assert.fail(Assert.java:50)
> at junit.framework.Assert.failNotEquals(Assert.java:287)
> at juni
sLoader.access$000(URLClassLoader.java:58)
at java.net.URLClassLoader$1.run(URLClassLoader.java:197)
at java.security.AccessController.doPrivileged(Native Method)
Chetan Mehrotra
On Fri, May 22, 2015 at 9:48 AM, Apache Jenkins Server
wrote:
> See
> <https://builds.apache.org/job/Apache%
Opened OAK-2898 to track this.
Testing that passed stream at Oak/JCR level is buffered would indeed be tricky!
Chetan Mehrotra
On Thu, May 21, 2015 at 2:16 PM, Thomas Mueller wrote:
> Hi,
>
> Sure.
>
> We should probably have some kind of test case to ensure the stream is
>
On Thu, May 21, 2015 at 1:31 PM, Thomas Mueller wrote:
> Yes, it would be better if we wrap it somewhere (not necessarily right
> there, but somewhere
I think we can do that in DataStoreBlobStore. Right?
Chetan Mehrotra
buffered variant?
Chetan Mehrotra
[1]
https://github.com/apache/jackrabbit/blob/trunk/jackrabbit-data/src/main/java/org/apache/jackrabbit/core/data/LazyFileInputStream.java#L102
s obtained
> + * from Adobe Systems Incorporated.
> + * ********
Chetan Mehrotra
further!
Chetan Mehrotra
[1]
https://svn.apache.org/repos/asf/jackrabbit/oak/tags/jackrabbit-oak-1.0.13/RELEASE-NOTES.txt
[2] https://lucene.apache.org/core/4_1_0/changes/Changes.html#4.1.0.new_features
[3] https://code.google.com/p/guava-libraries/wiki/Release18
[4] http://lucene.apache.org/solr
where Persistent Cache doc
is referred.
Unless we restructure the site like say Apache Drill [4] (which shows
nested link in side bar) I think we should also refer to all such
inner links directly.
Thoughts?
Chetan Mehrotra
[1] http://jackrabbit.apache.org/oak/docs/
[2] http://jackrabbit.apache.org
Opened OAK-2892 to track this
Chetan Mehrotra
On Wed, May 20, 2015 at 2:45 PM, Chetan Mehrotra
wrote:
> On Wed, May 20, 2015 at 2:34 PM, Ian Boston wrote:
>> And does that apply to all BlobStore implementations including those that
>> use Mongo as the BlobStore
>
> It
Mongo
primary would be visible to later reads from other clients.
Latency that occurs while reading from repository is different - That
happens due to the way DocumentNodeStore performs background reads. So
any change from other cluster node would only be picked up when the
background read happens.
On Wed, May 20, 2015 at 1:40 PM, Ian Boston wrote:
> Isn't there a latency issue for a cluster ?
Latency applies for content accessed via NodeStore API. Changes made
to BlobStore should be immediately visible on other cluster nodes
Chetan Mehrotra
ting support
in most BlobStore implementations. We just need to manage key names
Thoughts?
Chetan Mehrotra
es use of already extracted
text content
Chetan Mehrotra
java_pid12637.hprof ...
Heap dump file created [596273278 bytes in 22.357 secs]
Exception in thread "TarMK flush thread [target/tar.1179582742341],
avg 1ms, max 123ms" java.lang.OutOfMemoryError: GC overhead limit
exceeded
Chetan Mehrotra
On Tue, May 19, 2015 at 5:01 PM, Travis CI wrot
perform this
Chetan Mehrotra
On Mon, May 18, 2015 at 2:55 PM, Michael Dürig wrote:
> Maybe we can make the dependency optional for now?
That should work. Then OSGi imports would be marked optional and if
some one wants to make use of that in OSGi env then he can optionally
deploy the required bundles
Chetan Mehrotra
Hi Michael,
On Mon, May 18, 2015 at 2:02 PM, wrote:
>
> +
> + org.apache.commons
> + commons-math3
> +
Is adding a required dependency on commons-math3 required? Probably
MicroBenchmark can be made part of commons/test and move this
dependency to tes
of that to update the release notes and
also ensure that documentation gets updated
Chetan Mehrotra
/amazonaws/aws-java-sdk-core/1.9.11/aws-java-sdk-core-1.9.11.pom
Chetan Mehrotra
On Mon, May 4, 2015 at 2:22 PM, Michael Dürig wrote:
>
> Hi,
>
> I've seen this happen a few times lately but with no apparent pattern.
> Anyone has an idea what could be causing this?
>
>
so often struggled to get hold of underlying NodeStore from
given Oak instance. May be we should expose it as part of API itself.
After each Oak instance would always be back by NodeStore
Chetan Mehrotra
301 - 400 of 773 matches
Mail list logo
