Slow Index Init on server startup

2016-06-19 Thread ZeroCool86
Good Morning guys, we use Jackrabbit in an production environment and detected a wait time in Workspace Index Init process of about 10min. Nothing happens in the log between these two lines, anybody knows whats happening in this step and any ideas why this took so much time? This is the biggest

Re: Slow Index Init on server startup

2016-06-27 Thread ZeroCool86
Hi, thanks for your advice. I used JConsole to inspect the start. Here is the Stacktrace where the time is spent on startup. Is there any chance to decrease this process duration? org.apache.jackrabbit.core.query.lucene.CachingIndexReader$CacheInitializer$2.collect(CachingIndexReader.java:482)

Re: JcrMappingException: Could not create node - Lock wait timeout exceeded; try restarting transaction Help Needed

2016-09-16 Thread ZeroCool86
Hi, we set our journal in a separate DB to reduce the risk of Lock Problems due our normal Content Databases. We faced the same problem when we use the journal and content in same database due lot of revisions in a cluster environment. The Journal Revision Updates blocked our normal read/write

Journal Lock after Index Rebuild

2016-12-29 Thread ZeroCool86
Hi Guys, we faced the problem with "Unable to lock global revision table" after an Index rebuild. The Index build takes 1-2 days. After that process, the revisions of this range gets synchronized. Is there any possibility to make this revision sync after index build smoother? The syncDelay fo

Re: Journal Lock after Index Rebuild

2017-01-24 Thread ZeroCool86
Hi guys, no ideas about this behaviour? Thank you, best regards Michael   - Gesendet: Donnerstag, 29. Dezember 2016 um 09:54 Uhr Von: zerocoo...@web.de An: users@jackrabbit.apache.org Betreff: Journal Lock after Index Rebuild Hi Guys, we faced the problem with "Unabl

XPath jcr:contains(.,"value") exlude one or more Properties

2017-02-04 Thread ZeroCool86
Hi guys, is it possible to do a jcr:contains(.,"my value") Query and exclude one or more properties from being searched for the current search? Something like "search all possible properties but ignore property1 and property2". Or is it just possible in the other way explicit name the searched

Re: XPath jcr:contains(.,\"value\") exlude one or more Properties

2017-02-04 Thread ZeroCool86
Hi, thanks for your response. I know this normal search pattern in a specific property. I was hoping there is something like a blacklist -> exclude one property but search all others without name theme explicit. It seems not ;-) Greetings, Michael

Index rebuild necessary? / Jackrabbit Update

2017-07-18 Thread ZeroCool86
Hi guys, we have a scenario with an old Jackrabbit Version 2.2.5 Is it necessary to rebuild all the Lucene Indexes if we would like to upgrade to an up to date Version (at the moment 2.15.4) due the different Lucene Upgrades? Is there a best practise / checklist for upgrading the Jackrabbit Ve

Re: Index rebuild necessary? / Jackrabbit Update

2017-07-21 Thread ZeroCool86
Hi guys, does nobody have Infos about this Upgradeprocess especially in view of Lucene Index Backwardcompatibility due Lucene Updates in the Jackrabbit Versions and Jcr Data Backwardcompatibility from an old 2.2.5 Version to an up-to-date Version like 2.15.4 (or is there a Version which fits be

Re: Index rebuild necessary? / Jackrabbit Update

2017-07-21 Thread ZeroCool86
Hi, thanks Julian for the reply and the hint to the unstable Version. In view of OAK, i struggled with the hint in the Backward Compatibility Page about "Node Name Length Limit" and https://issues.apache.org/jira/browse/OAK-2644 The Index Question is still interesting for me, maybe someone els