[JBoss JIRA] (ISPN-5665) Query should not rely on the results of return values of write commands
by Sanne Grinovero (JIRA)
[ https://issues.jboss.org/browse/ISPN-5665?page=com.atlassian.jira.plugin.... ]
Sanne Grinovero reassigned ISPN-5665:
-------------------------------------
Assignee: Gustavo Fernandes
> Query should not rely on the results of return values of write commands
> -----------------------------------------------------------------------
>
> Key: ISPN-5665
> URL: https://issues.jboss.org/browse/ISPN-5665
> Project: Infinispan
> Issue Type: Bug
> Components: Embedded Querying
> Affects Versions: 8.0.0.Beta2
> Reporter: Dan Berindei
> Assignee: Gustavo Fernandes
> Fix For: 8.0.0.Final
>
>
> The query interceptor relies on the return value of the write commands to know the previous value of the modified entries. This is not correct, because some write commands do not return the previous value, e.g. {{remove(key, value)}}, {{replace(key, oldValue, newValue)}}, and {{putAll(map)}}.
> The query interceptor should instead look up the previous values in the invocation context, and also force the loading of old values in the invocation context if the command doesn't do it explicitly (e.g. {{putAll(map)}}, or {{put(k, v)}} with the {{IGNORE_RETURN_VALUES}} flag).
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
10 years, 8 months
[JBoss JIRA] (ISPN-5101) Handle database restart
by Sanne Grinovero (JIRA)
[ https://issues.jboss.org/browse/ISPN-5101?page=com.atlassian.jira.plugin.... ]
Sanne Grinovero commented on ISPN-5101:
---------------------------------------
Are there more details about the configuration please? I don't see how the database could affect the Infinispan state, without knowing how the two are connected: is there a CacheStore involved for persistence of the Infinispan caches?
> Handle database restart
> -----------------------
>
> Key: ISPN-5101
> URL: https://issues.jboss.org/browse/ISPN-5101
> Project: Infinispan
> Issue Type: Bug
> Components: Lucene Directory
> Affects Versions: 7.0.0.Alpha4
> Environment: database: postgress, using hibernate search with infinispan&lucene
> Reporter: yaniv oren
> Labels: lucene
>
> relates to https://issues.jboss.org/browse/ISPN-1680
> scenario:
> restarting database during application run.
> expected:
> App recovery.
> Actual:
> Received error:
> ..
> Caused by: org.hibernate.search.exception.SearchException: Unable to reopen IndexReader
> at org.hibernate.search.indexes.impl.SharingBufferReaderProvider$PerDirectoryLatestReader.refreshAndGet(SharingBufferReaderProvider.java:241) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.indexes.impl.SharingBufferReaderProvider.openIndexReader(SharingBufferReaderProvider.java:72) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.indexes.impl.SharingBufferReaderProvider.openIndexReader(SharingBufferReaderProvider.java:34) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.reader.impl.MultiReaderFactory.openReader(MultiReaderFactory.java:36) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.query.engine.impl.HSQueryImpl.buildSearcher(HSQueryImpl.java:617) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.query.engine.impl.HSQueryImpl.buildSearcher(HSQueryImpl.java:517) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.query.engine.impl.HSQueryImpl.queryEntityInfos(HSQueryImpl.java:251) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> at org.hibernate.search.query.hibernate.impl.FullTextQueryImpl.list(FullTextQueryImpl.java:198) ~[hibernate-search-orm-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> ...
> Caused by: java.io.IOException: Read past EOF
> at org.infinispan.lucene.impl.SingleChunkIndexInput.readByte(SingleChunkIndexInput.java:54) ~[infinispan-lucene-directory-7.0.0.Alpha4.jar:7.0.0.Alpha4]
> at org.apache.lucene.store.BufferedChecksumIndexInput.readByte(BufferedChecksumIndexInput.java:41) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.store.DataInput.readInt(DataInput.java:96) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.SegmentInfos.read(SegmentInfos.java:331) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.SegmentInfos$1.doBody(SegmentInfos.java:416) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:864) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:710) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.SegmentInfos.read(SegmentInfos.java:412) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.StandardDirectoryReader.isCurrent(StandardDirectoryReader.java:347) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.StandardDirectoryReader.doOpenNoWriter(StandardDirectoryReader.java:301) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.StandardDirectoryReader.doOpenIfChanged(StandardDirectoryReader.java:264) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.StandardDirectoryReader.doOpenIfChanged(StandardDirectoryReader.java:252) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.apache.lucene.index.DirectoryReader.openIfChanged(DirectoryReader.java:171) ~[lucene-core-4.8.1.jar:4.8.1 1594670 - rmuir - 2014-05-14 19:22:52]
> at org.hibernate.search.indexes.impl.SharingBufferReaderProvider$PerDirectoryLatestReader.refreshAndGet(SharingBufferReaderProvider.java:238) ~[hibernate-search-engine-5.0.0.Alpha4.jar:5.0.0.Alpha4]
> ... 121 common frames omitted
> ...
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
10 years, 8 months
[JBoss JIRA] (ISPN-5622) Make use of parallel iteration in the MassIndexer
by Sanne Grinovero (JIRA)
[ https://issues.jboss.org/browse/ISPN-5622?page=com.atlassian.jira.plugin.... ]
Sanne Grinovero commented on ISPN-5622:
---------------------------------------
Right but that type of contention isn't a big problem, if any it means the indexing couldn't be faster than what you achieved. I expect it to be strongly beneficial for distributed data?
Also to keep in mind: the {{indexwriter.max_thread_states}} property - although it defaults to 8 so probably not related to your test.
I think there was a plan for Lucene 5 to drop that limit, we'll need to check if that was done too -> HSEARCH-1962
> Make use of parallel iteration in the MassIndexer
> -------------------------------------------------
>
> Key: ISPN-5622
> URL: https://issues.jboss.org/browse/ISPN-5622
> Project: Infinispan
> Issue Type: Enhancement
> Components: Embedded Querying
> Reporter: Gustavo Fernandes
> Assignee: Gustavo Fernandes
>
> As the EntryRetriever is being deprecated, there's a possibility of speeding up the indexing process by using parallel streams
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
10 years, 8 months