I asked this previously but didn’t see a response in the follow ups. If were
were to improve upon the code so that it didn’t stack overflow when it can’t
read the index version on a specific member/bucket, ignore the members/buckets
that don’t return a result because they can’t read the index, could we not
aggregate the results of the members/buckets that can read the index and still
produce the expected results?
It seems like there should be sufficient coverage across all the nodes that if
the buckets are redundant, given they are replicating, that one node should be
able to read the index contained in that bucket and the other nodes can be
ignored for that bucket.
-Jake
> On Sep 27, 2021, at 2:33 PM, Nabarun Nag <n...@vmware.com> wrote:
>
> The solution for preventing the query executions to occur in the mixed
> version mode also caused some problems where the query function executions
> get repeatedly executed and that results in stack overflow.
>
>
> ________________________________
> From: Nabarun Nag <n...@vmware.com>
> Sent: Monday, September 27, 2021 2:30 PM
> To: dev@geode.apache.org <dev@geode.apache.org>
> Subject: Re: [DISCUSS] Upgrading to Lucene 7.1.0
>
> In simple words, if Lucene indexes were created by a new version (7.1.0),
> then replicated to others that are still in the older version, they won't
> understand the index, and the event processors start throwing exceptions.
>
> This can be simply seen by just re-enabling the query execution in the DUnit
> tests and commenting out the check blocks: [develop SHA:
> 68629356f561a932f5dfbace70b01d9971a42473]
>
> In LuceneEventListener
> if (cache.hasMemberOlderThan(KnownVersion.GEODE_1_15_0)) {
> logger.info("Some members are older than " +
> KnownVersion.GEODE_1_15_0.getName());
> return false;
> }
>
> In IndexRepositoryFactory:
> if (userRegion.getCache() != null
> &&
> userRegion.getCache().hasMemberOlderThan(KnownVersion.GEODE_1_15_0)) {
> logger.info("Some members are older than " +
> KnownVersion.GEODE_1_15_0.getName());
> return null;
> }
>
>
> This is the exception that will be encountered:
>
> [Exception]
>
> [vm2_v1.2.0] [warn 2021/09/27 14:24:42.251 PDT <Event Processor for
> GatewaySender_AsyncEventQueue_index#_aRegion_5> tid=102] An Exception
> occurred. The dispatcher will continue.
> [vm2_v1.2.0] org.apache.geode.InternalGemFireError: Unable to create index
> repository
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.lambda$computeRepository$0(AbstractPartitionedRepositoryManager.java:118)
> [vm2_v1.2.0] at
> java.util.concurrent.ConcurrentHashMap.compute(ConcurrentHashMap.java:1853)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.computeRepository(AbstractPartitionedRepositoryManager.java:108)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.getRepository(AbstractPartitionedRepositoryManager.java:137)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.getRepository(AbstractPartitionedRepositoryManager.java:76)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.LuceneEventListener.process(LuceneEventListener.java:87)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.LuceneEventListener.processEvents(LuceneEventListener.java:64)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.GatewaySenderEventCallbackDispatcher.dispatchBatch(GatewaySenderEventCallbackDispatcher.java:154)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.GatewaySenderEventCallbackDispatcher.dispatchBatch(GatewaySenderEventCallbackDispatcher.java:80)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.AbstractGatewaySenderEventProcessor.processQueue(AbstractGatewaySenderEventProcessor.java:609)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.AbstractGatewaySenderEventProcessor.run(AbstractGatewaySenderEventProcessor.java:1051)
> [vm2_v1.2.0] Caused by: org.apache.lucene.index.IndexFormatTooNewException:
> Format version is not supported (resource
> BufferedChecksumIndexInput(segments_2)): 7 (needs to be between 4 and 6)
> [vm2_v1.2.0] at
> org.apache.lucene.codecs.CodecUtil.checkHeaderNoMagic(CodecUtil.java:216)
> [vm2_v1.2.0] at
> org.apache.lucene.index.SegmentInfos.readCommit(SegmentInfos.java:302)
> [vm2_v1.2.0] at
> org.apache.lucene.index.SegmentInfos.readCommit(SegmentInfos.java:286)
> [vm2_v1.2.0] at
> org.apache.lucene.index.IndexWriter.<init>(IndexWriter.java:938)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.IndexRepositoryFactory.computeIndexRepository(IndexRepositoryFactory.java:84)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.PartitionedRepositoryManager.computeRepository(PartitionedRepositoryManager.java:42)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.lambda$computeRepository$0(AbstractPartitionedRepositoryManager.java:116)
> [vm2_v1.2.0] ... 10 more
>
>
>
> Also:
> [vm2_v1.2.0] [warn 2021/09/27 14:24:42.134 PDT <Event Processor for
> GatewaySender_AsyncEventQueue_index#_aRegion_7> tid=106] An Exception
> occurred. The dispatcher will continue.
> [vm2_v1.2.0] org.apache.geode.InternalGemFireError: Unable to create index
> repository
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.lambda$computeRepository$0(AbstractPartitionedRepositoryManager.java:118)
> [vm2_v1.2.0] at
> java.util.concurrent.ConcurrentHashMap.compute(ConcurrentHashMap.java:1853)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.computeRepository(AbstractPartitionedRepositoryManager.java:108)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.getRepository(AbstractPartitionedRepositoryManager.java:137)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.getRepository(AbstractPartitionedRepositoryManager.java:76)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.LuceneEventListener.process(LuceneEventListener.java:87)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.LuceneEventListener.processEvents(LuceneEventListener.java:64)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.GatewaySenderEventCallbackDispatcher.dispatchBatch(GatewaySenderEventCallbackDispatcher.java:154)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.GatewaySenderEventCallbackDispatcher.dispatchBatch(GatewaySenderEventCallbackDispatcher.java:80)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.AbstractGatewaySenderEventProcessor.processQueue(AbstractGatewaySenderEventProcessor.java:609)
> [vm2_v1.2.0] at
> org.apache.geode.internal.cache.wan.AbstractGatewaySenderEventProcessor.run(AbstractGatewaySenderEventProcessor.java:1051)
> [vm2_v1.2.0] Caused by: org.apache.lucene.index.IndexFormatTooNewException:
> Format version is not supported (resource
> BufferedChecksumIndexInput(segments_2)): 7 (needs to be between 4 and 6)
> [vm2_v1.2.0] at
> org.apache.lucene.codecs.CodecUtil.checkHeaderNoMagic(CodecUtil.java:216)
> [vm2_v1.2.0] at
> org.apache.lucene.index.SegmentInfos.readCommit(SegmentInfos.java:302)
> [vm2_v1.2.0] at
> org.apache.lucene.index.SegmentInfos.readCommit(SegmentInfos.java:286)
> [vm2_v1.2.0] at
> org.apache.lucene.index.IndexWriter.<init>(IndexWriter.java:938)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.IndexRepositoryFactory.computeIndexRepository(IndexRepositoryFactory.java:84)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.PartitionedRepositoryManager.computeRepository(PartitionedRepositoryManager.java:42)
> [vm2_v1.2.0] at
> org.apache.geode.cache.lucene.internal.AbstractPartitionedRepositoryManager.lambda$computeRepository$0(AbstractPartitionedRepositoryManager.java:116)
> [vm2_v1.2.0] ... 10 more
>
>
> ________________________________
> From: Jacob Barrett <jabarr...@vmware.com>
> Sent: Monday, September 27, 2021 2:08 PM
> To: dev@geode.apache.org <dev@geode.apache.org>
> Subject: Re: [DISCUSS] Upgrading to Lucene 7.1.0
>
>
>
>> On Sep 27, 2021, at 11:48 AM, nabarun nag <n...@apache.org> wrote:
>>
>> Recently, a commit was pushed to develop which upgraded the Lucene
>> version used in Apache Geode to 7.1.0. These new Lucene indexes are
>> not compatible with the previous versions hence it breaks the rolling
>> upgrade contract. We are no longer able to execute Lucene queries when
>> there are severs of mixed versions in the cluster.
>
>
> Can you describe the problem with a little more detail? Does this mean that
> while there is a mix the execution throws an exception on all servers or is
> there a subset for which it works? If there is a subset for which it works,
> are those instances sufficient to provide accurate results if the instances
> that fail are ignored?
>
> -Jake
>
