-
Type:
Improvement
-
Status: Resolved
-
Priority:
Minor
-
Resolution: Fixed
-
Affects Version/s: 9.10
-
Fix Version/s: 9.10-HF34, 10.10-HF11, 11.1, 2021.0
-
Component/s: Elasticsearch
When a document is checked in, the document and all its versions are reindexed in Elasticsearch.
The current implementation is not optimized because it was difficult at that time to retrieve the previous last version and last major version. But the consequence is a potential problem when a document has many versions : the Elasticsearch request to index the document is huge and the build of the ES request could time out
[org.nuxeo.ecm.core.work.AbstractWork] Exception during work: IndexingWorker(306e0e21-91bb-45c0-9f5e-274a6ab94c83, , Progress(?%, ?/0), null)
org.nuxeo.runtime.transaction.TransactionRuntimeException: Transaction has timed out
at org.nuxeo.runtime.transaction.TransactionHelper.checkTransactionTimeout(TransactionHelper.java:223)
at org.nuxeo.ecm.core.api.local.LocalSession.getSession(LocalSession.java:107)
at org.nuxeo.ecm.core.api.AbstractSession.resolveReference(AbstractSession.java:338)
at org.nuxeo.ecm.core.api.AbstractSession.exists(AbstractSession.java:844)
at org.nuxeo.elasticsearch.commands.IndexingCommand.getTargetDocument(IndexingCommand.java:162)
at org.nuxeo.elasticsearch.core.ElasticSearchIndexingImpl.buildEsIndexingRequest(ElasticSearchIndexingImpl.java:408)
