Solr Index Timeout on large Ldap-Groups

We have a rather big Ldap-Group with 1000+ Members. This Group is used in one or two webs and on some Topics. Solr-Indexer needs about 2-3 Minutes to index such a Topic. During a Full Index (Cronjob) the Indexer even terminates after a certain amount of time. We are NOT able to run a Delta-Index every 5 Minutes anymore. the Jobs Queue up and eventually just terminate after a while.

If we manually index ONLY the corresponding web or topic, the indexer succeeds, but needs to much time.

Ldap and SOLR running the latest Plugin-Versions. Cache is OK and has been rebuild several times during testing.

Where could we set a higher Timeout value (as a workaround) for the Solr-Indexer? Or do you have a tip what configuration we could trim to speed things up here?

Thanks in advance,

-- OliverSchaub - 14 Mar 2012

Noticed something... is it normal that the Memory-Usage of solrindex is steadily increasing? It's running now, not nearly half way through indexing and already at 160mb. Shouldn't it free memory after every commit?

-- OliverSchaub - 14 Mar 2012

See Tasks.Item11664

-- MichaelDaum - 16 Mar 2012

The referenced bug item is fixed as of today. Do the problems still occur?

-- MichaelDaum - 27 Feb 2015