ASS 2.0: Solr shards placement (enforced by user)

As we transition to the new Hyland Community, some post links may be temporarily disrupted as we finalize redirects. We apologize for the inconvenience and are working to resolve this as quickly as possible. Details.

luis_alventosa

Hello,

I need some clarification about how solr sharding works with Alfresco 7 + ASS 2.0 Community Edition.

I would like to place my shards on specific servers and tie them to some specific SSD disks using DB_ID_RANGE.

I managed to create a first deployment where I configure the shards on specific nodes and it works fine.

However, it looks like when trying to add new shards to an existing deployment alfresco repository is deciding where the new shards are deployed and the existing ones are moved around onto new servers and as such that breaking my SSD configuration.

It seems there's a java class in alfresco repository placing the shard instances (with replicas) following a given algorithm which imposes the solr shard placement and does not allow the user to decide where new shards sould be placed.

https://github.com/Alfresco/alfresco-core/blob/c886cc2c997d5be2c5007f252d4ac6d2dc8d1e47/src/main/java/org/alfresco/util/shard/ExplicitShardingPolicy.java#L62

Has anyone encountered the same issue?

Any help would be much appreciated.

Thanks in advance,

Luis

angelborroy

You need every shard for a DBID Shard configuration ("All shards required"), as described in documentation:

https://docs.alfresco.com/insight-engine/latest/config/sharding/#comparison-overview

Since Community only supports that Shard method, no alternative is provided.

For Enterprise deployments, DBID Range can be used in order to add new shards dynamically.