Task #5008
closed
Enable Infrastructure Gateway cluster on Production
100%
Description
infra-gateway1.d4science.org will be added to the cluster currently composed by the infra-gateway.d4science.org node only.
The Liferay instances running on both machines will be configured to sync via a multicast RMI protocol.
The 2 tomcats will NOT be configured in cluster, so that when a node fails the user is ported to the other instance and will have to login again. In the next release the 2 tomcats will be configured in cluster and the http session will be migrated transparently (we're not ready for this yet in gCube 4).
The whole operation should not require any interruption of the service.
Updated by Massimiliano Assante almost 9 years ago
@andrea.dellamico@isti.cnr.it could you confirm the following:
- the network for Multicast Clustering Setting on production will be 229.0.0.0/24
- infra gateway and infra gateway1 node can communicate without firewall via RMI
Updated by Andrea Dell'Amico almost 9 years ago
Yes, that's correct. The firewall rules are already active.
Updated by Massimiliano Assante almost 9 years ago
- Due date changed from Sep 14, 2016 to Sep 13, 2016
- Start date changed from Sep 14, 2016 to Sep 13, 2016
Updated by Massimiliano Assante almost 9 years ago
- Status changed from New to In Progress
together with this activity we are going to deploy a patch for fixing the unzip problem of workspace folder on Windows https://support.d4science.org/issues/5013
Updated by Massimiliano Assante almost 9 years ago
- Status changed from In Progress to Closed
- % Done changed from 0 to 100
Cluster composed by infra-gateway.d4s and infra-gateway1.d4s is now fully operational.
Updated by Massimiliano Assante almost 9 years ago
- Status changed from Closed to In Progress
- % Done changed from 100 to 90
the Liferay’s Documents and Media Library should be "clustered" as well https://support.d4science.org/issues/5022
Updated by Massimiliano Assante almost 9 years ago
- Status changed from In Progress to Closed
Updated by Massimiliano Assante over 8 years ago
- Status changed from Closed to In Progress
We need Clustering Lucene Indexes on All Nodes of the cluster, as Liferay API methods use that index for instance to get the list of users
According to the Liferay's documentation a property set should suffice, but we need server restart
Set this to true if you want the portal to replicate an index write across all members of the cluster. This is useful in some clustered environments where you wish each server instance to have its own copy of the Lucene search index. This is only relevant when using the default Lucene indexing engine.
Defaults:
lucene.replicate.write=false
Updated by Massimiliano Assante over 8 years ago
- Status changed from In Progress to Closed
it seems working, closing the ticket
17:14:57,303 INFO [liferay/search_writer/SYSTEM_ENGINE-1][EditServerAction:997] Thread name liferay/search_writer/SYSTEM_ENGINE-1 with thread ID 244 synchronized on latch. Waiting for others. 17:14:57,304 INFO [com.liferay.portlet.admin.action.EditServerAction-1][EditServerAction:1054] Monitor thread name com.liferay.portlet.admin.action.EditServerAction-1 with thread ID 996 unlocked latch. Notified peers to start index loading.
