Hi,
About the number of Zookeeper elements in an ensemble, you can find this
good information in this page. It applies to Solr.
https://www.cloudkarafka.com/blog/2018-07-04-cloudkarafka-how-many-zookeepers-in-a-cluster.html
- 1 Node: no fault tolerance, no maintenance possibilities
- 3 Nodes: An ensemble with 3 nodes will support one failure without
loss of service, which is probably fine for most users, and also the most
popular setup
- 5 Nodes (recommended for real fault tolerance): A five-node cluster
allows to take one server out for maintenance or upgrade and still be able
to take a second unexpected failure, without interrupting your service
- 7 Nodes: The same as for 5-node cluster but with the ability to bear
the failure of three nodes
The Zookeeper ensemble size is not dependent of the number of Solr nodes.
Zookeeper activity is not in relation with updates or queries volume. Il is
related with :
- solr node stop / start and so recovery
- solr and alias collection creation / destruction ...
- solr configset management
If your solr cluster is stable in terms of functional solr nodes and
collections, even with huge data updates and/or queries, the Zookeeper
ensemble won't be stressed.
About the upgrade method, I am not sure a hot operation is possible. In
order to minimise downtime :
Step 1 - Set up 2 new zookeeper with the 3 servers declaration
server.1=xxx:2881:3881
server.2=yyy:2882:3882
server.3=zzz:2883:3883
Step 2 - Add the 2 new servers declaration in the running zookeeper
server.2=yyy:2882:3882
server.3=zzz:2883:3883
Step 3 - Restart the running zookeeper and start the 2 new zookeeper
Step 4 - Wait for data synchronization
Step 5 - Stop, move, start the first zookeeper
Warning : do not use IP address, use server names in your Zookeeper
configuration
Regards
Dominique
Le ven. 14 août 2020 à 05:48, yaswanth kumar a
écrit :
> Hi Team
>
> Can someone let me know if we can do an upgrade to zookeeper ensemble
> from standalone ??
>
> I have 3 solr nodes with one zookeeper running on one of the node .. and
> it’s a solr cloud .. so now can I install zookeeper on another node just to
> make sure it’s not a single point of failure when the solr node that got
> zookeeper is down??
>
> Also want to understand what’s the best formula of choosing no of
> zookeeper that’s needed for solr cloud like for how many solr nodes .. how
> many zookeeper do we need to maintain for best fault tolerance
>
> Sent from my iPhone
