cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Marcus Olsson (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-11258) Repair scheduling - Resource locking API
Date Fri, 04 Mar 2016 14:57:40 GMT


Marcus Olsson commented on CASSANDRA-11258:

bq. WDYT of this approach?
I like it, it would reduce the amount of work for the write operations while keeping a simple
structure of the lock keyspace. However, *if* we want to visualize all data center locks to
the user, it would be more complex to do so.

bq. If you like this approach, we could do the initial version assuming a single DC with SimpleStrategy
replication + SERIAL consistency, while power users could still have multi-DC support by manually
changing replication settings of the lock keyspace. We could later add transparent/efficient
multi-DC support via CASSANDRA-11300 and LocalDataCenterReplicationStrategy.

Should we start with {{SimpleStrategy}} and a replication of 3 then? If we only have one node
there would be no need for repairs and if we have two it would be possible to do CAS requests.
If we include the {{LocalDataCenterReplicationStrategy}} class directly in the first version
but use SimpleStrategy until CASSANDRA-11300 is in place it should be possible to switch to
LDCRS automatically later on, assuming that repairs would be paused during that upgrade. If
we don't include LDCRS directly I think the switch would either have to be manual or done
by first verifying the version of the other nodes, so that the "older" nodes wouldn't complain
about an unknown replication class.

> Repair scheduling - Resource locking API
> ----------------------------------------
>                 Key: CASSANDRA-11258
>                 URL:
>             Project: Cassandra
>          Issue Type: Sub-task
>            Reporter: Marcus Olsson
>            Assignee: Marcus Olsson
>            Priority: Minor
> Create a resource locking API & implementation that is able to lock a resource in
a specified data center. It should handle priorities to avoid node starvation.

This message was sent by Atlassian JIRA

View raw message