cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stefano Ortolani (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (CASSANDRA-13667) DROP KEYSPACE or TABLE cause unrelated flushes and compactions on all tables
Date Wed, 05 Jul 2017 16:03:00 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-13667?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16074981#comment-16074981
] 

Stefano Ortolani edited comment on CASSANDRA-13667 at 7/5/17 4:02 PM:
----------------------------------------------------------------------

The cluster was serving reads/writes as usual (1 DC, 2 racks).

I have been able to reproduce it multiple times with a single drop.
Every time I could see spikes with 40/50 pending compactions, and all memtables being flushed
(maybe the actual cause for those compactions?). Also, {{nodetool describecluster}} kept showing
many nodes unreachable during the schema propagation (settling down after 10 minutes).

The schema/tables I was dropping were those created by cassandra-reaper (https://github.com/thelastpickle/cassandra-reaper)
(keyspace with {{class: SimpleStrategy}}).
All other keyspaces are {{class: NetworkTopologyStrategy}}.


was (Author: ostefano):
The cluster was serving reads/writes as usual (1 DC, 2 racks).

I have been able to reproduce it multiple times with a single drop.
Every time I could see spikes with 40/50 pending compactions, and all memtables being flushed
Also, {{nodetool describecluster}} kept showing many nodes unreachable during the schema propagation
(settling down after 10 minutes).

The schema/tables I was dropping were those created by cassandra-reaper (https://github.com/thelastpickle/cassandra-reaper)
(keyspace with {{class: SimpleStrategy}}).
All other keyspaces are {{class: NetworkTopologyStrategy}}.

> DROP KEYSPACE or TABLE cause unrelated flushes and compactions on all tables
> ----------------------------------------------------------------------------
>
>                 Key: CASSANDRA-13667
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13667
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Stefano Ortolani
>            Priority: Minor
>
> As soon as I drop a keyspace or a table, I see _all_ nodes struggling to acknowledge
the new schema because of several flushes and compactions happening on _all_ keyspaces and
compactions (completely unrelated to the dropped keyspace/table).



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org


Mime
View raw message