[
https://issues.apache.org/jira/browse/CASSANDRA-14618?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16596373#comment-16596373
]
Jason Brown commented on CASSANDRA-14618:
-----------------------------------------
bq. yeah lets move it to tools
sgtm. I would say you could probably just do that at commit time, but I think
the utests will be affected. But, yeah, let's do it as a last step.
bq. Since full querylogging has not been shipped in a released version, maybe
we can punt the versioning until after that?
Ahh, right. Then since this is not released yet, why do we need to support a
version of the log files that has no keyspace?
My last major thought is: There's an assumption about an ordered, one-to-one
mapping between targetHosts -> QueryReplayer.sessions /
ResultHandler.resultPaths / ComparableResultSet that is passed to
ResultHandler.handleResults / ResultsStore.queues. This feature is small enough
that I can keep most of it in my head, and I think it's correct, but I wonder
if there's another data structure to make it more fool-proof and more obvious
about the relationships between the hosts (or clusters) on which an event
occurred and how that event is referenced in relation to the target. I'm not
sure how difficult or invasive it is, or if there's a another/simpler way to
enforce the orderings/sizes. Lacking anything else, maybe just document it
everywhere (you've already got it in a bunch of places already).
On the whole I think we're looking good here.
> Create fqltool replay command
> -----------------------------
>
> Key: CASSANDRA-14618
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14618
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Marcus Eriksson
> Assignee: Marcus Eriksson
> Priority: Major
> Labels: fqltool
> Fix For: 4.x
>
>
> Make it possible to replay the full query logs from CASSANDRA-13983 against
> one or several clusters. The goal is to be able to compare different runs of
> production traffic against different versions/configurations of Cassandra.
> * It should be possible to take logs from several machines and replay them in
> "order" by the timestamps recorded
> * Record the results from each run to be able to compare different runs
> (against different clusters/versions/etc)
> * If {{fqltool replay}} is run against 2 or more clusters, the results should
> be compared as we go
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
