cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aleksey Yeschenko (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-14457) Add a virtual table with current compactions
Date Mon, 21 May 2018 11:27:00 GMT


Aleksey Yeschenko commented on CASSANDRA-14457:

This looks good, so I only have some bikeshedding to contribute:

1. The table doesn't really represent compaction statistics, so should probably not name it
compaction_stats? I know that the nodetool cmd is named compactionstats, but that was a bad
name too, imo. So perhaps {{compaction_state}} and {{CompactionStateTable}}, or {{compcation_status}}
and {{CompactionStatusTable}}, or even {{active_compactions}} and {{ActiveCompactionsTable}};
whichever sounds nicer to your American ear.
2. Should we perhaps use {{int}} as the type for current and total columns, instead of {{text}}?
3. Maybe don't concat the names of the keyspace and the table? Why not have them in separate
columns, for easier querying?
4. We tend to lowercase enums in table, usually. Can you slap a {{toLowerCase()}} on {{task_type}}
5. I would prefer to have ((keyspace, table), id) or ((keyspace), table, id) as PRIMARY KEY
here, personally.

Upon a quick look, it seems like the only case where we don't have a keyspace/table attached
to a compaction is summary redistribution, which is performed on all sstables. But it's not
really a compaction, so perhaps we should exclude it from the dataset?

> Add a virtual table with current compactions
> --------------------------------------------
>                 Key: CASSANDRA-14457
>                 URL:
>             Project: Cassandra
>          Issue Type: New Feature
>            Reporter: Chris Lohfink
>            Assignee: Chris Lohfink
>            Priority: Minor
>             Fix For: 4.x

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message