cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sam Tunnicliffe (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (CASSANDRA-12969) Index: index can significantly slow down boot
Date Mon, 05 Dec 2016 11:34:59 GMT

     [ https://issues.apache.org/jira/browse/CASSANDRA-12969?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Sam Tunnicliffe updated CASSANDRA-12969:
----------------------------------------
       Resolution: Fixed
    Fix Version/s:     (was: 3.x)
                   3.12
                   4.0
           Status: Resolved  (was: Patch Available)

bq.Hello, did it go well ?
Yes, thanks for the patch and sorry that I forgot to include the link to the CI jobs in my
previous comment:

||branch||testall||dtest||
|[12969-3.X|https://github.com/beobal/cassandra/tree/12969-3.X]|[testall|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-12969-3.X-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-12969-3.X-dtest]|
|[12969-trunk|https://github.com/beobal/cassandra/tree/12969-trunk]|[testall|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-12969-trunk-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-12969-trunk-dtest]|

The CI looks good relative to upstream (there are some dtest failures also present in trunk),
so I've committed to 3.X in {{afbc2e8502a8a8d1d6a319017dfc3c2a45bebaca}} and merged to trunk.


bq.Another sensible optimization will be CASSANDRA-12962
I'm sure someone will take a look at that as soon as they have chance. In the meantime, I
quickly glanced at the ticket and I'm afraid I don't really understand the problem you're
describing. Perhaps the description could use a little work to make it a bit clearer?

> Index: index can significantly slow down boot
> ---------------------------------------------
>
>                 Key: CASSANDRA-12969
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-12969
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: CQL
>            Reporter: Corentin Chary
>             Fix For: 4.0, 3.12
>
>         Attachments: 0004-index-do-not-re-insert-values-in-IndexInfo.patch
>
>
> During startup, each existing index is opened and marked as built by adding an entry
in "IndexInfo" and forcing a flush. Because of that we end up flushing one sstable per index.
On systems on HDD this can take minutes for nothing.
> Thw following patch allows to avoid creating useless new sstables if the index was already
marked as built and will greatly reduce the startup time (and improve availability during
restarts).



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message