You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Brad Schoening (Jira)" <ji...@apache.org> on 2025/04/07 21:23:00 UTC
[jira] [Comment Edited] (CASSANDRA-13000) slow query log analysis tool
[ https://issues.apache.org/jira/browse/CASSANDRA-13000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17941720#comment-17941720 ]
Brad Schoening edited comment on CASSANDRA-13000 at 4/7/25 9:22 PM:
--------------------------------------------------------------------
[~pedro_gordo] [~smiklosovic]
I reviewed Pedro's recent changes to the PR and while I appreciate the effort he has put into resolving this, I have general concerns about the csqldumpslow script:
(1) does this add value over simply using 'grep' on the CLI that justifies the code maintenance going forward?
(2) if so, would it be better if it was implemented as a virtual table?
(3) without unit tests, how will this be kept in-sync with any format changes to slow query logging?
Both [CASSANDRA-13000|sues.apache.org/jira/browse/CASSANDRA-13000] and Datastax's [node_slow_log|https://docs.datastax.com/en/opscenter/6.8/online-help/services/configuring-slow-query-log.html] table seem preferable approaches.
was (Author: bschoeni):
[~pedro_gordo] [~smiklosovic]
I reviewed Pedro's recent changes to the PR and while I appreciate the effort he has put into resolving this, I have general concerns about the csqldumpslow script:
(1) does this add value over simply using 'grep' on the CLI that justifies the code maintenance going forward?
(2) if so, would it be better if it was implemented as a virtual table?
(3) without unit tests, how will this be kept in-sync with any format changes to slow query logging?
Both [CASSANDRA-13000|sues.apache.org/jira/browse/CASSANDRA-13000] and Datastax's [{{{}node_slow_log{}}}|[https://docs.datastax.com/en/opscenter/6.8/online-help/services/configuring-slow-query-log.html]] table seem preferable approaches.
> slow query log analysis tool
> ----------------------------
>
> Key: CASSANDRA-13000
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13000
> Project: Apache Cassandra
> Issue Type: New Feature
> Components: Legacy/Observability
> Reporter: Jon Haddad
> Assignee: Pedro Gordo
> Priority: Normal
> Fix For: 5.x
>
> Attachments: 0001-Adds-a-cqldumpslow-tool-which-analyses-the-debug-log.patch, 0001-Adds-a-cqldumpslow-tool-which-analyses-the-debug-log.patch, csqldumpslow.py, slow_queries_parser_output.txt
>
>
> As a follow up to CASSANDRA-12403, it would be very helpful to have a tool to process the slow queries that are logged. In the MySQL world, there's a tool called mysqldumpslow, which processes a slow query log, abstracts the parameters to prepared statements, and shows the queries which are causing problems based on frequency. The {{mysqldumpslow}} utillity shows an aggregated count & time statistics spent on slow queries. For instance:
> {code}shell> mysqldumpslow
> Reading mysql slow query log from /usr/local/mysql/data/mysqld51-apple-slow.log
> Count: 1 Time=4.32s (4s) Lock=0.00s (0s) Rows=0.0 (0), root[root]@localhost
> insert into t2 select * from t1
> Count: 3 Time=2.53s (7s) Lock=0.00s (0s) Rows=0.0 (0), root[root]@localhost
> insert into t2 select * from t1 limit N
> Count: 3 Time=2.13s (6s) Lock=0.00s (0s) Rows=0.0 (0), root[root]@localhost
> insert into t1 select * from t1{code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org