[
https://issues.apache.org/jira/browse/CASSANDRA-14527?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16516003#comment-16516003
]
Jaydeepkumar Chovatia commented on CASSANDRA-14527:
---------------------------------------------------
{quote}I like the idea of making operational issues more visible to the user.
But the intention to create a common framework for logging certain messages,
doesn't sound very convincing to me. At such point, I'm always asking myself,
what kind of problem we precisely want to solve. What's the value in adding
abstraction layers around producing log messages for very specific, predefined
conditions?
{quote}
Thanks for the review [[email protected]]. Here are some of the reason behind
this intention:
1. As described in the doc, CASSANDRA-12403 [large partition warning
|https://github.com/apache/cassandra/blob/trunk/src/java/org/apache/cassandra/io/sstable/format/big/BigTableWriter.java#L208]
[tombstone warning
|https://github.com/apache/cassandra/blob/trunk/src/java/org/apache/cassandra/db/ReadCommand.java#L490]
are already having some sort of this logic to inform user about some uncommon
behavior but they are very specific for certain type of problems and with many
limitations like changing thresholds require restart, user cannot consume in
different ways, etc. Also adding yet another type of problem will require
duplication of work.
2. Improve C* operational aspect
3. Having a common way of detecting/reporting often encourages people to add
more anti-patterns and reduces duplicate code
{quote}
I'd also recommend to first take any such ideas to the dev mailing list before
spending time implementing such changes. The Google docs hosted proposal gives
a nice overview on the ideas behind this and would have been a good way to
start a discussion and get some early feedback.
{quote}
I agree, will send this to dev mailing list.
> Real time Bad query logging framework
> -------------------------------------
>
> Key: CASSANDRA-14527
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14527
> Project: Cassandra
> Issue Type: New Feature
> Components: Observability
> Reporter: Jaydeepkumar Chovatia
> Assignee: Jaydeepkumar Chovatia
> Priority: Major
> Fix For: 4.x
>
>
> If Cassandra is not used in right way then it can create adverse effect to
> the application. There are lots of bad queries when using Cassandra, but
> major problem is end user don’t know where and what exactly is the problem.
> Most of the times end user is ready to take actions on bad queries provided
> Cassandra gives all detailed which could have potential impact on cluster
> performance.
> There has been already lots of work done as part of CASSANDRA-12403, proposal
> as part of this JIRA is let’s have some common way of detecting and logging
> different problems in Cassandra cluster which could have potential impact on
> Cassandra cluster performance.
> Please visit this document which has details like what is currently
> available, motivation behind developing this common framework, architecture,
> samples, etc.
> [https://docs.google.com/document/d/1D0HNjC3a7gnuKnR_iDXLI5mvn1zQxtV7tloMaLYIENE/edit?usp=sharing]
> Here is the patch with this feature:
> ||trunk||
> |[!https://circleci.com/gh/jaydeepkumar1984/cassandra/tree/bqr.svg?style=svg!
> |https://circleci.com/gh/jaydeepkumar1984/cassandra/82]|
> |[patch
> |https://github.com/apache/cassandra/compare/trunk...jaydeepkumar1984:bqr]|
> Please review this doc and the patch, and provide your opinion and feedback
> about this effort.
> Thank you!
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
