[
https://issues.apache.org/jira/browse/HBASE-23112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16943274#comment-16943274
]
Daisuke Kobayashi commented on HBASE-23112:
-------------------------------------------
Thanks for raising this Stack! Yea, it'd be nice to have better explanation
regarding that change. In addition to that, given that the current HBCK2 has
both porcelain and plumbing commands, should we rethink, or reword, the
following philosophy too?
{code}
HBCK2 performs a single discrete 'fix' task each time it is run. It does not
presume a tool can analyze all about the running cluster and then repair 'all
problems' found as hbck1 used suggest. HBCK2 is a tool that is more in the vein
of plumbing than porcelain.
{code}
> [hbase-operator-tools] fixMeta in hbck2 is porcelain, in hbck1 it was
> plumbing; fix
> -----------------------------------------------------------------------------------
>
> Key: HBASE-23112
> URL: https://issues.apache.org/jira/browse/HBASE-23112
> Project: HBase
> Issue Type: Bug
> Components: hbase-operator-tools
> Reporter: Michael Stack
> Priority: Major
>
> From [~daisuke.kobayashi]:
> {code}
> Daisuke Kobayashi 6:45 PM
> ... One confusion here is fixMeta in hbck1 was a plumbing command whereas the
> new fixMeta in hbck2 is a porcelain command which fixes several issues behind
> the scene. Wonder if it's a good naming. Sorry for bothering you!
> if you don't wanna change the command name, it'd be better to have a clear
> documentation.
> {code}
> How about we add note to usage that makes it clear it is not same as hbck1
> and in the README where we talk about difference between hbck1 and hbck2, we
> talk up this command in particular; that users need to be careful expecting
> hbck1 behaviors even in case where commands are similarly named?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
