Ashwani Raina has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/20792 )

Change subject: [compaction] Log timestamp of two matching DELETE REDO 
mutations.
......................................................................


Patch Set 1:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/20792/1/src/kudu/tablet/compaction.cc
File src/kudu/tablet/compaction.cc:

http://gerrit.cloudera.org:8080/#/c/20792/1/src/kudu/tablet/compaction.cc@479
PS1, Line 479:     if (0 == ret) {
             :       // Log the timestamp value, invalid value could mean 
potential corruption.
             :       LOG(INFO) << Substitute("Both DELETE REDO mutations are 
stamped with same timestamp i.e. $0",
             :                               left_last->timestamp().ToString());
             :     }
> I guess it's not safe to continue from here having the invariant broken.
At first, I thought of using RowToString or StringifyMutationList to print 
timestamp and changelist for all mutations. It would probably be a futile 
exercise since one problematic entry vs many problematic entries would mostly 
require same workaround I guess.

But there is no harm in printing those mutations. I am not sure how logging 
handles long list of mutations. It should probably be ok as I can see it being 
used at couple of places where multiple mutation entries are logged.



--
To view, visit http://gerrit.cloudera.org:8080/20792
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I508254a83046818b81db4577bf07265b46a13c9a
Gerrit-Change-Number: 20792
Gerrit-PatchSet: 1
Gerrit-Owner: Ashwani Raina <[email protected]>
Gerrit-Reviewer: Abhishek Chennaka <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Ashwani Raina <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Comment-Date: Thu, 14 Dec 2023 20:08:51 +0000
Gerrit-HasComments: Yes

Reply via email to