David Ribeiro Alves has posted comments on this change.

Change subject: KUDU-1056 and KUDU-1020 Safe time for ksck checksum
......................................................................


Patch Set 2:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/6843/2/src/kudu/tools/ksck_remote-test.cc
File src/kudu/tools/ksck_remote-test.cc:

PS2, Line 154: a chance establish a timestamp.
> Without this change the test fails regularly with a message like:
I see what you mean, my comment was more about verbage.
Are you sure that you might get that number from 2 replicas, not just 1?
That lagging amount is interesting though, seems like the replica hasn't 
updated the safe time ever. Might actually be a bug.

Don't want to delay this though, so regarding verbage how about:
"Wait for the first 100 writes so be committed so that there is a very high 
chance that all replicas have committed at least one message in each tablet, 
otherwise safe time might not have been and replicas might refuse snapshot 
scans because of lag"

os something like that


-- 
To view, visit http://gerrit.cloudera.org:8080/6843
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Ib45be20dcfa37fb85185302adf84d2c4a55f8c1e
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Will Berkeley <[email protected]>
Gerrit-Reviewer: David Ribeiro Alves <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Will Berkeley <[email protected]>
Gerrit-HasComments: Yes

Reply via email to