impala-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dan Hecht (Code Review)" <ger...@cloudera.org>
Subject [Impala-CR](cdh5-trunk) IMPALA-3385: Fix crashes on accessing error_log
Date Fri, 22 Apr 2016 17:55:21 GMT
Dan Hecht has posted comments on this change.

Change subject: IMPALA-3385: Fix crashes on accessing error_log
......................................................................


Patch Set 5:

(4 comments)

http://gerrit.cloudera.org:8080/#/c/2829/5/be/src/util/error-util.cc
File be/src/util/error-util.cc:

Line 135:     if (v.second.count == 0) continue;
> right.. this was a left over of my previous change, where I changed appende
this isn't fixed in your new patchset


Line 159:     if (v.second.count == 0) continue;
> I will make it more general for readability..
This wasn't fixed in your new patchest.  did you upload the correct patch?


Line 167:         (*left)[v.first].count += v.second.count;
> left will never be cleared. This wont be called on per fragment's runtime s
This code shouldn't be making that assumption. let's make it work for any valid ErrorLogMap


Line 193:       errors.find(TErrorCode::GENERAL)->second.messages.size() - 1 : 0;
> Done. If we can avoid using ErrorCount than this wont be a problem.
This didn't change in your new patchset even though you say "Done".

Anyway, yes, I'm okay with keeping a separate count, but let's have it keep a running count
the number of distinct errors, which is what ErrorCount was trying to do.


-- 
To view, visit http://gerrit.cloudera.org:8080/2829
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I3a7e3d22e26147ada780aae5aed1f2e25a515afc
Gerrit-PatchSet: 5
Gerrit-Project: Impala
Gerrit-Branch: cdh5-trunk
Gerrit-Owner: Huaisi Xu <hxu@cloudera.com>
Gerrit-Reviewer: Bharath Vissapragada <bharathv@cloudera.com>
Gerrit-Reviewer: Dan Hecht <dhecht@cloudera.com>
Gerrit-Reviewer: Huaisi Xu <hxu@cloudera.com>
Gerrit-HasComments: Yes

Mime
View raw message