abdullah alamoudi has posted comments on this change.

Change subject: Ensure Metadata locks are acquired for SQL++ queries
......................................................................


Patch Set 14:

> LGTM.  I couldn't think of a deadlock case so far ( (A) no DDL that
 > write locks two instances of following things, dataverses,
 > datasets, functions, nodegroups, feeds, feeds-policy,
 > compaction-policy, datatypes, extensions -- so at most 1 write lock
 > for a DDL,  (B) the (only one) write lock is the last one to
 > acquire), but I give Till a chance to review it by +1:-)

Exactly, the write lock is always the last.
We assume that no additional read locks will be acquired after that for those 
statements (and they shouldn't), otherwise, there is a bug.

-- 
To view, visit https://asterix-gerrit.ics.uci.edu/1642
To unsubscribe, visit https://asterix-gerrit.ics.uci.edu/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I5f468599897a37cbcb12d8577d072f340f0d949c
Gerrit-PatchSet: 14
Gerrit-Project: asterixdb
Gerrit-Branch: master
Gerrit-Owner: abdullah alamoudi <[email protected]>
Gerrit-Reviewer: Jenkins <[email protected]>
Gerrit-Reviewer: Till Westmann <[email protected]>
Gerrit-Reviewer: Yingyi Bu <[email protected]>
Gerrit-Reviewer: abdullah alamoudi <[email protected]>
Gerrit-HasComments: No

Reply via email to