Yingyi Bu has posted comments on this change. Change subject: Ensure Metadata locks are acquired for SQL++ queries ......................................................................
Patch Set 12: (1 comment) https://asterix-gerrit.ics.uci.edu/#/c/1642/12/asterixdb/asterix-metadata/src/main/java/org/apache/asterix/metadata/lock/LockList.java File asterixdb/asterix-metadata/src/main/java/org/apache/asterix/metadata/lock/LockList.java: PS12, Line 28: ArrayList > Yingyi, I looked into this and it is not straight forward. For once, an IMe During query rewriting, we can run a rewrite rule many times until nothing changes further, because we use the fixpoint rule controller for many rules. Those iterative rules can call metadataProvider.findDataset(...) (and they do), which could potentially lead to memory issue here. >> For once, an IMetadataLock doesn't know what it locks. I see. Adding a string reference to MetadataLock could fix it, but that requires us to unify dataversesLocks, datasetLocks, functionsLocks etc and maybe to use prefix strings to differentiate them. If we merge this now, we should probably open an issue. -- 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: 12 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: Yes
