Alexey Serbin has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/18101 )

Change subject: [consensus] force fsync metadata on XFS
......................................................................


Patch Set 2:

(4 comments)

http://gerrit.cloudera.org:8080/#/c/18101/2/src/kudu/consensus/consensus_meta.cc
File src/kudu/consensus/consensus_meta.cc:

http://gerrit.cloudera.org:8080/#/c/18101/2/src/kudu/consensus/consensus_meta.cc@57
PS2, Line 57: a system crash.
nit: does it make sense to mention KUDU-2195 for reference?


http://gerrit.cloudera.org:8080/#/c/18101/2/src/kudu/consensus/consensus_meta.cc@58
PS2, Line 58: unsafe
Do we really want to be this strict?  Maybe, tagging it as 'experimental' could 
be a better choice here?


http://gerrit.cloudera.org:8080/#/c/18101/2/src/kudu/consensus/consensus_meta.cc@58
PS2, Line 58: TAG_FLAG(cmeta_fsync_override_on_xfs, unsafe);
Does it make sense to add the 'runtime' tag as well?


http://gerrit.cloudera.org:8080/#/c/18101/2/src/kudu/fs/fs_manager.cc
File src/kudu/fs/fs_manager.cc:

http://gerrit.cloudera.org:8080/#/c/18101/2/src/kudu/fs/fs_manager.cc@365
PS2, Line 365:   RETURN_NOT_OK_PREPEND(env_->IsOnXfsFilesystem(
             :       canonicalized_metadata_fs_root_.path, &meta_on_xfs_),
             :       Substitute("Could not determine file system of metadata 
directory $0",
             :                  canonicalized_metadata_fs_root_.path));
Is it possible to turn this into a warning instead of an error, keeping 
'meta_on_xfs_' to be false in case of it was not possible to determine the type 
of the FS that backs the metadata directory?



--
To view, visit http://gerrit.cloudera.org:8080/18101
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ie97138e3522f20325be104d3665d887693b79a10
Gerrit-Change-Number: 18101
Gerrit-PatchSet: 2
Gerrit-Owner: Andrew Wong <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Bankim Bhavsar <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Yingchun Lai <[email protected]>
Gerrit-Comment-Date: Wed, 15 Dec 2021 05:48:43 +0000
Gerrit-HasComments: Yes

Reply via email to