multithreaded promotion/demotion in tiering #53
Labels
Comments
dlambrig
changed the title
dlambrig
changed the title
|
@dlambrig the feature cannot make it to 3.10, but is still a valid feature to track for future releases, am I right? Reopening the issue based on the above assumption, and removing the 3.10 release milestone from the same. It will still reside as a part of the tier backlog though. |
|
tier backlog is the right place to track it. |
gluster-ant
pushed a commit
that referenced
this issue
Aug 20, 2018
…re.t Problem: In line #13 of the test case, it checks whether the file is present on first 2 bricks or not. If it is not present on even one of the bricks it will break the loop and checks for the dirty marking on the parent on the 3rd brick and checks for file not present on the 1st and 2nd bricks. The below scenario can happen in this case: - File gets created on 1st and 3rd bricks - In line #13 it sees file is not present on both 1st & 2nd bricks and breaks the loop - In line #51 test fails because the file will be present on the 1st brick - In line #53 test will fail because the file creation was not failed on quorum bricks and dirty marking will not be there on the parent on 3rd brick Fix: Don't break from the loop if file is present on either brick 1 or brick 2. Change-Id: I918068165e4b9124c1de86cfb373801b5b432bd9 fixes: bz#1612054 Signed-off-by: karthik-us <ksubrahm@redhat.com>
amarts
pushed a commit
to amarts/glusterfs_fork
that referenced
this issue
Sep 11, 2018
…re.t Problem: In line gluster#13 of the test case, it checks whether the file is present on first 2 bricks or not. If it is not present on even one of the bricks it will break the loop and checks for the dirty marking on the parent on the 3rd brick and checks for file not present on the 1st and 2nd bricks. The below scenario can happen in this case: - File gets created on 1st and 3rd bricks - In line gluster#13 it sees file is not present on both 1st & 2nd bricks and breaks the loop - In line gluster#51 test fails because the file will be present on the 1st brick - In line gluster#53 test will fail because the file creation was not failed on quorum bricks and dirty marking will not be there on the parent on 3rd brick Fix: Don't break from the loop if file is present on either brick 1 or brick 2. Change-Id: I918068165e4b9124c1de86cfb373801b5b432bd9 fixes: bz#1612054 Signed-off-by: karthik-us <ksubrahm@redhat.com>
|
tier is removed from the build from glusterfs-6. If reopen if anyone is interested to work. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Tiering currently promotes or demotes a single file at a time. The multithreaded code used in DHT rebalance is not used. This makes tier migration of files very slow.
This project will change the tiering migration code to use a thread pool, to move multiple files at a time. DHT multithreaded code shall be leveraged.
Draft description of implementation:
https://docs.google.com/document/d/10pvgU9uYINv0pXjGgIUNGATpoh93b4djExmyWG5Atn8/edit
The text was updated successfully, but these errors were encountered: