New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
jewel: mds: enforce a dirfrag limit on entries #10104
Conversation
This commit adds a new config option mds_bal_fragment_size_max = 10000*10 which is an order of magnitude larger than mds_bal_split_size. This limit prevents a fragment from getting too large which results in large omap directories. Right now the limit is enforced only in the RPC paths and in stray directory entry creation. Fixes http://tracker.ceph.com/issues/16164 Signed-off-by: Patrick Donnelly <pdonnell@redhat.com> (cherry picked from commit 60af83c)
e39312f
to
d244b7a
Compare
@xiaoxichen I'm setting milestone jewel since this is targeting jewel really. If that's a mistake please let me know. |
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
@jcsp does this backport look good to merge ? It passed the fs suite http://tracker.ceph.com/issues/16344#note-21 (except for unrelated valgrind failures and one transient failure that is, I believe, unrelated ). |
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
…ries Reviewed-by: Loic Dachary <ldachary@redhat.com>
This looks okay to me on a skim but hopefully Patrick can sign-off on this since John's on vacation another week. |
@batrick ^ |
|
http://tracker.ceph.com/issues/16560