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

Remove "experimental" warnings from multimds #15154

merged 13 commits into from Jun 23, 2017


None yet
2 participants

jcsp commented May 18, 2017

No description provided.

ukernel and others added some commits Jun 3, 2017

mds: fix stray dentry replication in cache rejoin ack
To replicate s stray dentry, we need to replicate all its ancestors.

Signed-off-by: "Yan, Zheng" <>
mds: create subtree root immediately after directory tree becomes frozen
When a directory tree become frozen, its WAIT_FROZEN contexts are
executed asynchronously. Before Migrator::export_frozen() set export
bounds, MDCache::try_subtree_merge_at() can merge newly imported
subtree into the frozen directory tree. This causes problem if there
are auth pins in newly imported subtree.

The fix is creating subtree root immediately after directory tree
becomes frozen. The new subtree root has dir_auth 'me, me', so it's
not meregeable.

Signed-off-by: "Yan, Zheng" <>
mds: don't mark nestlock dirty on improper inode
If inode is replica and it has no auth subtree dirfrag, We should
not mark its nestlock dirty.

Signed-off-by: "Yan, Zheng" <>
mds: avoid submitting log entry while adjusting subtree map
MDCache::eval_subtree_root() may tigger scatter-gather process, which
submits log entry. Submitting log entry while adjusting subtree map is
bad, because subtree map in intermediate state may get used/logged.

Signed-off-by: "Yan, Zheng" <>
mds: don't forge replica dirfrag
MDCache::forge_replica_dir() set wrong dir_auth if the forged replica
dirfrag is subtree root.

Signed-off-by: "Yan, Zheng" <>
mon: remove experimental warning on multimds
Signed-off-by: John Spray <>
mds: handle MDirUpdate race
mds may try discover several times for MDirUpdate, rename may kick
in and cause MDCache::path_traverse() to return error.

Signed-off-by: "Yan, Zheng" <>
mds: drop locks before waiting for export targets
Signed-off-by: "Yan, Zheng" <>
mds: don't call StrayManager::eval_stray() for undefined inode
Signed-off-by: "Yan, Zheng" <>
mds: enable multimds by default in new filesystems
Signed-off-by: John Spray <>
qa: no longer need to explicitly enable multimds
Signed-off-by: John Spray <>
doc: multimds is no longer experimental
Signed-off-by: John Spray <>
qa: update cephtool test for multimds on by default
Signed-off-by: John Spray <>

@jcsp jcsp changed the title from [DNM] Remove "experimental" warnings from multimds to Remove "experimental" warnings from multimds Jun 23, 2017

@jcsp jcsp merged commit 45b1a10 into ceph:master Jun 23, 2017

2 of 4 checks passed

arm64 make check arm64 make check triggered
make check running make check
Signed-off-by all commits in this PR are signed
Unmodified Submodules submodules for project are unmodified

@jcsp jcsp deleted the jcsp:wip-multimds-stable branch Jun 23, 2017

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment