[LU-903] Race condition while get_attr after cancel_lru_locks and sysctl drop_caches

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.4.0
• Affects Version/s: Lustre 2.4.0, Lustre 1.8.6
• Labels: dentry patch
• Environment: SLES11
• Severity: 3
• Bugzilla ID: 24,555

Epic

metadata

• Assignee: Keith Mannthey (Inactive)
• Reporter: Artem Blagodarenko (Inactive)

Description

Reproduction script is described in https://bugzilla.lustre.org/show_bug.cgi?id=24555
After some analysis the next bug picture drown:

1.
First thread makes lookup. Gets CR lock and terminates.
After that another thread a) make clear_lru cache b) sysctl that flush slab and another kernel
caches (dcache, icache, etc ...)

This results that sequence "shrink_dcache_memory -> "foreach_dentry_lru" -> prune_one_dentry ->
d_kill > d_iput()" is executed. After that ll_clear_inode executed that NULLed lock>l_ast_data.

2.
Some time after 1. another thread make get_attr on same inode. Gets another IBIT lock, but LOOKUP +
UPDATE.
Another client need cancel this lock, but 2 BL AST race arised. Second lock can't cancel first
lock because optimization, that was sown bellow and fist lock can't be canceled because its
inode == NULL.

Optimisation:

int ll_mdc_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc,
                        void *data, int flag)
...
               if ((bits & MDS_INODELOCK_LOOKUP) &&
                    ll_have_md_lock(inode, MDS_INODELOCK_LOOKUP, LCK_MINMODE))
                        bits &= ~MDS_INODELOCK_LOOKUP;
                if ((bits & MDS_INODELOCK_UPDATE) &&
                    ll_have_md_lock(inode, MDS_INODELOCK_UPDATE, LCK_MINMODE))
                        bits &= ~MDS_INODELOCK_UPDATE;
                if ((bits & MDS_INODELOCK_OPEN) &&
                    ll_have_md_lock(inode, MDS_INODELOCK_OPEN, mode))
                        bits &= ~MDS_INODELOCK_OPEN;
...
if (inode->i_sb->s_root &&
        inode != inode->i_sb->s_root->d_inode &&
        (bits & MDS_INODELOCK_LOOKUP))
        ll_unhash_aliases(inode);
iput(inode);

Issue Links

• Trackbacks: Lustre 1.8.x known issues tracker While testing against Lustre b18 branch, we would hit known bugs which were already reported in Lustre Bugzilla https://bugzilla.lustre.org/. In order to move away from relying on Bugzilla, we would create a JIRA

Comments

[Peter Jones]

Landed for 2.4

[Cory Spitz]

FYI, Cray has been using this patch for nearly a year.

[Nathan Rutman]

Xyratex-bug-id: MRP-269
Xyratex-bug-id: MRP-363

[Keith Mannthey (Inactive)]

I have reviewed the latest version of the patch.

[Artem Blagodarenko (Inactive)]

The patch is reviewed. Is waiting for landing.

[Artem Blagodarenko (Inactive)]

Patch for master is uploaded http://review.whamcloud.com/2627

[Andreas Dilger]

I'm all in favour of moving the inode reference onto the resource instead of on the lock. This patch should be landed to master first, not b1_8. Please submit a version of this patch against the master branch.