[LU-1234] Executing binary stored on Lustre results in " (deleted)" appended to /proc/self/exec

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Severity: 3
• Assignee: Lai Siyao
• Reporter: Prakash Surya (Inactive)

Description

We have reports that when running an executable stored on Lustre, the symlink /proc/self/exec will have " (deleted)" appended to it. This can be normal behavior under Linux, if the executable has in fact been overwritten or deleted. But I have been told this is not the case, and I would not be surprised if there is a Lustre bug here. Digging into the kernel sources, it appears the following snippets are relevant:

__d_path

 if (d_unlinked(dentry) &&
         (prepend(&end, &buflen, " (deleted)", 10) != 0))
                 goto Elong;

static inline int d_unhashed(struct dentry *dentry)
{
        return (dentry->d_flags & DCACHE_UNHASHED);
}

static inline int d_unlinked(struct dentry *dentry)
{
        return d_unhashed(dentry) && !IS_ROOT(dentry);
}

So, the string " (deleted)" is appended when the dentry in question has the DCACHE_UNHASHED flag set. Perhaps there's a situation where Lustre fails to take a reference on the dentry, allowing it to be dropped from the cache when it shouldn't be?

Also note, I haven't been able to reproduce this issue so it's been difficult to debug, or even determine exactly where the issue lies.

Comments

[Peter Jones]

Thanks Chris!

[Christopher Morrone (Inactive)]

Patch landed on b2_1. I assume this can be closed.

[Christopher Morrone (Inactive)]

Patch 2400 is getting stale, and conflicts on b2_1 are growing. Can you please rebase patch 2400 onto 2.1.4?

[Christopher Morrone (Inactive)]

Patch for 2.1 is still in limbo.

[Christopher Morrone (Inactive)]

LU-506 closed without this landing on b2_1. The b2_1 patch has two +1 reviews, but is just sitting. We are running the patch in production at LLNL on our 2.1 systems.

[Peter Jones]

Thanks Chris

[Christopher Morrone (Inactive)]

Sure.

[Peter Jones]

Chris

We will track landing this patch under LU-506. I gather that you are running this patch in production now so is it ok to close this as a duplicate of LU-506?

Peter

[Christopher Morrone (Inactive)]

Someone needs to attend to the Maloo failure.

[Build Master (Inactive)]

Integrated in lustre-reviews » x86_64,server,el6,inkernel #4594
LU-1234 dcache: don't drop invalid dentry arbitrarily (Revision c7619a3c9c0b5ddc040612169402f6ab75bb3a49)

Result = SUCCESS
Lai Siyao : c7619a3c9c0b5ddc040612169402f6ab75bb3a49
Files :

• lustre/llite/namei.c
• lustre/llite/llite_internal.h
• lustre/include/linux/lustre_patchless_compat.h
• lustre/llite/statahead.c
• lustre/include/linux/lustre_compat25.h
• lustre/llite/dcache.c
• lustre/llite/file.c