| Message ID | 20141231183013.GA3637@ret.masoncoding.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
Chris, you rule, I applied this to 3.16.7 and my problem went away. Tested-By: Marc MERLIN <marc@merlins.org> Happy new year! :) Marc On Wed, Dec 31, 2014 at 01:30:13PM -0500, Chris Mason wrote: > Commit 1d52c78afbb (Btrfs: try not to ENOSPC on log replay) added a > check to skip delayed inode updates during log replay because it > confuses the enospc code. But the delayed processing will end up > skipping delayed refs from log replay because the inode itself wasn't > put through the delayed code. > > This can end up triggering a warning at commit time: > > WARNING: CPU: 2 PID: 778 at fs/btrfs/delayed-inode.c:1410 btrfs_assert_delayed_root_empty+0x32/0x34() > > Which is repeated for each commit because we never process the delayed ref. > > The fix used here is to change btrfs_delayed_delete_inode_ref to return > an error if we're current in log replay. The caller will do the ref > deletion immediately and everything will work properly. > > This bug can cause lost files, whick fsck will find. --repair on > btrfs-progs 3.18 will fix them > > Signed-off-by: Chris Mason <clm@fb.com> > cc: stable@vger.kernel.org # v3.18 and any stable series that picked 1d52c78afbbf80b58299e076a159617d6b42fe3c > > diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c > index 054577b..de4e70f 100644 > --- a/fs/btrfs/delayed-inode.c > +++ b/fs/btrfs/delayed-inode.c > @@ -1857,6 +1857,14 @@ int btrfs_delayed_delete_inode_ref(struct inode *inode) > { > struct btrfs_delayed_node *delayed_node; > > + /* > + * we don't do delayed inode updates during log recovery because it > + * leads to enospc problems. This means we also can't do > + * delayed inode refs > + */ > + if (BTRFS_I(inode)->root->fs_info->log_root_recovering) > + return -EAGAIN; > + > delayed_node = btrfs_get_or_create_delayed_node(inode); > if (IS_ERR(delayed_node)) > return PTR_ERR(delayed_node); > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html >
Great! The root cause of missing file is fixed! Thanks a lot! Qu -------- Original Message -------- Subject: Re: [PATCH] Btrfs: don't delay inode ref updates during log replay From: Marc MERLIN <marc@merlins.org> To: Chris Mason <clm@fb.com>, Qu Wenruo <quwenruo@cn.fujitsu.com>, Roman Mamedov <rm@romanrm.net>, Btrfs BTRFS <linux-btrfs@vger.kernel.org> Date: 2015?01?02? 06:58 > Chris, you rule, I applied this to 3.16.7 and my problem went away. > > Tested-By: Marc MERLIN <marc@merlins.org> > > Happy new year! :) > > Marc > > On Wed, Dec 31, 2014 at 01:30:13PM -0500, Chris Mason wrote: >> Commit 1d52c78afbb (Btrfs: try not to ENOSPC on log replay) added a >> check to skip delayed inode updates during log replay because it >> confuses the enospc code. But the delayed processing will end up >> skipping delayed refs from log replay because the inode itself wasn't >> put through the delayed code. >> >> This can end up triggering a warning at commit time: >> >> WARNING: CPU: 2 PID: 778 at fs/btrfs/delayed-inode.c:1410 btrfs_assert_delayed_root_empty+0x32/0x34() >> >> Which is repeated for each commit because we never process the delayed ref. >> >> The fix used here is to change btrfs_delayed_delete_inode_ref to return >> an error if we're current in log replay. The caller will do the ref >> deletion immediately and everything will work properly. >> >> This bug can cause lost files, whick fsck will find. --repair on >> btrfs-progs 3.18 will fix them >> >> Signed-off-by: Chris Mason <clm@fb.com> >> cc: stable@vger.kernel.org # v3.18 and any stable series that picked 1d52c78afbbf80b58299e076a159617d6b42fe3c >> >> diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c >> index 054577b..de4e70f 100644 >> --- a/fs/btrfs/delayed-inode.c >> +++ b/fs/btrfs/delayed-inode.c >> @@ -1857,6 +1857,14 @@ int btrfs_delayed_delete_inode_ref(struct inode *inode) >> { >> struct btrfs_delayed_node *delayed_node; >> >> + /* >> + * we don't do delayed inode updates during log recovery because it >> + * leads to enospc problems. This means we also can't do >> + * delayed inode refs >> + */ >> + if (BTRFS_I(inode)->root->fs_info->log_root_recovering) >> + return -EAGAIN; >> + >> delayed_node = btrfs_get_or_create_delayed_node(inode); >> if (IS_ERR(delayed_node)) >> return PTR_ERR(delayed_node); >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html >> -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 054577b..de4e70f 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1857,6 +1857,14 @@ int btrfs_delayed_delete_inode_ref(struct inode *inode) { struct btrfs_delayed_node *delayed_node; + /* + * we don't do delayed inode updates during log recovery because it + * leads to enospc problems. This means we also can't do + * delayed inode refs + */ + if (BTRFS_I(inode)->root->fs_info->log_root_recovering) + return -EAGAIN; + delayed_node = btrfs_get_or_create_delayed_node(inode); if (IS_ERR(delayed_node)) return PTR_ERR(delayed_node);
Commit 1d52c78afbb (Btrfs: try not to ENOSPC on log replay) added a check to skip delayed inode updates during log replay because it confuses the enospc code. But the delayed processing will end up skipping delayed refs from log replay because the inode itself wasn't put through the delayed code. This can end up triggering a warning at commit time: WARNING: CPU: 2 PID: 778 at fs/btrfs/delayed-inode.c:1410 btrfs_assert_delayed_root_empty+0x32/0x34() Which is repeated for each commit because we never process the delayed ref. The fix used here is to change btrfs_delayed_delete_inode_ref to return an error if we're current in log replay. The caller will do the ref deletion immediately and everything will work properly. This bug can cause lost files, whick fsck will find. --repair on btrfs-progs 3.18 will fix them Signed-off-by: Chris Mason <clm@fb.com> cc: stable@vger.kernel.org # v3.18 and any stable series that picked 1d52c78afbbf80b58299e076a159617d6b42fe3c -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html