New: inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

Bug 201823 [email protected] 3 participants 12 messages Dec 1, 2018 – Dec 3, 2018

Share & tools

[email protected]

author

2018-12-01 19:28 UTC #0 lore raw

https://bugzilla.kernel.org/show_bug.cgi?id=201823

            Bug ID: 201823
           Summary: inconsistent behaviour of xfs_db, xfs_repair and
                    xfs_scrub
           Product: File System
           Version: 2.5
    Kernel Version: 4.19.5
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: XFS
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Hi,

I am using xfs on Manjaro with kernel 4.19 and latest xfsprogs 4.19.0-1.

I checked my root filesystem with xfs_db for fragmentation and got metadata corruption messages:

5# xfs_db -c frag -r /dev/sde1
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x7faac0/0x8000 Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x2341340/0x8000
aktuell 976616, ideal 970547, Unterteilungsfaktor 0,62%
Note, this number is largely meaningless.
Files on this filesystem average 1,01 extents per file

But xfs_repiar is not mentioning anything and hence not repairing anything. Also, xfs_scrub is not complaining:

8# xfs_scrub -v /
EXPERIMENTAL xfs_scrub program in use! Use at your own risk! Phase 1: Find filesystem geometry.
/: using 8 threads to scrub.
Phase 2: Check internal metadata.
Phase 3: Scan all inodes.
Phase 5: Check directory tree.
Phase 7: Check summary counters.
41,6GiB data used; 1,3M inodes used.
41,5GiB data found; 1,3M inodes found.
1,3M inodes counted; 1,3M inodes checked.

How do I have to interpret these results?
What can I do to get repair it?

I created an xfs dump just in case somebody wants to debug this.

Thank you for your help
Matthias

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-02 06:26 UTC #1 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #1 from [email protected] ---
I see that inode 0x234134 has a bad v3.crc:

xfs_db> inode 0x2341340
Metadata corruption detected at 0x56047ae6a3f3, xfs_inode block 0x1bab340/0x4000
Metadata CRC error detected for ino 36967232
xfs_db> p
core.magic = 0x3132
core.mode = 033062
core.version = 32
core.format = 56
core.nlinkv2 = 775108409
core.onlink = 14649
core.projid_lo = 12597
core.projid_hi = 8268
core.uid = 775371058
core.gid = 925645104
core.flushiter = 13361
core.atime.sec = Wed May 24 22:18:58 2000 core.atime.nsec = 775173426
core.mtime.sec = Thu Aug 17 12:56:03 1995 core.mtime.nsec = 540555575
core.ctime.sec = Fri Jul 29 00:54:45 1994 core.ctime.nsec = 875313459
core.size = 3328777196389539896
core.nblocks = 4123096260573147948
core.extsize = 959655481
core.nextents = 942880825
core.naextents = 8248
core.forkoff = 57
core.aformat = 57
core.dmevmask = 0x2e343530
core.dmstate = 13612
core.newrtbm = 0
core.prealloc = 1
core.realtime = 1
core.immutable = 0
core.append = 1
core.sync = 1
core.noatime = 0
core.nodump = 0
core.rtinherit = 1
core.projinherit = 0
core.nosymlinks = 0
core.extsz = 1
core.extszinherit = 1
core.nodefrag = 1
core.filestream = 0
core.gen = 775371064
next_unlinked = 959717452
v3.crc = 0x20393031 (bad)
v3.change_count = 3330188947660289070
v3.lsn = 0x3130373037204320
v3.flags2 = 0x3930322e33333631
v3.cowextsize = 741946158
v3.crtime.sec = Wed Mar 26 15:01:29 1997 v3.crtime.nsec = 858665010
v3.inumber = 4123108269418885678
v3.uuid = 34333636-2c39-322e-3238-333338203930 v3.reflink = 0
v3.cowextsz = 0
u3 = (leer)
a = (leer)

I would like to know which file belongs to that inode before I try recalculate it. But "blockget -n" is always running out of memory.

Anyways, why isnt xfs_repair fixing this? Why isnt xfs_scrub showing this?

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-02 08:27 UTC #2 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #4 from [email protected] ---
The error occurs also with unmounted drive. Sorry for posting the output of the "xfs_db -r" command this is misleading.

The first thing I did is checking it with a KNOPPIX 8.2 live CD. xfs_db is showing the erros but xfs_repair does not repair anything.

Here the relevant outputs. /dev/sde1 is not mounted:

######

knoppix@Microknoppix:~$ uname -a
Linux Microknoppix 4.16.5-64 #1 SMP PREEMPT Sun Apr 29 17:09:15 CEST 2018
x86_64 GNU/Linux

######

knoppix@Microknoppix:~$ xfs_db /dev/sde1
xfs_db> frag
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
Metadata corruption detected at xfs_inode block 0x2341340/0x8000
aktuell 977134, ideal 970700, Unterteilungsfaktor 0,66%
Note, this number is largely meaningless.
Files on this filesystem average 1,01 extents per file
xfs_db> inode 0x2341340
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata corruption detected at xfs_inode block 0x1bab340/0x2000
Metadata CRC error detected for ino 36967232
xfs_db> p
core.magic = 0x3132
core.mode = 033062
core.version = 32
core.format = 56
core.nlinkv2 = 775108409
core.onlink = 14649
core.projid_lo = 12597
core.projid_hi = 8268
core.uid = 775371058
core.gid = 925645104
core.flushiter = 13361
core.atime.sec = Wed May 24 22:18:58 2000
core.atime.nsec = 775173426
core.mtime.sec = Thu Aug 17 12:56:03 1995
core.mtime.nsec = 540555575
core.ctime.sec = Fri Jul 29 00:54:45 1994
core.ctime.nsec = 875313459
core.size = 3328777196389539896
core.nblocks = 4123096260573147948
core.extsize = 959655481
core.nextents = 942880825
core.naextents = 8248
core.forkoff = 57
core.aformat = 57
core.dmevmask = 0x2e343530
core.dmstate = 13612
core.newrtbm = 0
core.prealloc = 1
core.realtime = 1
core.immutable = 0
core.append = 1
core.sync = 1
core.noatime = 0
core.nodump = 0
core.rtinherit = 1
core.projinherit = 0
core.nosymlinks = 0
core.extsz = 1
core.extszinherit = 1
core.nodefrag = 1
core.filestream = 0
core.gen = 775371064
next_unlinked = 959717452
v3.crc = 0x20393031 (bad)
v3.change_count = 3330188947660289070
v3.lsn = 0x3130373037204320
v3.flags2 = 0x3930322e33333631
v3.cowextsize = 741946158
v3.crtime.sec = Wed Mar 26 15:01:29 1997
v3.crtime.nsec = 858665010
v3.inumber = 4123108269418885678
v3.uuid = 34333636-2c39-322e-3238-333338203930
v3.reflink = 0
v3.cowextsz = 0
u3 = (leer)
a = (leer)
xfs_db> q

######

knoppix@Microknoppix:~$ xfs_repair -V
xfs_repair Version 4.9.0

######

knoppix@Microknoppix:~$ xfs_repair /dev/sde1
Phase 1 - Superblock finden und überprüfen...
Phase 2 - ein internes Protokoll benutzen
        - Null-Protokoll...
        - freier Speicher und Inode-Karten des Dateisystems werden
gescannt...
        - Wurzel-Inode-Stück gefunden
Phase 3 - für jedes AG...
        - agi unverknüpfte Listen werden gescannt und bereinigt...
        - bekannte Inodes werden behandelt und Inode-Entdeckung wird
durchgeführt...
        - agno = 0
        - agno = 1
        - agno = 2
        - agno = 3
        - neu entdeckte Inodes werden behandelt...
Phase 4 - auf doppelte Blöcke überprüfen...
        - Liste mit doppeltem Ausmaß wird eingerichtet...
        - es wird geprüft ob Inodes Blocks doppelt beanspruchen...
        - agno = 3
        - agno = 2
        - agno = 1
        - agno = 0
Phase 5 - AG-Köpfe und Bäume werden erneut gebildet...
        - Superblock wird zurückgesetzt...
Phase 6 - Inode-Verbindbarkeit wird geprüft...
        - Inhalte der Echtzeit-Bitmaps und Zusammenfassungs-Inodes werden
zurückgesetzt
        - Dateisystem wird durchquert ...
        - durchqueren beendet ...
        - nicht verbundene Inodes werden nach lost+found verschoben ...
Phase 7 - Verweisanzahl wird geprüft und berichtigt...
erledigt

######

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

Dave Chinner

2018-12-02 09:57 UTC #3 lore raw

In reply to [email protected]

Quoted 26 lines · On Sat, Dec 01, 2018 at 08:16:13AM +0000, [email protected]…

https://bugzilla.kernel.org/show_bug.cgi?id=201823

            Bug ID: 201823
           Summary: inconsistent behaviour of xfs_db, xfs_repair and
                    xfs_scrub
           Product: File System
           Version: 2.5
    Kernel Version: 4.19.5
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: XFS
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Hi,

I am using xfs on Manjaro with kernel 4.19 and latest xfsprogs 4.19.0-1.

I checked my root filesystem with xfs_db for fragmentation and got metadata corruption messages:

Quoted 9 lines

5# xfs_db -c frag -r /dev/sde1
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x7faac0/0x8000 Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x2341340/0x8000
aktuell 976616, ideal 970547, Unterteilungsfaktor 0,62%
Note, this number is largely meaningless.
Files on this filesystem average 1,01 extents per file

That is expected if you are running xfs_db on a mounted filesystem. What is on disk at any given instant is not consistent. The combined in-memory state, the journal and what is on-disk is consistent, but you cannot see all that extra in-memory state from xfs_db. If the system crashes, journal replay is what brings it back to being consistent on disk. IOWs, this is expected behaviour.

Quoted 1 line

But xfs_repiar is not mentioning anything and hence not repairing anything.

xfs_repair won't run on a mounted filesystem.

Quoted 13 lines

Also, xfs_scrub is not complaining:
8# xfs_scrub -v /
EXPERIMENTAL xfs_scrub program in use! Use at your own risk! Phase 1: Find filesystem geometry.
/: using 8 threads to scrub.
Phase 2: Check internal metadata.
Phase 3: Scan all inodes.
Phase 5: Check directory tree.
Phase 7: Check summary counters.
41,6GiB data used; 1,3M inodes used.
41,5GiB data found; 1,3M inodes found.
1,3M inodes counted; 1,3M inodes checked.

Scrub runs online, so sees the in-memory state of the filesystem along with what is on disk, and hence it doesn't find any corruptions because there are none.

IOWs, there are no problems and everything is behaving as expected.

-Dave.

Signature · 2 lines

Dave Chinner
[email protected]

[email protected]

author

2018-12-02 09:57 UTC #4 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #2 from Dave Chinner ([email protected]) ---

Quoted 26 lines · On Sat, Dec 01, 2018 at 08:16:13AM +0000, [email protected]…

https://bugzilla.kernel.org/show_bug.cgi?id=201823

            Bug ID: 201823
           Summary: inconsistent behaviour of xfs_db, xfs_repair and
                    xfs_scrub
           Product: File System
           Version: 2.5
    Kernel Version: 4.19.5
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: XFS
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Hi,

I am using xfs on Manjaro with kernel 4.19 and latest xfsprogs 4.19.0-1.

I checked my root filesystem with xfs_db for fragmentation and got metadata corruption messages:

Quoted 10 lines

5# xfs_db -c frag -r /dev/sde1
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x7faac0/0x8000
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x20cb300/0x8000
Metadata corruption detected at 0x558ed74133f3, xfs_inode block 0x2341340/0x8000
aktuell 976616, ideal 970547, Unterteilungsfaktor 0,62% Note, this number is largely meaningless.
Files on this filesystem average 1,01 extents per file

Quoted 1 line

But xfs_repiar is not mentioning anything and hence not repairing anything.

xfs_repair won't run on a mounted filesystem.

Quoted 13 lines

Also, xfs_scrub is not complaining:
8# xfs_scrub -v /
EXPERIMENTAL xfs_scrub program in use! Use at your own risk! Phase 1: Find filesystem geometry.
/: using 8 threads to scrub.
Phase 2: Check internal metadata.
Phase 3: Scan all inodes.
Phase 5: Check directory tree.
Phase 7: Check summary counters.
41,6GiB data used; 1,3M inodes used.
41,5GiB data found; 1,3M inodes found.
1,3M inodes counted; 1,3M inodes checked.

Scrub runs online, so sees the in-memory state of the filesystem along with what is on disk, and hence it doesn't find any corruptions because there are none.

IOWs, there are no problems and everything is behaving as expected.

-Dave.

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-02 12:37 UTC #5 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

Eric Sandeen ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #3 from Eric Sandeen ([email protected]) ---

Quoted 1 line

# xfs_db -c frag -r /dev/sde1

I think Dave's analysis is correct, but to be sure, can you confirm that this command was run while mounted?

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-02 17:02 UTC #6 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #5 from Eric Sandeen ([email protected]) ---
If you'd like to send me a compressed metadump image (offline for maximum privacy) I'll take a look.

However, some of your messages above are a result user error: you are providing a block number as an inode number to xfs_db, so it is trying to parse a non-inode as an inode.

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

Matthias Bodenbinder

2018-12-02 17:28 UTC #7 lore raw

Re: [Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

Am 02.12.18 um 18:02 schrieb [email protected]:

Quoted 10 lines

https://bugzilla.kernel.org/show_bug.cgi?id=201823
--- Comment #5 from Eric Sandeen ([email protected]) ---
If you'd like to send me a compressed metadump image (offline for maximum privacy) I'll take a look.
However, some of your messages above are a result user error: you are providing a block number as an inode number to xfs_db, so it is trying to parse a non-inode as an inode.

I will send you a dropbox link to the metadump

[email protected]

author

2018-12-02 17:40 UTC #8 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #6 from [email protected] ---
Am 02.12.18 um 18:02 schrieb [email protected]:

Quoted 11 lines

https://bugzilla.kernel.org/show_bug.cgi?id=201823
--- Comment #5 from Eric Sandeen ([email protected]) ---
If you'd like to send me a compressed metadump image (offline for maximum privacy) I'll take a look.
However, some of your messages above are a result user error: you are providing
a block number as an inode number to xfs_db, so it is trying to parse a non-inode as an inode.

I will send you a dropbox link to the metadump

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-03 21:22 UTC #9 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #7 from Eric Sandeen ([email protected]) --- So, the frag command emits these warnings under a debug build:

# xfs_db -c frag -r /dev/sde1
bad inode magic/vsn daddr 8366784 #0 (magic=5844)
Metadata corruption detected at 0x4289c3, xfs_inode block 0x7faac0/0x8000
bad inode magic/vsn daddr 34386688 #0 (magic=0)
Metadata corruption detected at 0x4289c3, xfs_inode block 0x20cb300/0x8000
bad inode magic/vsn daddr 36967232 #0 (magic=0)
Metadata corruption detected at 0x4289c3, xfs_inode block 0x2341340/0x8000

If we look at each of those blocks (daddr) after a blockget -n,

xfs_db> daddr 8366784
xfs_db> blockuse
block 1045848 (0/1045848) type dir inode 8366719 xfs_db> daddr 34386688
xfs_db> blockuse
block 5292640 (1/1098336) type free2 xfs_db> daddr 36967232
xfs_db> blockuse
block 5615208 (1/1420904) type data inode 44934228

The "inode" corruption it's finding isn't actually in inode blocks...

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-03 22:16 UTC #10 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #8 from Eric Sandeen ([email protected]) ---
I think this is probably just a bug in the frag command, and not indicative of actual filesystem corruption. Another reason to kill off the worthless frag() command. ;)

I think the bug is probably in sparse inode chunk handling.

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.

[email protected]

author

2018-12-03 22:35 UTC #11 lore raw

[Bug 201823] inconsistent behaviour of xfs_db, xfs_repair and xfs_scrub

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=201823

--- Comment #9 from Eric Sandeen ([email protected]) --- Yep, that's the case. We're missing the equivalent of

ea8a48f xfs_check: process sparse inode chunks correctly

for the frag function's version of scanfun_ino. I'll send a patch to the list, but in the meantime I think you can relax: this is a bug in the xfs_db frag command, not filesystem corruption.

Signature · 2 lines

You are receiving this mail because:
You are watching the assignee of the bug.