mmfsckx command

Checks and repairs corruptions on a mounted IBM Storage Scale file system.

Synopsis


 mmfsckx Device Start of change[--repair]End of change[--check-reserved-files-only ][--max-threads MaxThreads]
            [--max-pagepool-percent MaxPagepoolPercent] [--qos QosClass]
            [-N {all | mount | Node[,Node...] | NodeFile | NodeClass}]

Availability

Available on all IBM Storage Scale editions.

Start of change

Description

The mmfsckx command is used to report or repair file system metadata corruptions. The report and repair actions can be done while the file system remains mounted. You can run the mmfsckx command for the following reasons:
  • MMFS_FSSTRUCT error entries in system log. For more information, see Operating system error logs.
  • Corrupted output when querying file system directories.
  • Disk startup failure.
  • File system mount failure.
  • Storage media failure.
Note: To analyze the impact of running the repair option, it is recommended that you first run the mmfsckx command in the report-only mode to identify the number and type of repairs that are required.
Unlike the mmfsck command, the mmfsckx command cannot repair all file system corruptions. The Table 1 lists all the corruptions that the command can repair.
Table 1. File system corruptions repaired by the mmfsckx command.
Type of Error Description Critical
Reserved file corruption Reserved files are files internal to GPFS and not visible to users. They contain all file system metadata. Yes
Improperly freed blocks. Blocks that are marked as available in the block allocation map system file but are referenced in some files. Yes
Duplicate disk addresses When two or more disk addresses in the same or different files point to overlapping sectors on the disk. Yes
Inconsistent inode allocation status In-use files are marked as available in the inode allocation map or vice-versa. No
Metadata corruption Includes the file inode, indirect blocks, and extended attribute blocks. No
Lost blocks Blocks that are marked as in-use in the block allocation map system file but do not belong to any file. No
The mmfsckx command does not check or repair the following directory related corruptions. To correct these corruptions, run the mmfsck command with the file system unmounted.
  • bad directory blocks
  • bad directory entries
  • fileset identifier validation
  • file link counts
  • orphaned inodes
When the mmfsckx command is running, the following commands and activities that change the file system are blocked from running:
  • mmchfs
  • mmchmgr
  • mmadddisk
  • mmdeldisk
  • mmchdisk
  • mmcrsnapshot
  • mmdelsnapshot
  • mmcrfileset
  • mmdelfileset
  • mmchfileset
  • Start of changemmlinkfilesetEnd of change
  • Start of changemmunlinkfilesetEnd of change
  • mmrestripefs
  • mmdefedquota
  • mmdefquotaon
  • mmrepquota
  • mmfsck
  • mmcheckquota
  • Start of changeAuto inode space expansionEnd of change
  • Start of changeCreation of new ACLsEnd of change
A single file system cannot have multiple mmfsckx commands run against it concurrently, however, multiple file systems can have an mmfsckx command run against them concurrently.

To run the mmfsckx command, all nodes that mount the file system, including remote mounts must be running IBM Storage Scale 5.2.0.0 or later.

When the mmfsckx command is executing, attempts to mount the file system on nodes running an IBM Storage Scale version older than 5.2.0.0 will be blocked until the mmfsckx command exits.

When run in report mode, the default, not all metadataOnly or dataAndMetadata NSDs need to be available. The metadata on NSDs that are not available do not get validated, and any disk addresses pointing to these unavailable NSDs is considered valid. However, when the mmfsck command runs with the --repair option, metadataOnly and dataAndMetadata NSDs must be available. If unavailable disks are physically damaged, then they must be deleted before running the mmfsckx command.

Before running the mmfsckx command, consider the issues and ongoing demands on the system. For example, running the command on a large file system might affect the system performance. In such cases, plan to run the command when the system demand is low.

End of change

Results

Start of changeIf the metadata in the file system is corrupted, the mmfsckx command displays information about the corruption in the command output along with other information like scan progress and statistics. All corruption-related information is prefixed with a ! to distinguish it from other information. End of change

On completion of the file system scan, the mmfsckx command prints a summary table of all the corrupted inodes that were identified in the scan. The mmfsckx command reports the inode numbers of corrupted files only. Run the mmfindinode command from samples/ilm/ directory in the IBM Storage Scale installation path to convert inode numbers to file paths.

Start of changeIf there are many inodes reported by the mmfsckx command that you need to convert to their corresponding path name, the mmfindinode command will accept a file with single inode on each line of the file. A file in this format can be generated from the mmfsckx command output using the following awk command:
awk ‘/^-/,/^$/{if (/^-|^$/)next; if ($2==0) print $3}’ /tmp/mmfsckx.out > /tmp/inodefile
End of change
The following scenarios might result in aborting a running instance of the mmfsckx operation:
  • A user who uses ctrl+c on the mmfsckx command line.
  • An environmental error like disk, network, or insufficient memory.
  • Start of changeThe node, where the file system manager is located for the file system, that is being checked either fails or panics the file system.End of change
  • Any node that is participating in the file system scan fails or panics the file system.
Note: Start of changeThe mmfsckx command is not aborted if a node, not participating in the file system scan but having the file system being checked mounted on it, fails. The command also does not get aborted if the file system being checked is unmounted from any of the nodes.End of change

Start of changeIf the mmfsckx command is aborted, the cause for the failure is included in the mmfsckx output, and logged on the node currently acting as the file system manager for the file system being checked. Retrying the mmfsckx command operation restarts the file system scan from the beginning. It does not resume from the last interrupted scan. End of change

Parameters

Device
The device name of the file system to be checked and repaired. File system names do not have to be fully qualified. fs0 is as acceptable as /dev/fs0.

This parameter must be the first parameter.

Start of change--repairEnd of change
Start of changeSpecifies that the mmfsckx command repairs corruptions that it finds in the file system.End of change
--check-reserved-files-only
Specifies that the mmfsckx command checks only the reserved files in the file system.
Note: Reserved files are special files used internally by IBM Storage Scale for maintaining file system metadata.

--maxthreads MaxThreads
The maximum number of parallel threads that scans the file system on a single node. The default is 32.

--maxpagepool-percent MaxPagepoolPercent
The maximum percentage of available page-pool memory on any node that is used by the mmfsckx command during its scan. The default is 50%.

--qos QosClass
Specifies the Quality of Service for I/O operations (QoS) class to which the instance of the command is assigned. If you do not specify this parameter, the instance of the command is assigned by default to the maintenance QoS class. This parameter has no effect unless the QoS service is enabled. For more information, see the topic mmchqos command. Specify one of the following QoS classes:
maintenance
This QoS class is typically configured to have a smaller share of file system IOPS. Use this class for I/O-intensive, potentially long-running IBM Storage Scale commands, so that they contribute less to reducing overall file system performance.
other
This QoS class is typically configured to have a larger share of file system IOPS. Use this class for administration commands that are not I/O-intensive.
For more information, see Setting the Quality of Service for I/O operations.
[-N {all | mount | Node[,Node...] | NodeFile | NodeClass}]
Specifies the nodes that are to participate in the checking of the file system. Only nodes that are at IBM Storage Scale 5.1.4 or higher can participate in the check. The file system manager node always participates in the check even if it is not specified in this parameter. This command supports all defined node classes. The default is all or the current value of the defaultHelperNodes parameter of the mmchconfig command.

To run the mmfsckx command on a single node, use the mmchmgr command to make the target node the file system manager and then run mmfsckx with -N targetNode.

For information on how to specify node names, see Specifying nodes as input to GPFS commands.

Exit status

0
Successful completion.
nonzero
A failure has occurred.

Security

You must have root authority to run the mmfsckx command.

The node on which the command is issued must be able to run remote shell commands on any other node in the cluster. It must run the commands without the use of a password and without producing any extraneous messages. For more information, see Requirements for administering a GPFS file system.

Examples

  1. An example output of mmfsckx command run on a file system without any corruptions.
    mmfsckx fs1
2022-01-01_00:13:18+0530: starting mmfsckx
mmfsckx /dev/fs1
  master node is h1
  GPFS version is 5.1.4.0
  system Linux x86_64, release 4.18.0-305.19.1.el8_4.x86_64
  maintenanceMode is off 
  maxThreads is 32 (default 32)
  maxPagePoolPercent is 50% (default 50%)
  nInodesToLock is 1008 (default 1008)
  lineNumAdjust is 0
File system info
  features are FastEA; StripedLogs; IExpandV2
  allocation map layout is vertical type is cluster
  replicas metadata is 1 of 2 data is 1 of 2 log is 0 of 2
  subblocksPerFullBlock is 32
  Sizes 
    sector is 512 B    
    metadata subblock is 2 KB (4 sectors)
    metadata fullblock is 64 KB (128 sectors)
    data subblock is 2 KB (4 sectors)
    data fullblock is 64 KB (128 sectors)
    inode is 4 KB (8 sectors) 
    indirect block is 8 KB (16 sectors)
    ea overflow block is 64 KB (128 sectors)
    directory block is 64 KB (128 sectors)
Recovering file system
Waiting for async recovery to complete
Initializing mounted nodes
Creating fsckx snapshot fsckx_2022-01-01_00.13.18
Created fsckx snapshot with snapId 13
Initializing caches on master node
Initializing reserved files
Checking inode of reserved files
Reserved files info
  snapshots global are 2 fileset are 0
  inode0 file
    inodeNum is 0 subblocksPerFullBlock is 32 recordsPerBlock is 16
    inodeSpaceMask is 0x0 inodeBlockMask is 0x1FFF
  block alloc map file of system pool
    inodeNum is 1 subblocksPerFullBlock is 32 recordsPerBlock is 16
    disks are 2 regions are 773 segments are 1 holdsData yes
  inode alloc map file
    inodeNum is 2 subblocksPerFullBlock is 32 recordsPerBlock is 32
    type is cluster regions are 257 segments are 1
    iallocSpaceMask is 0x0 iallocSegmtMask is 0x0
  access control list file
    inodeNum is 4 subblocksPerFullBlock is 32 recordsPerBlock is 1
  extended attribute file
    inodeNum is 5 subblocksPerFullBlock is 8 recordsPerBlock is 1
  fileset metadata file
    inodeNum is 38 subblocksPerFullBlock is 32 recordsPerBlock is 16
    independent are 1 afm are 0 dependent are 0
Selecting worker nodes
  Selected 1 worker nodes
Combined pagepool usage across all scanner nodes
         Total pagepool: 4.00 GB
     Stealable pagepool: 3.39 GB
  Fsckx Usable pagepool: 1.69 GB (50%)
        Needed pagepool: 26.43 MB (1%)
Initializing worker nodes
Checking reserved files metadata pass 1 of 1
Initializing caches on worker nodes
Checking user files metadata pass 1 of 1
2022-01-01_00:13:22: checked 0 of 65792 inodes
2022-01-01_00:13:25: checked 65792 of 65792 inodes
Checking inode allocation map for inodes 0 to 131071
Checking block allocation map of poolId 0 for segment 0 regions 387 to 772 on node h1
Checking block allocation map of poolId 0 for segment 0 regions 0 to 386 on node h2
Deleting fsckx snapshots
Deleting fsckx snapshot fsckx_2022-01-01_00.13.18

Inodes:
    65792 total (in active file system)
       62 used
        1   directories
        3   user-files
    65920 scanned (including snapshots)
        0   fatal
        0   data-loss
        0   inconsistent
        0   bad-alloc
        0   lost-alloc

Disk-sectors:
  8388608 total (all disks)
  8388608 used
      512   directories
   274120   user-files
  1601040 read from disk (all passes)
        0   corrupt
        0   duplicate (critical)
        0   bad-alloc (critical)
        0   lost-alloc

Disk-addresses:
    16708 valid (including snapshots)
        0 corrupt
        0 duplicate
        0 deleted-disk
        0 reserved-file-hole
        0 metadata-replica-mismatches

File system has no metadata corruptions 
2022-01-01_00:13:28+0530: mmfsckx completed after 10 seconds.
  2. An example output of mmfsckx command run on a file system with corruptions.
    mmfsckx fs1
2022-01-01_00:34:19+0530: starting mmfsckx
mmfsckx /dev/fs1
  master node is h1
  GPFS version is 5.1.4.0
  system Linux x86_64, release 4.18.0-305.19.1.el8_4.x86_64
  maintenanceMode is off 
  maxThreads is 32 (default 32)
  maxPagePoolPercent is 50% (default 50%)
  nInodesToLock is 1008 (default 1008)
  lineNumAdjust is 0
File system info
  features are FastEA; StripedLogs; IExpandV2
  allocation map layout is vertical type is cluster
  replicas metadata is 1 of 2 data is 1 of 2 log is 0 of 2
  subblocksPerFullBlock is 32
  Sizes 
    sector is 512 B    
    metadata subblock is 2 KB (4 sectors)
    metadata fullblock is 64 KB (128 sectors)
    data subblock is 2 KB (4 sectors)
    data fullblock is 64 KB (128 sectors)
    inode is 4 KB (8 sectors) 
    indirect block is 8 KB (16 sectors)
    ea overflow block is 64 KB (128 sectors)
    directory block is 64 KB (128 sectors)
Recovering file system
Waiting for async recovery to complete
Initializing mounted nodes
Creating fsckx snapshot fsckx_2022-01-01_00.34.19
Created fsckx snapshot with snapId 19
Initializing caches on master node
Initializing reserved files
Reserved files info
  snapshots global are 2 fileset are 0
  inode0 file
    inodeNum is 0 subblocksPerFullBlock is 32 recordsPerBlock is 16
    inodeSpaceMask is 0x0 inodeBlockMask is 0x1FFF
  block alloc map file of system pool
    inodeNum is 1 subblocksPerFullBlock is 32 recordsPerBlock is 16
    disks are 2 regions are 773 segments are 1 holdsData yes
  inode alloc map file
    inodeNum is 2 subblocksPerFullBlock is 32 recordsPerBlock is 32
    type is cluster regions are 257 segments are 1
    iallocSpaceMask is 0x0 iallocSegmtMask is 0x0
  access control list file
    inodeNum is 4 subblocksPerFullBlock is 32 recordsPerBlock is 1
  extended attribute file
    inodeNum is 5 subblocksPerFullBlock is 8 recordsPerBlock is 1
  fileset metadata file
    inodeNum is 38 subblocksPerFullBlock is 32 recordsPerBlock is 16
    independent are 1 afm are 0 dependent are 0
Selecting worker nodes
  Selected 1 worker nodes
Combined pagepool usage across all scanner nodes
         Total pagepool: 4.00 GB
     Stealable pagepool: 3.39 GB
  Fsckx Usable pagepool: 1.69 GB (50%)
        Needed pagepool: 26.43 MB (1%)
Initializing worker nodes
Checking reserved files metadata pass 1 of 1
Initializing caches on worker nodes
Checking user files metadata pass 1 of 1
2022-01-01_00:34:21: checked 0 of 65792 inodes
!Inode 8288 snap 0 fset 0 "user file" inode @1:2195712: disk address 1:100 in slot 0 replica 0 pointing to data block 0 code 1928 is corrupt
!Inode 8288 snap 0 fset 0 "user file" inode @1:2195712: disk address 1:100 in slot 1 replica 1 pointing to data block 0 code 1928 is corrupt
!Inode 8288 snap 0 fset 0 "user file" inode @1:2195712: lastBlockSubblocks 12 should be 0
!Inode 10560 snap 0 inode @1:2204800: deleted inode fileSize 23744 should be 0
!Inode 11136 snap 0 fset 0 "user file" inode @1:2207104: nFullBlocks 0 should be 1
2022-01-01_00:34:25: checked 65792 of 65792 inodes
Checking inode allocation map for inodes 0 to 131071
!Inode 10560 allocation state used should be free
Checking block allocation map of poolId 0 for segment 0 regions 387 to 772 on node h1
Checking block allocation map of poolId 0 for segment 0 regions 0 to 386 on node h2
!Block 1:131584 has map status: 0x00000000 expected: 0xFFFFFFFF
!Block 1:131712 has map status: 0x00000000 expected: 0xFFFFFFFF
!Block 1:132096 has map status: 0x000FFFFF expected: 0x00000000
!Block 1:924672 has map status: 0x000FFFFF expected: 0xFFFFFFFF
!Block 1:1761408 has map status: 0x00000000 expected: 0x0FFFFF00
!Block 1:1761536 has map status: 0x00000000 expected: 0xFFFFFFFF
!Block 1:1761664 has map status: 0x00000000 expected: 0xFFFFFFFF
!Block 1:1761792 has map status: 0x00000000 expected: 0xFFFFFFFF
!Block 1:1761920 has map status: 0x00000000 expected: 0xFFFFFFFF
Deleting fsckx snapshots
Deleting fsckx snapshot fsckx_2022-01-01_00.34.19

No. SnapId     InodeNum FileType                 Corruption   Severity
--- ---------- -------- ------------------------ ------------ ------------
1   0          8288     user file                inconsistent non-critical
2   0          10560    deleted inode            fatal        non-critical
3   0          11136    user file                inconsistent non-critical 

Inodes:
    65792 total (in active file system)
       61 used
        1   directories
        4   user-files
    65952 scanned (including snapshots)
        1   fatal
        0   data-loss
        2   inconsistent
        0   bad-alloc
        1   lost-alloc

Disk-sectors:
  4194304 total (all disks)
  4193328 used
      256   directories
   101668   user-files
  1328736 read from disk (all passes)
        8   corrupt
        0   duplicate (critical)
       80   bad-alloc (critical)
      896   lost-alloc

Disk-addresses:
     8072 valid (including snapshots)
        2 corrupt
        0 duplicate
        0 deleted-disk
        0 reserved-file-hole
        0 metadata-replica-mismatches

File system has critical metadata corruptions
File system is NOT safe to use without repair
2022-01-01_00:34:30+0530: mmfsckx completed after 11 seconds
  3. Start of changeAn example output of mmfsckx command to repair a file system with corruptions.
    mmfsckx fs1 --repair
2023-09-20_13:02:22-0400: starting mmfsckx
mmfsckx /dev/fs1 --repair 
  master node is node3
  GPFS version is 5.1.9.0
  system Linux x86_64, release 3.10.0-693.el7.x86_64
  maintenanceMode is off
  maxThreads is 32 (default 32)
  maxPagePoolPercent is 50% (default 50%)
  nInodesToLock is 1024 (default 1024)
  lineNumAdjust is 0
File system info
  format version is 33.00 (5.1.9.0)
  features are FastEA; StripedLogs; 4KAligned; IExpandV2
  allocation map layout is vertical  type is scatter
  replicas  metadata is 1 of 2  data is 1 of 2  log is 0 of 2
  subblocksPerFullBlock is 512
  Sizes
    sector is 512 B
    metadata subblock is 8 KB (16 sectors)
    metadata fullblock is 4 MB (8192 sectors)
    data subblock is 8 KB (16 sectors)
    data fullblock is 4 MB (8192 sectors)
    inode is 4 KB (8 sectors)
    indirect block is 32 KB (64 sectors)
    ea overflow block is 64 KB (128 sectors)
    directory block is 256 KB (512 sectors)
Checking mounted nodes
Creating fsckx snapshot fsckx_2023-09-20_13.02.22
Created fsckx snapshot with snapId 3
Initializing mounted nodes
Initializing reserved files
Initializing caches on master node
Checking inode of reserved files
Reserved files info
  snapshots  global are 1  fileset are 0
  inode0 file
    inodeNum is 0  subblocksPerFullBlock is 512  recordsPerBlock is 1024
    inodeSpaceMask is 0x300  inodeBlockMask is 0xFF
  block alloc map file of system pool
    inodeNum is 1  subblocksPerFullBlock is 512  recordsPerBlock is 512
    disks are 14  regions are 787  segments are 2  holdsData yes
  inode alloc map file
    inodeNum is 2  subblocksPerFullBlock is 16  recordsPerBlock is 16
    type is cluster  regions are 257  segments are 4
    iallocSpaceMask is 0x3  iallocSegmtMask is 0x0
  access control list file
    inodeNum is 4  subblocksPerFullBlock is 512  recordsPerBlock is 1
  extended attribute file
    inodeNum is 5  subblocksPerFullBlock is 2  recordsPerBlock is 1
  fileset metadata file
    inodeNum is 38  subblocksPerFullBlock is 512  recordsPerBlock is 1024
    independent are 4  afm are 0  dependent are 0
Selecting worker nodes
  Selected 1 worker nodes
    node4
Combined pagepool usage across all scanner nodes
         Total pagepool: 2.00 GB
     Stealable pagepool: 1.43 GB
  Fsckx Usable pagepool: 734.68 MB (50%)
        Needed pagepool: 595.63 MB (41%)
Initializing worker nodes
Checking reserved files metadata pass 1 of 1
Initializing caches on worker nodes
Checking user files metadata pass 1 of 1
  2023-09-20_13:02:23: checking 0 of 444416 inodes
Checking inode allocation map for inodes 0 to 131071
Checking inode allocation map for inodes 131072 to 262143
!Inode 280540 snap 0 fset 1 "user file" inode @9:7962336: checksum 0x0 should be 0x4AC09003; inode will be deleted
Checking inode allocation map for inodes 262144 to 393215
!Inode 280540 allocation state used should be free
Checking inode allocation map for inodes 393216 to 524287
Checking inode allocation map for inodes 524288 to 655359
Checking inode allocation map for inodes 655360 to 786431
  2023-09-20_13:02:28: checked 444416 of 444416 inodes
Checking inode allocation map for inodes 786432 to 886783
Checking block allocation map of poolId 0 for segmentNum 0 regionNums 0 to 393 on node node3
!Block 3:188416 has map status:
    0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 
  expected:
    0x0001FFFFF8000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 
!Block 5:188416 has map status:
    0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 
  expected:
    0x0000000000000000 0x3FFFFFFFC0000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 
Checking block allocation map of poolId 0 for segmentNum 0 regionNums 394 to 786 on node node4
Checking for duplicate and bad alloc references in pass 1
  2023-09-20_13:02:35: checking 0 of 444416 inodes
Checking inode allocation map for inodes 0 to 131071
!Inode 280544 snap 0 fset 1 "user file" inode @9:7962368: disk address 7:332464 in slot 0 replica 0 pointing to data block 0 code 2136 is a duplicate
!Inode 280544 snap 0 fset 1 "user file" inode @9:7962368: lastBlockSubblocks 22 should be 512
Checking inode allocation map for inodes 262144 to 393215
Checking inode allocation map for inodes 524288 to 655359
!Inode 792543 snap 0 fset 3 "user file" inode @1:16678648: disk address 7:332464 in slot 0 replica 0 pointing to data block 0 code 2136 is a duplicate
!Inode 792543 snap 0 fset 3 "user file" inode @1:16678648: lastBlockSubblocks 36 should be 512
  2023-09-20_13:02:39: checked 444416 of 444416 inodes
Checking inode allocation map for inodes 786432 to 886783
Checking block allocation map of poolId 0 for segmentNum 0 regionNums 0 to 393 on node node3
!Block 7:327680 has map status:
    0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 
  expected:
    0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x00000000001FFFFF 0xFFFE000000000000 0x0000000000000000 0x0000000000000000 
Checking block allocation map of poolId 0 for segmentNum 0 regionNums 394 to 786 on node node4
Deleting fsckx snapshots
Deleting fsckx snapshot fsckx_2023-09-20_13.02.22

No. SnapId InodeNum FileType                 Corruption   Severity     
--- ------ -------- ------------------------ ------------ ------------ 
1   0      280540   user file                fatal        non-critical 
2   0      280544   user file                data-loss    critical     
3   0      792543   user file                data-loss    critical     

Inodes:
     444416 total (in active file system)
      26098 used
         40   directories
      26019   user-files
     890880 scanned (including snapshots)
          1   fatal
          2   data-loss
          0   inconsistent
          0   bad-alloc
          1   lost-alloc

Disk-sectors:
  285293992 total (all disks)
   15442264 used
       3776   directories
    7864880   user-files
   11486008 read from disk (all passes)
          0   bad
        352   duplicate
          0   bad-alloc
       1440   lost-alloc

Disk-addresses:
      26206 valid (including snapshots)
          0 corrupt
          2 duplicate (critical)
          0 bad-alloc (critical)
          0 deleted-disk
          0 reserved-file-hole
          0 metadata-replica-mismatches

File system had critical metadata corruptions that were repaired.

2023-09-20_13:02:45-0400: mmfsckx completed after 23 seconds
    End of change

See also

mmfsck command

Location

/usr/lpp/mmfs/bin