Topic
  • 2 replies
  • Latest Post - ‏2013-07-05T20:18:36Z by JohnKaufmann
JohnKaufmann
JohnKaufmann
6 Posts

Pinned topic Unexplainable fs mount error... looking for explaination

‏2013-07-05T14:37:47Z |

We have many clusters, they usually are very similar. But one I created the file systems for a new cluster I was building, I had an issue where one of the file systems created would not mount. It was name mm. Its mount point was /mm, but when you try to mount it, it give you the oddest error message (below) and won't mount.

Below i have done mmlsfs of the file system, shown that it is not mounted (df -h) and then tried to mount it.

My question is, what record stores the individual information for how a file system is built? I would like to look at it (or them) and see if it has the incorrect mount point some place.

Here is the output (with the error message in bold at the bottom):

[root@rso-isfile001a ~]# mmlsfs mm
flag                value                    description
------------------- ------------------------ -----------------------------------
 -f                 8192                     Minimum fragment size in bytes
 -i                 512                      Inode size in bytes
 -I                 16384                    Indirect block size in bytes
 -m                 1                        Default number of metadata replicas
 -M                 2                        Maximum number of metadata replicas
 -r                 1                        Default number of data replicas
 -R                 2                        Maximum number of data replicas
 -j                 cluster                  Block allocation type
 -D                 nfs4                     File locking semantics in effect
 -k                 all                      ACL semantics in effect
 -n                 128                      Estimated number of nodes that will mount file system
 -B                 262144                   Block size
 -Q                 none                     Quotas enforced
                    none                     Default quotas enabled
 --filesetdf        No                       Fileset df enabled?
 -V                 13.01 (3.5.0.0)          File system version
 --create-time      Wed May 15 16:40:40 2013 File system creation time
 -u                 Yes                      Support for large LUNs?
 -z                 No                       Is DMAPI enabled?
 -L                 4194304                  Logfile size
 -E                 Yes                      Exact mtime mount option
 -S                 No                       Suppress atime mount option
 -K                 whenpossible             Strict replica allocation option
 --fastea           Yes                      Fast external attributes enabled?
 --inode-limit      11591168                 Maximum number of inodes
 -P                 system;mm_pool           Disk storage pools in file system
 -d                 nsd7;nsd107              Disks in file system
 --perfileset-quota no                       Per-fileset quota enforcement
 -A                 yes                      Automatic mount option
 -o                 none                     Additional mount options
 -T                 /mm                      Default mount point
 --mount-priority   0                        Mount priority
[root@rso-isfile001a ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/VolGroup-lv_root
                      251G  6.7G  232G   3% /
tmpfs                  12G  8.0K   12G   1% /dev/shm
/dev/sda1             485M   33M  427M   8% /boot
/dev/asthin           512G  2.3M  512G   1% /asthin
/dev/cnfs              10G  2.3M   10G   1% /cnfs
/dev/dbashare          82G   11M   82G   1% /dbashare
/dev/em                21T  131G   21T   1% /em
/dev/emthin            21T  2.3M   21T   1% /emthin
/dev/isthin            26T  2.3M   26T   1% /isthin
/dev/mmthin           7.3T  2.3M  7.3T   1% /mmthin
/dev/nethome          300G  469M  300G   1% /nethome
/dev/pd                25T  2.3M   25T   1% /pd
/dev/pdthin            17T  2.3M   17T   1% /pdthin
/dev/pgthin           615G  2.3M  615G   1% /pgthin
/dev/vmexports        6.6T  211G  6.4T   4% /vmexports
[root@rso-isfile001a ~]# mmmount mm
Fri Jul  5 13:42:00 GMT 2013: mmmount: Mounting file systems ...
mount: mount point gpfs does not exist
mmmount: Command failed.  Examine previous error messages to determine cause.
 

  • sdenham
    sdenham
    70 Posts

    Re: Unexplainable fs mount error... looking for explaination

    ‏2013-07-05T20:06:39Z  

    It looks to me like you might have hit some unintended reserved use of "mm".  Have you tried either changing the default mountpoint

    mmchfs mm -T /test

    or forcing the mount to a different mount point?

    mmmount mm /test

    What is the contents of the line starting with /dev/mm in /etc/fstab ?

     

     

  • JohnKaufmann
    JohnKaufmann
    6 Posts

    Re: Unexplainable fs mount error... looking for explaination

    ‏2013-07-05T20:18:36Z  
    • sdenham
    • ‏2013-07-05T20:06:39Z

    It looks to me like you might have hit some unintended reserved use of "mm".  Have you tried either changing the default mountpoint

    mmchfs mm -T /test

    or forcing the mount to a different mount point?

    mmmount mm /test

    What is the contents of the line starting with /dev/mm in /etc/fstab ?

     

     

    lol... you were exactly correct!!

    For some reason there was an entry for /dev/mm with no mount point, then later on down the entries there was one with /dev/mm     /mm  (for the mount point). Not sure how the first one got there, it was generated by gpfs and was on both the nodes in the cluster.

    I guess i am so used to it working flawlessly that I don't even think to look there... but I will now!

     

    Thanks so very much for you help and advice,

     

    John