[Linux-cluster] gfs2_grow - Error writing new rindex entries; aborted.

[Marco Huang]

Hi,

We are running Cenos5.5 (gfs2-utils.x86_64 v0.1.62-20.el5) cluster. We want to add another 5T disk space on the filesystem - expand from 19T to 25T, however it doesn't grow over 20T. There was no error when I did test run. So just wondering if there is a limitation of gfs2_grow which only can grow gfs2 filesystem upto 20T ? Is there anyone had same experience?

Test run
# gfs2_grow -T /mnt/fsbackup
(Test mode--File system will not be changed)
FS: Mount Point: /mnt/fsbackup
FS: Device:      /dev/mapper/fsbackup-fsbackup01
FS: Size:        4882811901 (0x12309cbfd)
FS: RG size:     524244 (0x7ffd4)
DEV: Size:       6103514112 (0x16bcc3c00)
The file system grew by 4768368MB.
gfs2_grow complete.

Actual run
# gfs2_grow /mnt/fsbackup
FS: Mount Point: /mnt/fsbackup
FS: Device:      /dev/mapper/fsbackup-fsbackup01
FS: Size:        4882811901 (0x12309cbfd)
FS: RG size:     524244 (0x7ffd4)
DEV: Size:       6103514112 (0x16bcc3c00)
The file system grew by 4768368MB.
Error writing new rindex entries;aborted.
gfs2_grow complete.


Before gfs2_grow
# df -h /mnt/fsbackup/
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/fsbackup-fsbackup01
                       19T   19T  651M 100% /mnt/fsbackup

# gfs2_tool df
/mnt/fsbackup:
  SB lock proto = "lock_dlm"
  SB lock table = "FSC:fsbackup01"
  SB ondisk format = 1801
  SB multihost format = 1900
  Block size = 4096
  Journals = 8
  Resource Groups = 10112
  Mounted lock proto = "lock_dlm"
  Mounted lock table = "FSC:fsbackup01"
  Mounted host data = "jid=0:id=65539:first=1"
  Journal number = 0
  Lock module flags = 0
  Local flocks = FALSE
  Local caching = FALSE

  Type           Total Blocks   Used Blocks    Free Blocks    use%           
  ------------------------------------------------------------------------
  data           5300270360     4882310027     417960333      92%
  inodes         447901122      29940789       417960333      7%

After gfs2_grow
#  df -h /mnt/fsbackup/
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/fsbackup-fsbackup01
                       20T   19T  1.6T  93% /mnt/fsbackup


# gfs2_tool df
/mnt/fsbackup:
  SB lock proto = "lock_dlm"
  SB lock table = "FSC:fsbackup01"
  SB ondisk format = 1801
  SB multihost format = 1900
  Block size = 4096
  Journals = 8
  Resource Groups = 9314
  Mounted lock proto = "lock_dlm"
  Mounted lock table = "FSC:fsbackup01"
  Mounted host data = "jid=0:id=65539:first=1"
  Journal number = 0
  Lock module flags = 0
  Local flocks = FALSE
  Local caching = FALSE

  Type           Total Blocks   Used Blocks    Free Blocks    use%           
  ------------------------------------------------------------------------
  data           4882476584     4882310009     166575         100%
  inodes         30107364       29940789       166575         99%



cheers
--
Marco

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20110407/68e1ffb2/attachment.htm>