Bug 702050 - gfs2_grow -T actually grows the FS
Summary: gfs2_grow -T actually grows the FS
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Red Hat Enterprise Linux 5
Classification: Red Hat
Component: gfs2-utils
Version: 5.6
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: rc
: ---
Assignee: Robert Peterson
QA Contact: Cluster QE
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-05-04 16:07 UTC by Madison Kelly
Modified: 2011-05-11 01:23 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2011-05-11 01:23:17 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Description Madison Kelly 2011-05-04 16:07:00 UTC
Description of problem:

Running 'gfs2_grow -T' on a CLVM LV says that it will test, but actually grows the FS.

Version-Release number of selected component (if applicable):

cman-2.0.115-68.el5_6.3
gfs2-utils-0.1.62-28.el5

How reproducible:

Only tested once, uncertain.

Steps to Reproduce:
1. Use 'lvextend' to extend a clustered LV.
2. Use 'gfs2_grow -T <args>' to test.
3. use 'df -h' to see that the FS is already grown, despite the '-T'est.
  
Actual results:

Grew the file system.

Expected results:

Not to grow the file system.

Additional info:

[root@xenmaster003 ~]# lvextend -L +50G /dev/drbd_sh1_vg0/cluster_files /dev/drbd3
  Extending logical volume cluster_files to 250.00 GB
  Logical volume cluster_files successfully resized

[root@xenmaster003 ~]# gfs2_grow -T /dev/drbd_sh1_vg0/cluster_files /cluster_files/
(Test mode--File system will not be changed)
FS: Mount Point: /cluster_files
FS: Device:      /dev/mapper/drbd_sh1_vg0-cluster_files
FS: Size:        52428798 (0x31ffffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       65536000 (0x3e80000)
The file system grew by 51200MB.
FS: Mount Point: /cluster_files
FS: Device:      /dev/mapper/drbd_sh1_vg0-cluster_files
FS: Size:        52428798 (0x31ffffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       65536000 (0x3e80000)
The file system grew by 51200MB.
gfs2_grow complete.

[root@xenmaster003 ~]# gfs2_grow /dev/drbd_sh1_vg0/cluster_files /cluster_files/
FS: Mount Point: /cluster_files
FS: Device:      /dev/mapper/drbd_sh1_vg0-cluster_files
FS: Size:        52428798 (0x31ffffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       65536000 (0x3e80000)
The file system grew by 51200MB.
Error: The device has grown by less than one Resource Group (RG).
The device grew by 0MB.  One RG is 255MB for this file system.
gfs2_grow complete.

[root@xenmaster003 ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/md2               57G  2.7G   51G   6% /
/dev/md0              251M   52M  187M  22% /boot
tmpfs                 7.7G     0  7.7G   0% /dev/shm
/dev/mapper/drbd_sh0_vg0-xen_shared
                       56G  259M   56G   1% /xen_shared
/dev/mapper/drbd_sh1_vg0-cluster_files
                      250G  145G  106G  58% /cluster_files

Comment 1 Steve Whitehouse 2011-05-06 10:20:43 UTC
Did you run the df -h after the initial grow -T or only after the second (real) grow command?

It looks like the second command (the real one) did the right thing. I'm not sure why the first (test) command printed the info twice, which is a bit strange.

I would be quite surprised if the gfs2_grow command did manage to grow the filesystem when in test mode, since it opens all its files in read only mode for safety in that case. If you are able to reproduce, can you get an strace of the gfs2_grow -T command?

Also a df -h both before and after the grow -T would be helpful.

Comment 2 Robert Peterson 2011-05-06 18:57:11 UTC
I tried and was unable to recreate the problem:

[root@roth-01 ../gfs2/mkfs]# rpm -q gfs2-utils
gfs2-utils-0.1.62-28.el5
[root@roth-01 ../gfs2/mkfs]# lvresize -L50G /dev/roth_vg/roth_lv 
  WARNING: Reducing active logical volume to 50.00 GB
  THIS MAY DESTROY YOUR DATA (filesystem etc.)
Do you really want to reduce roth_lv? [y/n]: y
  Reducing logical volume roth_lv to 50.00 GB
  Logical volume roth_lv successfully resized
[root@roth-01 ../gfs2/mkfs]# mkfs.gfs2 -O -t bobs_roth:roth_lv -p lock_dlm -j 4 /dev/roth_vg/roth_lv
Device:                    /dev/roth_vg/roth_lv
Blocksize:                 4096
Device Size                50.00 GB (13107200 blocks)
Filesystem Size:           50.00 GB (13107198 blocks)
Journals:                  4
Resource Groups:           200
Locking Protocol:          "lock_dlm"
Lock Table:                "bobs_roth:roth_lv"
UUID:                      4B910756-5C6D-F16A-706D-1C4A9B9F101D

[root@roth-01 ../gfs2/mkfs]# mount -tgfs2 /dev/roth_vg/roth_lv /mnt/gfs2
[root@roth-01 ../gfs2/mkfs]# lvextend -L +50G /dev/roth_vg/roth_lv 
  Extending logical volume roth_lv to 100.00 GB
  Logical volume roth_lv successfully resized
[root@roth-01 ../gfs2/mkfs]# gfs2_grow -T /dev/roth_vg/roth_lv 
(Test mode--File system will not be changed)
FS: Mount Point: /mnt/gfs2
FS: Device:      /dev/mapper/roth_vg-roth_lv
FS: Size:        13107198 (0xc7fffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       26214400 (0x1900000)
The file system grew by 51200MB.
gfs2_grow complete.
[root@roth-01 ../gfs2/mkfs]# gfs2_grow /dev/roth_vg/roth_lv 
FS: Mount Point: /mnt/gfs2
FS: Device:      /dev/mapper/roth_vg-roth_lv
FS: Size:        13107198 (0xc7fffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       26214400 (0x1900000)
The file system grew by 51200MB.
gfs2_grow complete.
[root@roth-01 ../gfs2/mkfs]# 

The first gfs2_grow gave two reports of the file system size,
and the only way that should be able to happen is if two
file systems were specified on one gfs2_grow.  However, once
-T mode is specified, all the specified file systems are
treated the same way: as read-only.  So even when specified
twice, the problem did not recreate.  Here's what I get when
I specified the same file system twice:

[root@roth-01 ../gfs2/mkfs]# umount /mnt/gfs2
[root@roth-01 ../gfs2/mkfs]# lvresize -L50G /dev/roth_vg/roth_lv 
  WARNING: Reducing active logical volume to 50.00 GB
  THIS MAY DESTROY YOUR DATA (filesystem etc.)
Do you really want to reduce roth_lv? [y/n]: y
  Reducing logical volume roth_lv to 50.00 GB
  Logical volume roth_lv successfully resized
[root@roth-01 ../gfs2/mkfs]# mkfs.gfs2 -O -t bobs_roth:roth_lv -p lock_dlm -j 4 /dev/roth_vg/roth_lv
Device:                    /dev/roth_vg/roth_lv
Blocksize:                 4096
Device Size                50.00 GB (13107200 blocks)
Filesystem Size:           50.00 GB (13107198 blocks)
Journals:                  4
Resource Groups:           200
Locking Protocol:          "lock_dlm"
Lock Table:                "bobs_roth:roth_lv"
UUID:                      47739572-5BD8-E05F-8B1F-F913346D0FF1

[root@roth-01 ../gfs2/mkfs]# mount -tgfs2 /dev/roth_vg/roth_lv /mnt/gfs2
[root@roth-01 ../gfs2/mkfs]# lvextend -L +50G /dev/roth_vg/roth_lv 
  Extending logical volume roth_lv to 100.00 GB
  Logical volume roth_lv successfully resized
[root@roth-01 ../gfs2/mkfs]# gfs2_grow -T /dev/roth_vg/roth_lv /dev/roth_vg/roth_lv
(Test mode--File system will not be changed)
FS: Mount Point: /mnt/gfs2
FS: Device:      /dev/mapper/roth_vg-roth_lv
FS: Size:        13107198 (0xc7fffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       26214400 (0x1900000)
The file system grew by 51200MB.
FS: Mount Point: /mnt/gfs2
FS: Device:      /dev/mapper/roth_vg-roth_lv
FS: Size:        13107198 (0xc7fffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       26214400 (0x1900000)
The file system grew by 51200MB.
gfs2_grow complete.
[root@roth-01 ../gfs2/mkfs]# gfs2_grow /dev/roth_vg/roth_lv 
FS: Mount Point: /mnt/gfs2
FS: Device:      /dev/mapper/roth_vg-roth_lv
FS: Size:        13107198 (0xc7fffe)
FS: RG size:     65535 (0xffff)
DEV: Size:       26214400 (0x1900000)
The file system grew by 51200MB.
gfs2_grow complete.
[root@roth-01 ../gfs2/mkfs]# 

When -T is specified, the device is opened O_RDONLY, so
writes to it should be impossible.  The code looks correct.
So at this point I'm baffled.  How did you do that?

Comment 3 Madison Kelly 2011-05-06 19:08:44 UTC
I'll retest it shortly and run 'df -h' after each step. If I am not able to reproduce it, I'll close it as NOTABUG. Though, this still confuses me:

The file system grew by 51200MB.
Error: The device has grown by less than one Resource Group (RG).
The device grew by 0MB.  One RG is 255MB for this file system.
gfs2_grow complete.

I'll try to get a new test done tonight.

Comment 4 Madison Kelly 2011-05-11 01:05:53 UTC
I can't reproduce it, either. Sorry about the line noise, please close as NOTABUG.

Comment 5 Robert Peterson 2011-05-11 01:23:17 UTC
No problem.  Closed as per comment #4.


Note You need to log in before you can comment on or make changes to this bug.