Misleading message while mounting lvm

[Mohammed Firdosh Nasim]

Hi All,
I am trying to mount a device/file system, which is already mounted to another mount point. While doing so I got an error message (as expected), but its misleading.
Device: /dev/srm_fs/vxfs
old mount point: /mnt/vxfs_mount_old
new mount point: /mnt/vxfs_mount_new

----------- screen dump start --------
# uname -a
HP-UX spea67 B.11.23 U ia64 1517336840 unlimited-user license
# mount -F vxfs /dev/srm_fs/vxfs /mnt/vxfs_mount_old
# mount | grep vxfs_mount_old
/mnt/vxfs_mount_old on /dev/srm_fs/vxfs ioerror=mwdisable,delaylog,nodatainlog,dev=40030002 on Tue Mar 10 17:44:09 2009
# mount -F vxfs /dev/srm_fs/vxfs /mnt/vxfs_mount_new
mount: /dev/srm_fs/vxfs is already mounted on /mnt/vxfs_mount_new
# echo $?
1
# mount | grep vxfs_mount_
/mnt/vxfs_mount_old on /dev/srm_fs/vxfs ioerror=mwdisable,delaylog,nodatainlog,dev=40030002 on Tue Mar 10 17:44:09 2009
#
----------- screen dump end --------

The error message returned mentions "dev/srm_fs/vxfs is already mounted on /mnt/vxfs_mount_new", which is not correct, as the device is still mounted on /mnt/vxfs_mount_old.

Any help pertaining to this will be appreciated.

Thanks and Regards,
Mohammed.

[James R. Ferguson]

Hi:

While the message may include the wrong name of the mount-point, the important event is that (1) you get a return code indicating and error; and (2) that (at least) the first part of the message indicates that the error is for the reason that the device is already mounted. Clearly you can then decipher *where* so I think all you can do is file a bug report.

Regards!

...JRF...

[Torsten.]

I'm not sure what /dev/srm_fs/vxfs is, but under some special circumstances you may mount clustered file systems more than once - what do you receive by running


# mount | grep vxfs_mount


Just grep for all "vxfs_mount".

[Mohammed Firdosh Nasim]

JRF,
Thanks for the response. As I am a new member, I would appreciate if you can let me know the procedure/link to file a bug report.



Torsten,

/dev/srm_fs/vxfs : Well, here I have created a volume group "srm_fs" and a logical volume "vxfs" in that VG. And then I have created a "vxfs" file system on that LV.

I do see only one mount point, i.e. the one which was created first.
----------- screen dump start --------
# uname -a
HP-UX spea67 B.11.23 U ia64 1517336840 unlimited-user license
# mount | grep vxfs_mount
/mnt/vxfs_mount_old on /dev/srm_fs/vxfs ioerror=mwdisable,delaylog,nodatainlog,dev=40030002 on Tue Mar 10 17:44:09 2009
----------- screen dump end --------

Thanks,
Mohammed

[Torsten.]

OK, thanks.

Before complaining about anything I would first check the patches (even via patch session here on ITRC or with swa - software assistant).

[smatador]

Hi,

And what about the mnttab.
grep vxfs_mount /etc/mnttab
In the past, I have an issue with mnttab/mount. I don't know why, my customer have made multiple mounting point with the same lv.

[Mohammed Firdosh Nasim]

Torsten,

Couldn't find "swa" on the host. But could get the patch info using "swlist"
----------- screen dump start --------
# uname -a
HP-UX spea67 B.11.23 U ia64 1517336840 unlimited-user license
# which mount
/usr/sbin/mount
# ls -l /usr/sbin/mount
lrwxr-xr-x 1 bin bin 11 Oct 15 15:47 /usr/sbin/mount -> /sbin/mount
# ls -l /sbin/mount
-r-xr-xr-x 1 bin bin 788268 Aug 27 2004 /sbin/mount
# swlist -l file | grep "/sbin/mount"
OS-Core.UX-CORE: /usr/sbin/mount
OS-Core.UX-CORE: /usr/sbin/mountall
OS-Core.UX2-CORE: /sbin/mount
OS-Core.UX2-CORE: /sbin/mountall
PHCO_31545.UX-CORE: /usr/sbin/mount
PHCO_31545.UX2-CORE: /sbin/mount
#
----------- screen dump end --------

I have attached the complete "swlist" output as attachment (spea67_HP-UXIA64_11.23.txt).

Just wanted to add that, in another host I am not seeing correct error message thrown while mounting the same device in other mount point. This is 11.31 host.
----------- screen dump start --------
spea119 #uname -a
HP-UX spea119 B.11.31 U ia64 1176152452 unlimited-user license
spea119 #mount -F vxfs /dev/test_vg/vxfs /mnt/vxfs_mount_old
spea119 #mount | grep vxfs_mount
/mnt/vxfs_mount_old on /dev/test_vg/vxfs ioerror=mwdisable,delaylog,dev=40030001 on Wed Mar 11 17:31:58 2009
spea119 #mount -F vxfs /dev/test_vg/vxfs /mnt/vxfs_mount_new
mount: /dev/test_vg/vxfs is already mounted on /mnt/vxfs_mount_old
spea119 #echo $?
1
spea119 #
----------- screen dump end --------

Does the initial issue reported by me, boils down to patch update?

Thanks,
Mohammed.

[Ganesh.A]

Hi ,

Rename or delete your current mount point .... and create it freshly and give a try it might work.

fuser - see if any process using it.