Re: Mount the cloned LUN from a host member of the mapped igroup

guillaume_pennec · ‎2009-07-18

Hi all,

I need small help on SNAPdrive, I have created a lun clone on filler(naante1). And, was able to map the lun_clone on host(my host is a solaris machine).

following is the procedure which i have followed :

snapdrive storage create –fs /mnttest –lun naante1:/vol/vol3/testlun –lunsize 4g (parent lun)

on filler :

- LUN cloning

Naante1>lun clone create /vol/vol3/testlun_clone -b /vol/vol3/testlun trial

- LUN mapping with Igroup

Naante1>/vol/vol3/testlun_clone

this is the sample out put on Filler:

naante1> lun show -m
---crap----
/vol/vol3/testlun                   ts1_fcp_SdIg            8       FCP
/vol/vol3/testlun_clone             ts1_fcp_SdIg            9       FCP

In solaris, using >snapdrive snap list -all (I can see the lun clone). The sample output is :

bash-3.00# snapdrive storage list -all

Connected LUNs and devices:

device filename        adapter path    size    proto   state   clone      lun path                     backing snapshot
----------------        ------- ----    ----    -----   -----   -----      --------                     ----------------

/dev/dsk/c3t60A98000486E5876644A515666462D43d0s2          -     -       4g      fcp     online lun-clone naante1:/vol/vol3/testlun_clone .snapshot/trial/testlun
-------------crap-----------------

Host devices and file systems:

dg: mnttest_SdDg                dgtype SVM
hostvol: /dev/md/mnttest_SdDg/dsk/d0    state: AVAIL
fs: /dev/md/mnttest_SdDg/dsk/d0 mount point: /mnttest (persistent) fstype ufs

device filename        adapter path    size    proto   state   clone      lun path                     backing snapshot
----------------        ------- ----    ----    -----   -----   -----      --------                     ----------------
/dev/dsk/c3t60A98000486E5876644A51565830532Dd0s2          -     -       4g      fcp     online No         naante1:/vol/vol3/testlun

Can Any one have any clue on the Mount ing the LUN Clone in SOLARIS....??

eric_barlier · ‎2009-07-19

Salut Guillaume,

Im not sure I have understood your problem but it looks like you want to mount a LUN clone.

Naante1>lun clone create /vol/vol3/testlun_clone -b /vol/vol3/testlun trial

All you need to do at this stage is use the mount command and point it to

/vol/vol3/testlun_clone

and forget about all the .snapshot output you can see.

eric

guillaume_pennec · ‎2009-07-19

Hi Edson,

Please, find the answer of NetApp.

Thanks

Guillaume

eric_barlier · ‎2009-07-19

Guillaume,

Sorry, but your last comment does not make any sense to me. I am not NetApp, most people here are not. This forum must not

be misunderstood as being NetApp tech support. Have you solved your problem?

Regards,

Eric, storage admin.

guillaume_pennec · ‎2009-07-19

Hi Eric,

Thanks for your mail.

Sorry for the misunbderstanding.

Ok now I understand.

Thanks for your answer and your support.

Thanks

Guillaume

guillaume_pennec · ‎2009-07-20

Hi Eric,

Thanks, for your prompt response. Yes, your right. I want to mount the LUN clone on solaris, so that i can access and restore the files from clone LUN as well.

I would like to share some more info on the procedure i followed, for better understanding the issue :

1) Using snapdrive we have mounted a LUN which is on Solaris /mnttest, following is the syntax command:

bash-3.00# snapdrive storage create -fs /mnttest -lun naante1:/vol/vol1/qtree2/testlun -lunsize 4g

LUN naante1:/vol/vol1/qtree2/testlun ... created

mapping new lun(s) ... done
discovering new lun(s) ... done

LUN to device file mappings:
- naante1:/vol/vol1/qtree2/testlun => /dev/dsk/c3t60A98000486E5876644A516C7873696Ed0s2

        disk group mnttest-1_SdDg created
        host volume d0 created
        file system /mnttest created

2) We have created a snapshot, for this snapdrive storage mount :/mnttest:-
bash-3.00# snapdrive snap create -fs /mnttest -snapname test_snap
Starting snap create /mnttest
WARNING: DO NOT CONTROL-C!
            If snap create is interrupted, incomplete snapdrive
                 generated data may remain on the filer volume(s)
                 which may interfere with other snap operations.
Successfully created snapshot test_snap on naante1:/vol/vol1

        snapshot test_snap contains:
        disk group mnttest-1_SdDg containing host volumes
                d0 (filesystem: /mnttest)

>>VERIFING the snapdrive storage :

bash-3.00# snapdrive storage show -all

Connected LUNs and devices:

Host devices and file systems:

dg: mnttest-1_SdDg dgtype SVM
hostvol: /dev/md/mnttest-1_SdDg/dsk/d0 state: AVAIL
fs: /dev/md/mnttest-1_SdDg/dsk/d0 mount point: /mnttest (persistent) fstype ufs

device filename        adapter path    size    proto   state   clone      lun path                     backing snapshot
----------------        ------- ----    ----    -----   -----   -----      --------                     ----------------
/dev/dsk/c3t60A98000486E5876644A516C7873696Ed0s2          -     P       4g      fcp     online No         naante1:/vol/vol1/qtree2/testlun    -
=============================================================================================

--On filer, side we have created a LUN clone, using a LUN command syntax :

naante1> lun show -m

LUN path Mapped to LUN ID Protocol //This is the Original LUN
-----------------------------------------------------------------------
/vol/vol1/qtree2/testlun ts1_fcp_SdIg 8 FCP

naante1> lun clone create /vol/vol1/qtree2/testlun_clone -b /vol/vol1/qtree2/testlun test_snap //Creation of LUN clone
naante1> lun show all
           /vol/vol1/qtree2/testlun       4g (4294967296)    (r/w, online, mapped)
        /vol/vol1/qtree2/testlun_clone      4g (4294967296)    (r/w, online)    //CLONED LUN

naante1> lun map /vol/vol1/qtree2/testlun_clone ts1_fcp_SdIg
lun map: auto-assigned ts1_fcp_SdIg=9

naante1> lun show -m

LUN path                            Mapped to          LUN ID Protocol
-----------------------------------------------------------------------
/vol/vol1/qtree2/testlun            ts1_fcp_SdIg            8       FCP
/vol/vol1/qtree2/testlun_clone      ts1_fcp_SdIg            9       FCP //Mapped to same hostname(solaris)

As we know, LUN is not a physicall drive. Its a Logical Volume Number, so as per your suggestion,

if i mount the LUN using "bash-3.00# mount naante1:/vol/vol1/qtree2/testlun_clone /test", it will chnage the mount point from directory to a file.

bash-3.00# cd /test
bash: cd: /test: Not a directory
bash-3.00# ls -la
total 1064
drwxr-xr-x 27 root     root         512 Jul 20 09:34 .
drwxr-xr-x 27 root     root         512 Jul 20 09:34 ..
-rw-------   1 root     root       10719 Jul 18 15:45 .bash_history
drwx------   2 root     root         512 Oct 14 2008 .ssh
drwx------   3 root     root         512 Aug 8 2008 .sunw
lrwxrwxrwx   1 root     root           9 Aug 8 2008 bin -> ./usr/bin
drwxr-xr-x   2 root     nobody       512 Oct 3 2008 cdrom
drwxr-xr-x   2 root     root         512 Aug 28 2008 data01
drwxr-xr-x 17 root     sys         5120 Jul 20 09:40 dev
drwxr-xr-x   2 root     sys          512 Jun 18 13:26 devices
drwxr-xr-x 85 root     sys         4608 Jul 20 09:40 etc
drwxr-xr-x   3 root     sys          512 Aug 8 2008 export
dr-xr-xr-x   1 root     root           1 Jun 18 13:27 home
drwxr-xr-x 15 root     sys          512 Aug 8 2008 kernel
drwxr-xr-x   7 root     bin         5632 Aug 8 2008 lib
drwx------   2 root     root        8192 Aug 8 2008 lost+found
drwxr-xr-x   2 root     sys          512 Aug 8 2008 mnt
drwxr-xr-x   3 root     root         512 Jul 20 09:34 mnttest
dr-xr-xr-x   1 root     root           1 Jun 18 13:27 net
drwxr-xr-x   9 root     sys          512 Aug 26 2008 opt
drwxr-xr-x 40 root     sys         1536 Aug 8 2008 platform
dr-xr-xr-x 344 root     root      480032 Jul 20 09:41 proc
drwxr-xr-x   2 root     sys         1024 Aug 8 2008 sbin
drwxr-xr-x   4 root     root         512 Aug 8 2008 system
-rw-------   1 root     root     4291821568 Jul 20 2009 test
drwxrwxrwt   7 root     sys          466 Jul 20 09:37 tmp
drwxrwxr-x 43 root     sys         1024 Aug 26 2008 usr
drwxr-xr-x 46 root     sys         1024 Aug 8 2008 var
dr-xr-xr-x   6 root     root         512 Jun 18 13:27 vol

I hope this clarifies, my issue in much clear way, I want to mount the LUN CLONE using SNAPDRIVE tool, to solaris...It will be appreciatable and helpfull if you can guide me through the process, to get this thing workout.

guillaume_pennec · ‎2009-07-20

Hi Nabil,

please note that we managed to mount the lun using a snapdrive on solaris 10 but we have issues whenever we want to mount lun clone. follow the output bellow:

Mount the cloned LUN from a host member of the mapped igroup

I want to mount the LUN clone on solaris, so that i can access and restore the files from clone LUN as well.

I would like to share some more info on the procedure i followed, for better understanding the issue :

1) Using snapdrive we have mounted a LUN which is on Solaris /mnttest, following is the syntax command:

bash-3.00# snapdrive storage create -fs /mnttest -lun naante1:/vol/vol1/qtree2/testlun -lunsize 4g

LUN naante1:/vol/vol1/qtree2/testlun ... created

mapping new lun(s) ... done

discovering new lun(s) ... done

LUN to device file mappings:

- naante1:/vol/vol1/qtree2/testlun => /dev/dsk/c3t60A98000486E5876644A516C7873696Ed0s2

disk group mnttest-1_SdDg created

host volume d0 created

file system /mnttest created

2) We have created a snapshot, for this snapdrive storage mount :/mnttest:-

bash-3.00# snapdrive snap create -fs /mnttest -snapname test_snap

Starting snap create /mnttest

WARNING: DO NOT CONTROL-C!

If snap create is interrupted, incomplete snapdrive

generated data may remain on the filer volume(s)

which may interfere with other snap operations.

Successfully created snapshot test_snap on naante1:/vol/vol1

snapshot test_snap contains:

disk group mnttest-1_SdDg containing host volumes

d0 (filesystem: /mnttest)

>>VERIFING the snapdrive storage :

bash-3.00# snapdrive storage show -all

Connected LUNs and devices:

Host devices and file systems:

dg: mnttest-1_SdDg dgtype SVM

hostvol: /dev/md/mnttest-1_SdDg/dsk/d0 state: AVAIL

fs: /dev/md/mnttest-1_SdDg/dsk/d0 mount point: /mnttest (persistent) fstype ufs

device filename adapter path size proto state clone lun path backing snapshot

eric_barlier · ‎2009-07-20

Guillaume,

I can see that you are mounting a LUN and you are succesful, but you seem to think there is a problem?

"bash-3.00# mount naante1:/vol/vol1/qtree2/testlun_clone /test", it will chnage the mount point from directory to a file.

A LUN is a file, not a file system. That is correct. Once you have mounted the LUN you will have to format it, put a file system

on it. Is that maybe where the confusion lies? Sorry if I sound dumb but I could be confused to where the problem is.

To give you an example of how I would do it with Windows and iSCSI, I create the LUN and I specify the host OS

lun create -i (iscsi) -t windows /vol/lun_path/ 50g

I then map to to my igroup and on my windows box I will rescan my disks and I will see a blank disk. This disk I will now

have to format to put NTFS on it. have you done this? is this where the problem lies?

Cdlt,

Eric

ZilchNada · ‎2014-12-02

@eric_barlier wrote:
Guillaume,

I can see that you are mounting a LUN and you are succesful, but you seem to think there is a problem?

"bash-3.00# mount naante1:/vol/vol1/qtree2/testlun_clone /test", it will chnage the mount point from directory to a file.

A LUN is a file, not a file system. That is correct. Once you have mounted the LUN you will have to format it, put a file system
on it. Is that maybe where the confusion lies? Sorry if I sound dumb but I could be confused to where the problem is.

To give you an example of how I would do it with Windows and iSCSI, I create the LUN and I specify the host OS

lun create -i (iscsi) -t windows /vol/lun_path/ 50g

I then map to to my igroup and on my windows box I will rescan my disks and I will see a blank disk. This disk I will now
have to format to put NTFS on it. have you done this? is this where the problem lies?

Cdlt,
Eric

OP is working with a LUN clone and expects, after mounting the LUN clone, to see the same files that are in the original LUN.

I am looking at a similar proble, in which the LUN clone is mounted but is empty. In my case, the LUN clone is being mounted by the same host that has already mounted to original LUN.