Backup Central

I am trying to recover a directory and recover goes to sleep and I suspect the
problem has to do with the old client alias.

I used to backup a filer "rifle" using NDMP with Solaris Networker 6.12. Six
months ago I migrated my networker server to another server upgrading to
Solaris 7.1 Build 325. I had no problem migrating the indexes. At the same
time a NetApp R200 was integrated into the backup sheme. The files from the
original Netapp client are now mirrored to the R200 and we back up the R200
again using NDMP.

The original NDMP client was left as is with the NDMP flag set to off in
Networker definition (for licensing). The backups have always been made using
DAR. Now the NetApp that was the original client was renamed and has a number
of new interfaces that the existing client in Networker does not have. The
existing client in Networker still resolves but only to one of the interfaces
on the renamed host. I added the ipadress of the renamed host that still
resolves to the old hostname to the aliases in the old client.

Existing aliases for client rifle in networker is:
rifle rifle-44 rifle-44.cs.concordia.ca rifle.cs.concordia.ca
rifle-2.cs.concordia.ca
to which I added 132.205.109.154

mercy 12:52 [124] # nslookup rifle returns

Name: valour-109.encs.concordia.ca
Address: 132.205.109.154
Aliases: rifle.encs.concordia.ca

Renamed host rifle -> valour is actually
Name: valour.services.encs.concordia.ca
Addresses: 132.205.96.252, 132.205.2.242, 132.205.11.19, 132.205.44.1
132.205.67.154, 132.205.93.2, 132.205.94.154, 132.205.109.154
Aliases: valour.encs.concordia.ca



I am now trying to recover a directory from rifle (the old client implementation).


mercy 11:40 [426] # /usr/bin/nsr/recover -c rifle
/usr/sbin/nsr/ not in index
<return> will exit.
Enter directory to browse: /vol/vol0/home/
recover: Current working directory is /vol/vol0/home/
recover> cd s/shiro
recover> versions disk1

Versions of `/vol/vol0/home/s/shiro/disk1/':

0 drwx------ shiro shiro 4096 Apr 28 2003 disk1/
save time: Sun Dec 7 02:30:07 2003
location: NetAppPerm.007 at theon, NetAppPerm.009 at theon

recover> add disk1
/vol/vol0/home/s/shiro/disk1
/vol/vol0/home/s/shiro/disk1/tez
/vol/vol0/home/s/shiro/disk1/tez/AzDiskette
/vol/vol0/home/s/shiro/disk1/tez/Problog
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP05JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP06JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP10Jul
/vol/vol0/home/s/shiro/disk1/tez/xx
/vol/vol0/home/s/shiro/disk1/tezbkup
/vol/vol0/home/s/shiro/disk1/tkde97
195 file(s) marked for recovery
recover> relocate /tmp
recover> verbose
verbose mode off
recover> volumes
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
recover> recover
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Direct Retrieval.

and it goes to sleep with a browsing message.

Same result if I use the added ipaddress

mercy 12:12 [436] # /usr/bin/nsr/recover -c 132.205.109.154
/usr/sbin/nsr/ not in index
<return> will exit.
Enter directory to browse: /vol/vol0/home/s/shiro/
recover: Current working directory is /vol/vol0/home/s/shiro/


recover> versions disk1

Versions of `/vol/vol0/home/s/shiro/disk1/':

0 drwx------ shiro shiro 4096 Apr 28 2003 disk1/
save time: Sun Dec 7 02:30:07 2003
location: NetAppPerm.007 at theon, NetAppPerm.009 at theon

recover> add disk1
/vol/vol0/home/s/shiro/disk1
/vol/vol0/home/s/shiro/disk1/tez
/vol/vol0/home/s/shiro/disk1/tez/AzDiskette
/vol/vol0/home/s/shiro/disk1/tez/Problog
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP05JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP06JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP10Jul
/vol/vol0/home/s/shiro/disk1/tez/xx
/vol/vol0/home/s/shiro/disk1/tezbkup
/vol/vol0/home/s/shiro/disk1/tkde97
195 file(s) marked for recovery
recover> relocate /tmp
recover> volumes
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
recover> recover
Recovering 195 files within /vol/vol0/home/s/shiro/ into /tmp
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Requesting 195 file(s), this may take a while...
Direct Retrieval.
and back again to sleep
--------------------------
mercy 12:05 [118] # ps auxww | grep nsr
root 20831 15.1 0.1 7368 5272 ? O 12:06:36 0:42 /usr/sbin/nsr/nsrd
root 21011 8.8 0.1 4800 2960 ? S 12:14:17 0:22
/usr/sbin/nsr/ansrd rifle:/ recovering
root 20836 2.5 0.1 7568 5584 ? S 12:06:37 0:08
/usr/sbin/nsr/nsrmmdbd
root 20827 1.4 0.1 4824 1896 ? S 12:06:21 0:03
/usr/sbin/nsr/nsrexecd
root 21009 0.2 0.1 5760 3048 pts/1 S 12:14:17 0:00
/usr/sbin/nsr/nsrndmp_recover -s mercy.services.encs.concordia.ca -c rifle -F
-m /tmp
root 21014 0.0 0.0 1000 792 pts/3 S 12:15:23 0:00 grep nsr
root 20828 0.0 0.1 4944 2704 ? S 12:06:21 0:00
/usr/sbin/nsr/nsrexecd
root 20837 0.0 0.1 4808 2856 ? S 12:06:39 0:00
/usr/sbin/nsr/nsrindexd
root 20976 0.0 0.1 5184 2816 ? S 12:07:13 0:00
/usr/sbin/nsr/nsrmmd -n 1
root 20977 0.0 0.1 5184 2816 ? S 12:07:15 0:00
/usr/sbin/nsr/nsrmmd -n 2
root 20978 0.0 0.1 5184 2816 ? S 12:07:17 0:00
/usr/sbin/nsr/nsrmmd -n 3
root 20979 0.0 0.1 5184 2816 ? S 12:07:19 0:00
/usr/sbin/nsr/nsrmmd -n 4 -r anamnesis.mercy-anamnesis.private
root 21004 0.0 0.1 5712 3240 pts/1 S 12:12:44 0:00
/usr/bin/nsr/recover -c 132.205.109.154
root 21005 0.0 0.1 5536 3312 ? S 12:12:44 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
root 21006 0.0 0.1 4760 2816 ? S 12:12:44 0:00
/usr/sbin/nsr/ansrd 132.205.109.154:root recovering
root 21010 0.0 0.1 4704 2616 ? S 12:14:17 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca


mercy 12:15 [119] # truss -p 21010
poll(0xFFBFD8A8, 1, -1) (sleeping...)
^Cmercy 12:17 [120] # truss -p 21006
poll(0xFFBFD888, 1, -1) (sleeping...)
^Cmercy 12:18 [121] # truss -p 21004
nanosleep(0xFFBFD588, 0xFFBFD580) = 0
poll(0xFFBFCB58, 1, 0) = 0
waitid(P_ALL, 0, 0xFFBFD440, WEXITED|WTRAPPED|WNOHANG) = 0
nanosleep(0xFFBFD588, 0xFFBFD580) = 0
poll(0xFFBFCB58, 1, 0) = 0
waitid(P_ALL, 0, 0xFFBFD440, WEXITED|WTRAPPED|WNOHANG) = 0


Any suggestions.


| Joel Krajden | Rm: LB-915, Tel: 514 848-2424 3052 |
| | Fax: 514 848-2830 |
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca |
| Engineering & Computer Sc.| http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and the clowns |
| Montreal, Canada | are supposed to make you laugh, not cry. |

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu

I have been able to recover to a different NDMP client by fooling the
backup server (changing the local hosts file) and then performing a new
jbconfig. Example:

Original: jbconfig has two NDMP drives:

celerra2:c0t1l0
celerra3:c0t2l0

You need to recover old NDMP client called celerra1. So, add celerra1
to /etc/hosts and give it the same IP as say celerra2. Delete the jukebox
and perform a new jbconfig. This time use celerra1:c0t1l0 as the NDMP
device. Strange, but I guess when Networker contacts the server (this case
it is celerra2 on port 10000) it does not verify the hostname!! This works
beautifully for me every time. A little annoying to keep doing jbconfigs
but it works.

Now this is using an EMC Celerra, you might get different behavior with
Netapps. Hope this helps,

Donovan

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu

Hi Joel,
In my experience, patience is the magic solution...
The way the recovery works, is that the recover first has to read the
directories/files/inodes maps.
Only after completing this stage will your files be recovered.
The initial stage can take quite a long time.
I would suggest that you verify on the Netapp side if the Netapp is
reading the tape or indeed sleeping.
You could also turn on ndmpd debugging to verify if the filer is just
sitting around or really working.
Itzik=20

-----Original Message-----
From: Legato NetWorker discussion=20
[mailto:NETWORKER < at > LISTMAIL.TEMPLE.EDU] On Behalf Of Joel Krajden
Sent: Tuesday, December 07, 2004 20:15
To: NETWORKER < at > LISTMAIL.TEMPLE.EDU
Subject: [Networker] Problem recovering directory from NDMP=20
client no longer in use
=20
I am trying to recover a directory and recover goes to sleep=20
and I suspect the problem has to do with the old client alias.
=20
I used to backup a filer "rifle" using NDMP with Solaris=20
Networker 6.12. Six months ago I migrated my networker server=20
to another server upgrading to Solaris 7.1 Build 325. I had=20
no problem migrating the indexes. At the same time a NetApp=20
R200 was integrated into the backup sheme. The files from the=20
original Netapp client are now mirrored to the R200 and we=20
back up the R200 again using NDMP.
=20
The original NDMP client was left as is with the NDMP flag=20
set to off in Networker definition (for licensing). The=20
backups have always been made using DAR. Now the NetApp that=20
was the original client was renamed and has a number of new=20
interfaces that the existing client in Networker does not=20
have. The existing client in Networker still resolves but=20
only to one of the interfaces on the renamed host. I added=20
the ipadress of the renamed host that still resolves to the=20
old hostname to the aliases in the old client.
=20
Existing aliases for client rifle in networker is:
rifle rifle-44 rifle-44.cs.concordia.ca rifle.cs.concordia.ca=20
rifle-2.cs.concordia.ca to which I added 132.205.109.154
=20
mercy 12:52 [124] # nslookup rifle returns
=20
Name: valour-109.encs.concordia.ca
Address: 132.205.109.154
Aliases: rifle.encs.concordia.ca
=20
Renamed host rifle -> valour is actually
Name: valour.services.encs.concordia.ca
Addresses: 132.205.96.252, 132.205.2.242, 132.205.11.19, 132.205.44.1
132.205.67.154, 132.205.93.2, 132.205.94.154,=20
132.205.109.154
Aliases: valour.encs.concordia.ca
=20
=20
=20
I am now trying to recover a directory from rifle (the old=20
client implementation).
=20
=20
mercy 11:40 [426] # /usr/bin/nsr/recover -c rifle=20
/usr/sbin/nsr/ not in index <return> will exit.
Enter directory to browse: /vol/vol0/home/
recover: Current working directory is /vol/vol0/home/
recover> cd s/shiro
recover> versions disk1
=20
Versions of `/vol/vol0/home/s/shiro/disk1/':
=20
0 drwx------ shiro shiro 4096 Apr 28 2003 disk1/
save time: Sun Dec 7 02:30:07 2003
location: NetAppPerm.007 at theon, NetAppPerm.009 at theon
=20
recover> add disk1
/vol/vol0/home/s/shiro/disk1
/vol/vol0/home/s/shiro/disk1/tez
/vol/vol0/home/s/shiro/disk1/tez/AzDiskette
/vol/vol0/home/s/shiro/disk1/tez/Problog
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP05JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP06JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP10Jul
/vol/vol0/home/s/shiro/disk1/tez/xx
/vol/vol0/home/s/shiro/disk1/tezbkup
/vol/vol0/home/s/shiro/disk1/tkde97
195 file(s) marked for recovery
recover> relocate /tmp
recover> verbose
verbose mode off
recover> volumes
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
recover> recover
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Direct Retrieval.
=20
and it goes to sleep with a browsing message.
=20
Same result if I use the added ipaddress
=20
mercy 12:12 [436] # /usr/bin/nsr/recover -c 132.205.109.154=20
/usr/sbin/nsr/ not in index <return> will exit.
Enter directory to browse: /vol/vol0/home/s/shiro/
recover: Current working directory is /vol/vol0/home/s/shiro/
=20
=20
recover> versions disk1
=20
Versions of `/vol/vol0/home/s/shiro/disk1/':
=20
0 drwx------ shiro shiro 4096 Apr 28 2003 disk1/
save time: Sun Dec 7 02:30:07 2003
location: NetAppPerm.007 at theon, NetAppPerm.009 at theon
=20
recover> add disk1
/vol/vol0/home/s/shiro/disk1
/vol/vol0/home/s/shiro/disk1/tez
/vol/vol0/home/s/shiro/disk1/tez/AzDiskette
/vol/vol0/home/s/shiro/disk1/tez/Problog
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP05JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP06JL
/vol/vol0/home/s/shiro/disk1/tez/Problog/BKUP10Jul
/vol/vol0/home/s/shiro/disk1/tez/xx
/vol/vol0/home/s/shiro/disk1/tezbkup
/vol/vol0/home/s/shiro/disk1/tkde97
195 file(s) marked for recovery
recover> relocate /tmp
recover> volumes
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
recover> recover
Recovering 195 files within /vol/vol0/home/s/shiro/ into /tmp=20
Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Requesting 195 file(s), this may take a while...
Direct Retrieval.
and back again to sleep
--------------------------
mercy 12:05 [118] # ps auxww | grep nsr
root 20831 15.1 0.1 7368 5272 ? O 12:06:36 0:42=20
/usr/sbin/nsr/nsrd
root 21011 8.8 0.1 4800 2960 ? S 12:14:17 0:22
/usr/sbin/nsr/ansrd rifle:/ recovering
root 20836 2.5 0.1 7568 5584 ? S 12:06:37 0:08
/usr/sbin/nsr/nsrmmdbd
root 20827 1.4 0.1 4824 1896 ? S 12:06:21 0:03
/usr/sbin/nsr/nsrexecd
root 21009 0.2 0.1 5760 3048 pts/1 S 12:14:17 0:00
/usr/sbin/nsr/nsrndmp_recover -s=20
mercy.services.encs.concordia.ca -c rifle -F -m /tmp
root 21014 0.0 0.0 1000 792 pts/3 S 12:15:23 0:00 grep nsr
root 20828 0.0 0.1 4944 2704 ? S 12:06:21 0:00
/usr/sbin/nsr/nsrexecd
root 20837 0.0 0.1 4808 2856 ? S 12:06:39 0:00
/usr/sbin/nsr/nsrindexd
root 20976 0.0 0.1 5184 2816 ? S 12:07:13 0:00
/usr/sbin/nsr/nsrmmd -n 1
root 20977 0.0 0.1 5184 2816 ? S 12:07:15 0:00
/usr/sbin/nsr/nsrmmd -n 2
root 20978 0.0 0.1 5184 2816 ? S 12:07:17 0:00
/usr/sbin/nsr/nsrmmd -n 3
root 20979 0.0 0.1 5184 2816 ? S 12:07:19 0:00
/usr/sbin/nsr/nsrmmd -n 4 -r anamnesis.mercy-anamnesis.private
root 21004 0.0 0.1 5712 3240 pts/1 S 12:12:44 0:00
/usr/bin/nsr/recover -c 132.205.109.154
root 21005 0.0 0.1 5536 3312 ? S 12:12:44 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on=20
mercy.services.encs.concordia.ca
root 21006 0.0 0.1 4760 2816 ? S 12:12:44 0:00
/usr/sbin/nsr/ansrd 132.205.109.154:root recovering
root 21010 0.0 0.1 4704 2616 ? S 12:14:17 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on=20
mercy.services.encs.concordia.ca
=20
=20
mercy 12:15 [119] # truss -p 21010
poll(0xFFBFD8A8, 1, -1) (sleeping...)
^Cmercy 12:17 [120] # truss -p 21006
poll(0xFFBFD888, 1, -1) (sleeping...)
^Cmercy 12:18 [121] # truss -p 21004
nanosleep(0xFFBFD588, 0xFFBFD580) =3D 0
poll(0xFFBFCB58, 1, 0) =3D 0
waitid(P_ALL, 0, 0xFFBFD440, WEXITED|WTRAPPED|WNOHANG) =3D 0
nanosleep(0xFFBFD588, 0xFFBFD580) =3D 0
poll(0xFFBFCB58, 1, 0) =3D 0
waitid(P_ALL, 0, 0xFFBFD440, WEXITED|WTRAPPED|WNOHANG) =3D 0
=20
=20
Any suggestions.
=20
=20
--
| Joel Krajden | Rm: LB-915, Tel: 514 848-2424=20
3052 |
| | Fax: 514 848-2830 =20
|
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca =20
|
| Engineering & Computer Sc.|=20
http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and=20
the clowns |
| Montreal, Canada | are supposed to make you=20
laugh, not cry. |
=20
--
Note: To sign off this list, send a "signoff networker"=20
command via email to listserv < at > listmail.temple.edu or visit=20
the list's Web site at=20
http://listmail.temple.edu/archives/networker.html where you=20
regarding this list should be sent to stan < at > temple.edu=20
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D=
*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D
=20

*************************************************************************=
*************************
The contents of this email and any attachments are confidential.
It is intended for the named recipient(s) only.
If you have received this email in error please notify the system manager=
or the=20
sender immediately and do not disclose the contents to any one or make co=
pies.

MBI - System Team
*************************************************************************=
*************************

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D

Thanks Donovan, but I think the problem may not have anything to do with
resolving clients. I just realized that I was trying to recover to /tmp on my
NSR server from a NetAQpp. That does not work. Now I have to figure out how to
recover the directory from my old Netapp filer client but send it to the
same filer with a new hostname or send it to the NearStore. The tape drives
are connected to the NearStore and I only have one license for NDMP.

Anyone played with nsrndmp_recover in detail.

Joel




Donovan O'Brien wrote:
I have been able to recover to a different NDMP client by fooling the
backup server (changing the local hosts file) and then performing a new
jbconfig. Example:

Original: jbconfig has two NDMP drives:

celerra2:c0t1l0
celerra3:c0t2l0

You need to recover old NDMP client called celerra1. So, add celerra1
to /etc/hosts and give it the same IP as say celerra2. Delete the jukebox
and perform a new jbconfig. This time use celerra1:c0t1l0 as the NDMP
device. Strange, but I guess when Networker contacts the server (this case
it is celerra2 on port 10000) it does not verify the hostname!! This works
beautifully for me every time. A little annoying to keep doing jbconfigs
but it works.

Now this is using an EMC Celerra, you might get different behavior with
Netapps. Hope this helps,

Donovan

--
Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu


| Joel Krajden | Rm: LB-915, Tel: 514 848-2424 3052 |
| | Fax: 514 848-2830 |
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca |
| Engineering & Computer Sc.| http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and the clowns |
| Montreal, Canada | are supposed to make you laugh, not cry. |

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu

Joel,
I do not think you need a license for the old Netapp.
Just grant remote access from the old one to the new one and recover the
NDMP files to the new Netapp.
If the NDMP tape drives are on the old Netapp and you want to recover to
the new one you may need two licenses.
In such a case you could use an evaluation license that is available in
the documentation accompanying the Legato software kits.
Itzik=20

-----Original Message-----
From: Legato NetWorker discussion=20
[mailto:NETWORKER < at > LISTMAIL.TEMPLE.EDU] On Behalf Of Joel Krajden
Sent: Tuesday, December 07, 2004 23:22
To: NETWORKER < at > LISTMAIL.TEMPLE.EDU
Subject: Re: [Networker] Problem recovering directory from=20
NDMP client no longer in use
=20
Thanks Donovan, but I think the problem may not have anything=20
to do with resolving clients. I just realized that I was=20
trying to recover to /tmp on my NSR server from a NetAQpp.=20
That does not work. Now I have to figure out how to
recover the directory from my old Netapp filer client but=20
send it to the same filer with a new hostname or send it to=20
the NearStore. The tape drives are connected to the NearStore=20
and I only have one license for NDMP.
=20
Anyone played with nsrndmp_recover in detail.
=20
Joel
=20
=20
=20
=20
Donovan O'Brien wrote:
I have been able to recover to a different NDMP client by=20
fooling the=20
backup server (changing the local hosts file) and then performing a=20
new jbconfig. Example:

Original: jbconfig has two NDMP drives:

celerra2:c0t1l0
celerra3:c0t2l0

You need to recover old NDMP client called celerra1. So,=20
add celerra1=20
to /etc/hosts and give it the same IP as say celerra2. =20
Delete the jukebox
and perform a new jbconfig. This time use celerra1:c0t1l0=20
as the NDMP
device. Strange, but I guess when Networker contacts the=20
server (this=20
case it is celerra2 on port 10000) it does not verify the=20
hostname!!=20
This works beautifully for me every time. A little=20
annoying to keep=20
doing jbconfigs but it works.

Now this is using an EMC Celerra, you might get different behavior=20
with Netapps. Hope this helps,

Donovan

--
Note: To sign off this list, send a "signoff networker" command via=20
Web site at=20
http://listmail.temple.edu/archives/networker.html where=20
you can also=20
view and post messages to the list. Questions regarding this list=20
should be sent to stan < at > temple.edu=20
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D=
*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D
=20
=20
--
| Joel Krajden | Rm: LB-915, Tel: 514 848-2424=20
3052 |
| | Fax: 514 848-2830 =20
|
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca =20
|
| Engineering & Computer Sc.|=20
http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and=20
the clowns |
| Montreal, Canada | are supposed to make you=20
laugh, not cry. |
=20
--
Note: To sign off this list, send a "signoff networker"=20
command via email to listserv < at > listmail.temple.edu or visit=20
the list's Web site at=20
http://listmail.temple.edu/archives/networker.html where you=20
regarding this list should be sent to stan < at > temple.edu=20
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D=
*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D
=20

*************************************************************************=
*************************
The contents of this email and any attachments are confidential.
It is intended for the named recipient(s) only.
If you have received this email in error please notify the system manager=
or the=20
sender immediately and do not disclose the contents to any one or make co=
pies.

MBI - System Team
*************************************************************************=
*************************

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D

Thanks Itzik. I am trying to recover to the nearstore which has the attached
tapes - so far this is what I see.


Recovering 195 files within /vol/vol0/home/s/shiro/ into
anamnesis.mercy-anamnesis.private::/vol/misc02Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Requesting 195 file(s), this may take a while...
Direct Retrieval.

zzz... does not look like it is reading the index and no attempts to mount a tape.

mercy 17:19 [100] # ps auxww | grep nsr
root 22536 15.3 0.1 7368 5272 ? O 16:10:09 4:30 /usr/sbin/nsr/nsrd
root 22773 9.2 0.1 4800 2960 ? S 17:12:25 2:33
/usr/sbin/nsr/ansrd rifle:/ recovering
root 22541 2.5 0.1 7592 5608 ? S 16:10:10 0:45
/usr/sbin/nsr/nsrmmdbd
root 22532 1.6 0.1 4824 1896 ? S 16:09:54 0:25
/usr/sbin/nsr/nsrexecd
root 22771 0.1 0.1 5760 3048 pts/1 S 17:12:25 0:02
/usr/sbin/nsr/nsrndmp_recover -s mercy.services.encs.concordia.ca -c rifle -F
-m anamnesis.mercy-anamnesis.private
root 22794 0.0 0.0 1000 792 pts/3 S 17:19:36 0:00 grep nsr
root 22533 0.0 0.1 4968 2712 ? S 16:09:54 0:00
/usr/sbin/nsr/nsrexecd
root 22542 0.0 0.1 4808 2856 ? S 16:10:12 0:00
/usr/sbin/nsr/nsrindexd
root 22681 0.0 0.1 5184 2816 ? S 16:10:35 0:00
/usr/sbin/nsr/nsrmmd -n 1
root 22682 0.0 0.1 5184 2816 ? S 16:10:37 0:00
/usr/sbin/nsr/nsrmmd -n 2
root 22683 0.0 0.1 5184 2816 ? S 16:10:39 0:00
/usr/sbin/nsr/nsrmmd -n 3
root 22684 0.0 0.1 5184 2816 ? S 16:10:41 0:00
/usr/sbin/nsr/nsrmmd -n 4 -r anamnesis.mercy-anamnesis.private
root 22763 0.0 0.1 5616 3104 pts/1 S 17:08:23 0:00
/usr/bin/nsr/recover -c rifle
root 22764 0.0 0.1 5032 2920 ? S 17:08:23 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
root 22765 0.0 0.1 4760 2816 ? S 17:08:23 0:00
/usr/sbin/nsr/ansrd rifle:root recovering
root 22772 0.0 0.1 4704 2616 ? S 17:12:25 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
mercy 17:19 [101] # truss -p 22772
poll(0xFFBFD8A8, 1, -1) (sleeping...)

Itzik Meirson wrote:
Joel,
I do not think you need a license for the old Netapp.
Just grant remote access from the old one to the new one and recover the
NDMP files to the new Netapp.
If the NDMP tape drives are on the old Netapp and you want to recover to
the new one you may need two licenses.
In such a case you could use an evaluation license that is available in
the documentation accompanying the Legato software kits.
Itzik



Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu

I killed the recover since nothing was happening.

How does nsrndmp_recover know to what to do if the client rifle no longer has
tape drives to mount the tapes found in the index. How can I tell anamnesis
(the nearstore client with the attached tape drives) to do the reading and
writing.

What is actaully happening when I run this command
/usr/sbin/nsr/nsrndmp_recover -s mercy.services.encs.concordia.ca -c
rifle -F -m anamnesis.mercy-anamnesis.private

joel

Joel Krajden wrote:
Thanks Itzik. I am trying to recover to the nearstore which has the
attached
tapes - so far this is what I see.


Recovering 195 files within /vol/vol0/home/s/shiro/ into
anamnesis.mercy-anamnesis.private::/vol/misc02Volumes needed (all on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Requesting 195 file(s), this may take a while...
Direct Retrieval.

zzz... does not look like it is reading the index and no attempts to
mount a tape.

mercy 17:19 [100] # ps auxww | grep nsr
root 22536 15.3 0.1 7368 5272 ? O 16:10:09 4:30
/usr/sbin/nsr/nsrd
root 22773 9.2 0.1 4800 2960 ? S 17:12:25 2:33
/usr/sbin/nsr/ansrd rifle:/ recovering
root 22541 2.5 0.1 7592 5608 ? S 16:10:10 0:45
/usr/sbin/nsr/nsrmmdbd
root 22532 1.6 0.1 4824 1896 ? S 16:09:54 0:25
/usr/sbin/nsr/nsrexecd
root 22771 0.1 0.1 5760 3048 pts/1 S 17:12:25 0:02
/usr/sbin/nsr/nsrndmp_recover -s mercy.services.encs.concordia.ca -c
rifle -F
-m anamnesis.mercy-anamnesis.private
root 22794 0.0 0.0 1000 792 pts/3 S 17:19:36 0:00 grep nsr
root 22533 0.0 0.1 4968 2712 ? S 16:09:54 0:00
/usr/sbin/nsr/nsrexecd
root 22542 0.0 0.1 4808 2856 ? S 16:10:12 0:00
/usr/sbin/nsr/nsrindexd
root 22681 0.0 0.1 5184 2816 ? S 16:10:35 0:00
/usr/sbin/nsr/nsrmmd -n 1
root 22682 0.0 0.1 5184 2816 ? S 16:10:37 0:00
/usr/sbin/nsr/nsrmmd -n 2
root 22683 0.0 0.1 5184 2816 ? S 16:10:39 0:00
/usr/sbin/nsr/nsrmmd -n 3
root 22684 0.0 0.1 5184 2816 ? S 16:10:41 0:00
/usr/sbin/nsr/nsrmmd -n 4 -r anamnesis.mercy-anamnesis.private
root 22763 0.0 0.1 5616 3104 pts/1 S 17:08:23 0:00
/usr/bin/nsr/recover -c rifle
root 22764 0.0 0.1 5032 2920 ? S 17:08:23 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
root 22765 0.0 0.1 4760 2816 ? S 17:08:23 0:00
/usr/sbin/nsr/ansrd rifle:root recovering
root 22772 0.0 0.1 4704 2616 ? S 17:12:25 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
mercy 17:19 [101] # truss -p 22772
poll(0xFFBFD8A8, 1, -1) (sleeping...)

Itzik Meirson wrote:

Joel,
I do not think you need a license for the old Netapp.
Just grant remote access from the old one to the new one and recover the
NDMP files to the new Netapp.
If the NDMP tape drives are on the old Netapp and you want to recover to
the new one you may need two licenses.
In such a case you could use an evaluation license that is available in
the documentation accompanying the Legato software kits.
Itzik



--
Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu


| Joel Krajden | Rm: LB-915, Tel: 514 848-2424 3052 |
| | Fax: 514 848-2830 |
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca |
| Engineering & Computer Sc.| http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and the clowns |
| Montreal, Canada | are supposed to make you laugh, not cry. |

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu

I got the restore to work. To rehash I had a NetApp client backed up using
ndmp. That client is no longer backed up and has no tape drives since my
backups are now done from another netapp client which has the attached drives
and tape library shared with the networker server.

I needed to restore files from the original client and whatever I tried the
recover or nsrndmp_recover would hang. The trick (and I have found no
documentation for it) is:

1) Make sure the original client has the setting for NDMP configured to on.
This may cause Networker to complain about only having one license in the logs
but it can be ignored.
2) The storage node setting for the original client (and this is the key) must
have an entry for the working netapp client which is acting as the storage
node. That is how networker knows where the tape needs to be mounted.

The documentation for dealing with ndmp clients remains woefully inadequate
three years after release and the the lack of error or status messages when
running nsrndmp_recover is poor programming to say the least.


joel


Joel Krajden wrote:
I killed the recover since nothing was happening.

How does nsrndmp_recover know to what to do if the client rifle no
longer has
tape drives to mount the tapes found in the index. How can I tell anamnesis
(the nearstore client with the attached tape drives) to do the reading and
writing.

What is actaully happening when I run this command
/usr/sbin/nsr/nsrndmp_recover -s mercy.services.encs.concordia.ca -c
rifle -F -m anamnesis.mercy-anamnesis.private

joel

Joel Krajden wrote:

Thanks Itzik. I am trying to recover to the nearstore which has the
attached
tapes - so far this is what I see.


Recovering 195 files within /vol/vol0/home/s/shiro/ into
anamnesis.mercy-anamnesis.private::/vol/misc02Volumes needed (all
on-line):
NetAppPerm.007 at theon
NetAppPerm.009 at theon
Requesting 195 file(s), this may take a while...
Direct Retrieval.

zzz... does not look like it is reading the index and no attempts to
mount a tape.

mercy 17:19 [100] # ps auxww | grep nsr
root 22536 15.3 0.1 7368 5272 ? O 16:10:09 4:30
/usr/sbin/nsr/nsrd
root 22773 9.2 0.1 4800 2960 ? S 17:12:25 2:33
/usr/sbin/nsr/ansrd rifle:/ recovering
root 22541 2.5 0.1 7592 5608 ? S 16:10:10 0:45
/usr/sbin/nsr/nsrmmdbd
root 22532 1.6 0.1 4824 1896 ? S 16:09:54 0:25
/usr/sbin/nsr/nsrexecd
root 22771 0.1 0.1 5760 3048 pts/1 S 17:12:25 0:02
/usr/sbin/nsr/nsrndmp_recover -s mercy.services.encs.concordia.ca -c
rifle -F
-m anamnesis.mercy-anamnesis.private
root 22794 0.0 0.0 1000 792 pts/3 S 17:19:36 0:00 grep nsr
root 22533 0.0 0.1 4968 2712 ? S 16:09:54 0:00
/usr/sbin/nsr/nsrexecd
root 22542 0.0 0.1 4808 2856 ? S 16:10:12 0:00
/usr/sbin/nsr/nsrindexd
root 22681 0.0 0.1 5184 2816 ? S 16:10:35 0:00
/usr/sbin/nsr/nsrmmd -n 1
root 22682 0.0 0.1 5184 2816 ? S 16:10:37 0:00
/usr/sbin/nsr/nsrmmd -n 2
root 22683 0.0 0.1 5184 2816 ? S 16:10:39 0:00
/usr/sbin/nsr/nsrmmd -n 3
root 22684 0.0 0.1 5184 2816 ? S 16:10:41 0:00
/usr/sbin/nsr/nsrmmd -n 4 -r anamnesis.mercy-anamnesis.private
root 22763 0.0 0.1 5616 3104 pts/1 S 17:08:23 0:00
/usr/bin/nsr/recover -c rifle
root 22764 0.0 0.1 5032 2920 ? S 17:08:23 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
root 22765 0.0 0.1 4760 2816 ? S 17:08:23 0:00
/usr/sbin/nsr/ansrd rifle:root recovering
root 22772 0.0 0.1 4704 2616 ? S 17:12:25 0:00
/usr/sbin/nsr/nsrindexd -a READ rifle index for backup on
mercy.services.encs.concordia.ca
mercy 17:19 [101] # truss -p 22772
poll(0xFFBFD8A8, 1, -1) (sleeping...)

Itzik Meirson wrote:

Joel,
I do not think you need a license for the old Netapp.
Just grant remote access from the old one to the new one and recover the
NDMP files to the new Netapp.
If the NDMP tape drives are on the old Netapp and you want to recover to
the new one you may need two licenses.
In such a case you could use an evaluation license that is available in
the documentation accompanying the Legato software kits.
Itzik



--
Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu



--
| Joel Krajden | Rm: LB-915, Tel: 514 848-2424 3052 |
| | Fax: 514 848-2830 |
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca |
| Engineering & Computer Sc.| http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and the clowns |
| Montreal, Canada | are supposed to make you laugh, not cry. |

--
Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu


| Joel Krajden | Rm: LB-915, Tel: 514 848-2424 3052 |
| | Fax: 514 848-2830 |
| Senior Systems Analyst | Email: joelk < at > cs.concordia.ca |
| Engineering & Computer Sc.| http://www.cs.concordia.ca/~staffcs/joelk |
| Concordia University | Remember it's a circus and the clowns |
| Montreal, Canada | are supposed to make you laugh, not cry. |

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu