I am trying to do an ndmpcopy from a 7m 8.1.4P1 to an Oracle ZS3.
I am not even sure if this is possible, but It authenticates successfully then dies with a network communication error.
Looking at a pkkt capture, the green is the netapp and the red is the zs3. There is 2 way protocol communication for about 15 seconds then the netapp sends a reset.
--Jordan
toaster*> ndmpcopy -d -da user:pass /vol/vm_store101 zs31:/export/test123/ Ndmpcopy: Starting copy [ 24 ] ... Ndmpcopy: toaster: Notify: Connection established Ndmpcopy: zs31: Notify: Connection established Ndmpcopy: toaster: Connect: Authentication successful Ndmpcopy: zs31: Connect: Authentication successful Ndmpcopy: zs31: Log: Running Restore without Direct Access RestoreNdmpcopy: zs31: Log: Restoring to "/export/test123/". Ndmpcopy: zs31: Log: Tape server: remote at 10.41.1.205:33866. Ndmpcopy: toaster: Log: DUMP: creating "/vol/vm_store101/../snapshot_for_backup.3267" snapshot. Ndmpcopy: toaster: Log: DUMP: Using Full Volume Dump Ndmpcopy: toaster: Log: DUMP: Date of this level 0 dump: Thu Jul 31 11:07:51 2014. Ndmpcopy: toaster: Log: DUMP: Date of last level 0 dump: the epoch. Ndmpcopy: toaster: Log: DUMP: Dumping /vol/vm_store101 to NDMP connection Ndmpcopy: toaster: Log: DUMP: mapping (Pass I)[regular files] Ndmpcopy: toaster: Log: DUMP: mapping (Pass II)[directories] Ndmpcopy: toaster: Log: DUMP: estimated 2616811716 KB. Ndmpcopy: toaster: Log: DUMP: dumping (Pass III) [directories] Ndmpcopy: toaster: Log: DUMP: dumping (Pass IV) [regular files] Ndmpcopy: zs31: Notify: restore successful Ndmpcopy: toaster: Log: DUMP: Network communication error Ndmpcopy: toaster: Log: DUMP: DUMP IS ABORTED Ndmpcopy: toaster: Log: DUMP: Deleting "/vol/vm_store101/../snapshot_for_backup.3267" snapshot. Ndmpcopy: toaster: Log: DATA: Backup terminated: EVENT: I/O ERROR (for /vol/vm_store101) Ndmpcopy: toaster: Connection halted: HALT: Internal error! Ndmpcopy: Notify: Transfer failed
[cid:image001.png@01CFACB0.B7B49340]
For the seventh consecutive year, Independent Health was named one of the best companies to work for in New York State.
The Buffalo News also named Independent Health one of the Top Workplaces in Western New York for 2014.
Go to www.independenthealth.com to view careers and available positions.
We are an Equal Opportunity Employer.
CONFIDENTIALITY NOTICE: This e-mail and its attachments (collectively referred to as "e-mail") may contain confidential information that is privileged and protected from disclosure by Federal and State confidentiality laws, rules or regulations. This e-mail is intended for the designated addressee only. If you are not the designated addressee, you are notified that any disclosure, copying or distribution of this e-mail may be unlawful and may subject you to legal consequences. If you have received this e-mail in error, please contact me immediately by telephone at (716) 631 - 3001 and delete the e-mail from your computer immediately. Thank you for your attention.
Jordan> I am trying to do an ndmpcopy from a 7m 8.1.4P1 to an Oracle Jordan> ZS3. I am not even sure if this is possible, but It Jordan> authenticates successfully then dies with a network Jordan> communication error.
It's not possible. NDMP is just a wrapper around the vendor specific dump/restore tools. I find this annoying too, but it's better than nothing.
Basically, the Netapp is just using the regular netapp 'dump' command but instead of going to a local tape drive, it's going to an NDMP target (which *can* be a local FC tape drive, but not a SCSI one as I recall) or can be over the network to another block device which understands NDMP.
The protocol then sends an index stream of to the controlling NDMP process so it can see which file(s) are backed up.
This is also why you can't do a full NDMP backup, and then unlimited incrementals for a month. You need to make sure you have weekly differential backups in the mix as well, since Netapp dump is really just old style unix dump with upto nine levels of backups.
Does this answer you're question?
The only way you'll make this work is mount the Netapp filesystem onto a system which also mount the Oracle ZS3 filesystem and then use tar or rsync or even just 'cp -r ....' to move the data.
John
Log into the ZS3 system.
rsh <netapp> dump 0f - /vol/volumename | (cd destination ; restore rf -)
Stupid but would get your copy over.
Using rsh reduces CPU load on the NeApp controller.
— Mike - old school nerd who misses SUNW
Mike> Log into the ZS3 system. Mike> rsh <netapp> dump 0f - /vol/volumename | (cd destination ; restore rf -)
Mike> Stupid but would get your copy over.
Mike> Using rsh reduces CPU load on the NeApp controller.
Have you actually tried doing this? The format of Netapp dump and ZS3 restore aren't compatible in my understanding. Luckily, I do have a Solaris 10 ZFS system I can use for testing tomorrow if I get a chance to double check.
John
They should be. It is an industry standard dump.
--tmac
*Tim McCarthy*
On Tue, Sep 2, 2014 at 8:05 PM, John Stoffel john@stoffel.org wrote:
Mike> Log into the ZS3 system. Mike> rsh <netapp> dump 0f - /vol/volumename | (cd destination ; restore rf -)
Mike> Stupid but would get your copy over.
Mike> Using rsh reduces CPU load on the NeApp controller.
Have you actually tried doing this? The format of Netapp dump and ZS3 restore aren't compatible in my understanding. Luckily, I do have a Solaris 10 ZFS system I can use for testing tomorrow if I get a chance to double check.
John _______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters
tmac> They should be. It is an industry standard dump.
Really? So how does it save and restore volumes with mixed security? Or CIFS volumes to a Unix based restore program? Just curious how that would be expected work properly.
Mixed security is simply the system picking one security style or the other. It still leverages NTFS *or* UNIX style ACLs. There¹s not a special form of ACL that mixed security uses. All standard ACLs.
If you are copying files with NTFS ACLs to UNIX file systems, you will lose the NTFS ACLs. This would happen in any and all environments.
On 9/2/14, 9:41 PM, "John Stoffel" john@stoffel.org wrote:
tmac> They should be. It is an industry standard dump.
Really? So how does it save and restore volumes with mixed security? Or CIFS volumes to a Unix based restore program? Just curious how that would be expected work properly. _______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters
NDMP is a backup control protocol which is industry-standard, but it does not define data format. It's only designed to have one backup device tell another backup device to back itself up, or restore itself. It allows products like CommVault to contact a NetApp, and say "Hi there, do a backup of path X and send it to device Y" and then sit back and wait.
You actually have 2 formats that you can select with ONTAP. One of them is native snapmirror format, which obviously can't be read by anything else, and the other format is similar to ufs dump.
There was a time when Solaris could read ONTAP dump format, although permissions would all be lost, but that's close to 6 years since I tried that. I would expect the formats have diverged enough that it won't work any longer.
-----Original Message----- From: toasters-bounces@teaparty.net [mailto:toasters-bounces@teaparty.net] On Behalf Of John Stoffel Sent: Wednesday, September 03, 2014 2:05 AM To: Mike Horwath Cc: Toasters@teaparty.net Subject: Re: ndmpcopy from netapp to oracle zs3?
Mike> Log into the ZS3 system. Mike> rsh <netapp> dump 0f - /vol/volumename | (cd destination ; restore Mike> rf -)
Mike> Stupid but would get your copy over.
Mike> Using rsh reduces CPU load on the NeApp controller.
Have you actually tried doing this? The format of Netapp dump and ZS3 restore aren't compatible in my understanding. Luckily, I do have a Solaris 10 ZFS system I can use for testing tomorrow if I get a chance to double check.
John _______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters
On Sep 2, 2014, at 7:05 PM, John Stoffel john@stoffel.org wrote:
Mike> Log into the ZS3 system. Mike> rsh <netapp> dump 0f - /vol/volumename | (cd destination ; restore rf -)
Mike> Stupid but would get your copy over.
Mike> Using rsh reduces CPU load on the NeApp controller.
Have you actually tried doing this? The format of Netapp dump and ZS3 restore aren't compatible in my understanding. Luckily, I do have a Solaris 10 ZFS system I can use for testing tomorrow if I get a chance to double check.
No idea but I know it works with Solaris 9 and above.
— Mike - old NetApp user
-
John Stoffel -
Jordan Slingerland -
Mike Horwath -
Parisi, Justin -
Steiner, Jeffrey -
tmac