toasters

toasters@lists.teaparty.net

3 participants
13513 discussions

Rubrik - restores to NetApp -- Symlink Modtime not being restored
by John Adams 16 Apr '21

16 Apr '21

Hi All -- I have a question for others that are using Rubrik for backups.. I've noticed that when symlink's are restored.. they are restored with time/date of the restore.. not the time/date of the original.. For example: I moved test to test-hide and then restored test from Rubrik (overwrite original) . The restored Symlink has the time I did the restore.. not the original mod date. lrwxrwxrwx 1 uid gid 22 Apr 14 07:53 test-> /testfile lrwxrwxrwx 1 uid gid 22 Mar 29 2010 test-hide -> /testfile The backup was done with Netapp SnapDiff enabled. The NetApp is running 9.7P5. The Rubrik is running 5.3 Any others using Rubrik that see this? Thanks.

1 0

SnapDiff/Rubrik
by Steiner, Jeffrey 12 Apr '21

12 Apr '21

NDMP is a backup control protocol, but it's not a data format. Even ONTAP has multiple formats for NDMP backups. In the first release of NDMP, the backup was a near mirror of ufsdump, and it was pretty easy to restore an ONTAP backup to something like Solaris. You'd get some weird permissions, but it did work. The problem with ufsdump formats is the controller has to create the files and stream them off to the backup destination. About 10 years back, ONTAP added the ability to do NDMP backups using a snapmirror format. It was just the raw WAFL data, and it was WAY faster. It also meant that unless a system could real WAFL, there would be no way to use those backups. It was a nice option for full backups with a low chance of needing to be recovered, because recovery would requiring restoring the entire volume object that you backed up. Snapdiff allows a backup application to do incremental-forever backups. It could ask ONTAP to identify changed files from the previous backup and provide the list of files to be backed up to the backup app, which would be read from an NFS/SMB share. I didn't think NDMP was involved. Does the Rubrik documentation mention a NFS or SMB share path? If the ONTAP system went away someday, I would expect you could use a Cloud Volumes ONTAP instance in Azure or AWS to restore the data. There used to be some sort of limitation with NDMP and CVO, but I can't see why Snapdiff wouldn't work as usual on CVO. I'll ping a couple of guys... From: Toasters <toasters-bounces(a)teaparty.net> On Behalf Of Heino Walther Sent: Saturday, April 10, 2021 5:34 AM To: Brian Parent <bparent(a)ucsd.edu>; toasters(a)teaparty.net Subject: SV: NetApp Security WARNING: This is an external email. Do not click links or open attachments unless you recognize the sender and know the content is safe. As fas as I can remember snapdiff is using NDMP which should be a standard and open protocol, but NetApp, EMC and others have made their specific changes to their version of NDMP which basically make them not interchangeable. So I guess you will need a NetApp controller in order to restore your data. There is a slim chance that files can be restored to other NDMP devices (CIFS or NFS), but LUNs are required to be restored to a NDMP. You can either just keep a NetApp controller ready for this (with no service on it) or maybe just setup an ONTAP Select virtual instance in vSphere which you will be able to restore to, and it's not as expensive and a NetApp Hardware box. /Heino Fra: Toasters <toasters-bounces(a)teaparty.net<mailto:toasters-bounces@teaparty.net>> på vegne af Brian Parent via Toasters <toasters(a)teaparty.net<mailto:toasters@teaparty.net>> Dato: lørdag, 10. april 2021 kl. 02.12 Til: toasters(a)teaparty.net<mailto:toasters@teaparty.net> <toasters(a)teaparty.net<mailto:toasters@teaparty.net>> Emne: <no subject> _______________________________________________ Toasters mailing list Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> https://www.teaparty.net/mailman/listinfo/toasters

3 3

SV:
by Heino Walther 10 Apr '21

10 Apr '21

As fas as I can remember snapdiff is using NDMP which should be a standard and open protocol, but NetApp, EMC and others have made their specific changes to their version of NDMP which basically make them not interchangeable. So I guess you will need a NetApp controller in order to restore your data. There is a slim chance that files can be restored to other NDMP devices (CIFS or NFS), but LUNs are required to be restored to a NDMP. You can either just keep a NetApp controller ready for this (with no service on it) or maybe just setup an ONTAP Select virtual instance in vSphere which you will be able to restore to, and it’s not as expensive and a NetApp Hardware box. /Heino Fra: Toasters <toasters-bounces(a)teaparty.net> på vegne af Brian Parent via Toasters <toasters(a)teaparty.net> Dato: lørdag, 10. april 2021 kl. 02.12 Til: toasters(a)teaparty.net <toasters(a)teaparty.net> Emne: <no subject> _______________________________________________ Toasters mailing list Toasters(a)teaparty.net https://www.teaparty.net/mailman/listinfo/toasters

3 6

restoring Rubrik backups of NetApp without a NetApp
by Brian Parent 10 Apr '21

10 Apr '21

What happens if I use Rubrik to backup a NetApp (OnTap 9.7P12) using the "backup with snapdiff" method, and I need to restore something after the NetApp has been decommissioned? The Rubrik docs specify that a NetApp NAS is required as a restore target to restore from Rubrik backups made using backup with snapdiff. So, if the NetApp has been decommissioned, can I run a simulator to restore from that backup? Other options? -- Brian Parent Information Technology Services Department ITS Computing Infrastructure Operations Group its-ci-ops-help(a)ucsd.edu (team email address for Service Now) UC San Diego (858) 534-6090

1 0

Removing files directly on filer?
by Scott Classen 07 Apr '21

07 Apr '21

Hello Fellow Toasters, I have several files that have an unusual characters in their names. Here are 3 examples. 99027496 -rwxr--r--. 1 fereyes sibyls 138284 Dec 17 2012 _Volumes_fereyes_Pm12_img12_data1_pxuser_cech3_px07-0106_Pm12_7_process_mr_AutoMR_run_9__combined_refine_001.pdb_Mon_Dec_17_151710_2012_modification_0.pdb.gz 99027497 -rwxr--r--. 1 fereyes sibyls 137894 Dec 17 2012 _Volumes_fereyes_Pm12_img12_data1_pxuser_cech3_px07-0106_Pm12_7_process_mr_AutoMR_run_9__combined_refine_001.pdb_Mon_Dec_17_151710_2012_modification_1.pdb.gz 99027498 -rwxr--r--. 1 fereyes sibyls 137467 Dec 17 2012 _Volumes_fereyes_Pm12_img12_data1_pxuser_cech3_px07-0106_Pm12_7_process_mr_AutoMR_run_9__combined_refine_001.pdb_Mon_Dec_17_151710_2012_modification_2.pdb.gz I would like to delete them. I have tried all the tricks I know of to delete them from the NFS client side including the usually-bullet-proof method of deleting them by inode: ls -li To get the inode then: find . -inum 99027498 -exec rm -i {} \; and this is not working. Is there some way to remove folders/files from the NetApp itself? Thanks, Scott

5 5

Re:
by Tim McCarthy 26 Mar '21

26 Mar '21

That sounds like a service processor issue I ran into it on a fas2650 and an a220 where the cluster ports would not link. Dropped to the sp and power cycled the node. As soon as I did that the ports came back Get Outlook for iOS<https://aka.ms/o0ukef> ________________________________ From: Toasters <toasters-bounces(a)teaparty.net> on behalf of Brian Parent via Toasters <toasters(a)teaparty.net> Sent: Friday, March 26, 2021 3:38:39 AM To: Alexander Griesser <AGriesser(a)anexia-it.com> Cc: toasters(a)teaparty.net <toasters(a)teaparty.net> Subject: _______________________________________________ Toasters mailing list Toasters(a)teaparty.net https://www.teaparty.net/mailman/listinfo/toasters

1 0

NICs fail to connect after upgrade to OnTap 9.3P21
by Brian Parent 26 Mar '21

26 Mar '21

I upgraded our AFF8020 (2 nodes) yesterday from 9.1P10 to 9.3P21, as a first step towards 9.7P12. Unfortunately, the two 10Gb interfaces (e0c, e0d) on the second controller won't connect after the upgrade. It seems like a layer one problem, with no link lights on either the NetApp, or the switch side. Yet it seems unlikely that we have two physical paths fail at the same time, and too coincidental that it happened (apparently) during the giveback. And, reseating the fiber jumpers, as well as the SFP optics hasn't helped. Has anyone else seen anything like this? I've engaged (third party) support, but so far, no joy. I've tried shut/no shut on the switch side, (Cisco 9K) also tried the same sort of thing on the NetApp side (advanced mode: network port modify -node <nodename> -port <portname> -up-admin false, on both ports, then back to true) Also, the e0c and e0d were bundled into a if_group (a0a) on the NetApp side, and a VPC port channel on the Cisco side, (identically on both nodes, so the config seems fine since it still works on node1. I tried deleting node2's a0a, but that failed due to a lif having set it's home to that a0a. So, I migrated that lif "permanently" to the working node, and then deleted node2's a0a, and then tried the -up-admin=false/true trick again on the e0c and e0d ports, but still no joy. I noticed the MTU was set to 9216 on the Cisco side and 1500 on the NetApp side, but again, this is identical on both nodes, and is working on node1, so even though it looks suspicious, it's probably not the problem. I tried setting the NetApp side to 9000 (it didn't accept 9216), but that didn't help either. I thinking my next step is to delete node1's a0a, and remove the corresponding port channels on the Cisco side, then try swapping known good fibers and optics to the problem ports to which piece is the root cause. What does the toasters brain trust say to all this? -- Brian Parent Information Technology Services Department ITS Computing Infrastructure Operations Group its-ci-ops-help(a)ucsd.edu (team email address for Service Now) UC San Diego (858) 534-6090

2 2

New BURT in Ontap 9.8P1, unexpected Node Reboot
by Josef Kropf 25 Mar '21

25 Mar '21

Good morning, Seems like we've run into a new BUG last weekend with one of our A300 systems: BURT 1380642 http://mysupport.netapp.com/NOW/cgi-bin/bol?Type=Detail <http://mysupport.netapp.com/NOW/cgi-bin/bol?Type=Detail&Display=1380642> &Display=1380642 Node reboots unexpectedly, happened after 3 weeks being productive on Sunday evening at exactly 10 PM No explanation from NA until now, bug is "under further investigation" Currently no known recommended solution. Also there is no workaround at the moment. Br Josef

5 4

AW: Re: Move fabric pool enabled aggregate to new disks
by Alexander Griesser 22 Mar '21

22 Mar '21

Ah, that’s interesting, seem to have missed that. Thanks for the pointer, looks exactly like what we’re looking for 😊 Best, Alexander Griesser Head of Systems Operations ANEXIA Internetdienstleistungs GmbH E-Mail: AGriesser(a)anexia-it.com<mailto:AGriesser@anexia-it.com> Web: http://www.anexia-it.com<http://www.anexia-it.com/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 Von: Tim Stiller <tim.stiller(a)gmail.com> Gesendet: Montag, 22. März 2021 14:09 An: Alexander Griesser <AGriesser(a)anexia-it.com> Cc: toasters(a)teaparty.net Betreff: Re: Move fabric pool enabled aggregate to new disks Hi Alexander, if you are on ONTAP 9.6+ and you want to use the same S3 bucket with your new disks, vol move will only move the hot data to the new aggregates. The cold data will remain on the cloud tier. just take a look at this KB: https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTA… best regards Tim Am Mo., 22. März 2021 um 14:04 Uhr schrieb Alexander Griesser <AGriesser(a)anexia-it.com<mailto:AGriesser@anexia-it.com>>: Hey there, we’ve quite some fabric pool enabled aggregates and would want to move them to different disks. As far as I can see, there’s no good way to do that. Using vol moves will require to download all the data from our S3 backend and re-upload it on the destination aggregate then. So say we want to „upgrade“ our fabric pool aggregate from 12x1TB SSD to 12x4TB SSD and let’s further assume that we have 20 volumes on the FP-enabled aggregate with a total of 1PB of storage in the S3 backend, do we really have to go through downloading and re-uploading all the data or is there an easier way (besides adding the new disks to the aggregate in the first step as spare disks and manually failing disk by disk) and afterwards increase the capacity in the raidgroup by addiding additional partitions? Best, Alexander Griesser Head of Systems Operations ANEXIA Internetdienstleistungs GmbH E-Mail: AGriesser(a)anexia-it.com<mailto:AGriesser@anexia-it.com> Web: http://www.anexia-it.com<http://www.anexia-it.com/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 _______________________________________________ Toasters mailing list Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> https://www.teaparty.net/mailman/listinfo/toasters Am Mo., 22. März 2021 um 14:04 Uhr schrieb Alexander Griesser <AGriesser(a)anexia-it.com<mailto:AGriesser@anexia-it.com>>: Hey there, we’ve quite some fabric pool enabled aggregates and would want to move them to different disks. As far as I can see, there’s no good way to do that. Using vol moves will require to download all the data from our S3 backend and re-upload it on the destination aggregate then. So say we want to „upgrade“ our fabric pool aggregate from 12x1TB SSD to 12x4TB SSD and let’s further assume that we have 20 volumes on the FP-enabled aggregate with a total of 1PB of storage in the S3 backend, do we really have to go through downloading and re-uploading all the data or is there an easier way (besides adding the new disks to the aggregate in the first step as spare disks and manually failing disk by disk) and afterwards increase the capacity in the raidgroup by addiding additional partitions? Best, Alexander Griesser Head of Systems Operations ANEXIA Internetdienstleistungs GmbH E-Mail: AGriesser(a)anexia-it.com<mailto:AGriesser@anexia-it.com> Web: http://www.anexia-it.com<http://www.anexia-it.com/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 _______________________________________________ Toasters mailing list Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> https://www.teaparty.net/mailman/listinfo/toasters

1 0

Move fabric pool enabled aggregate to new disks
by Alexander Griesser 22 Mar '21

22 Mar '21

Hey there, we've quite some fabric pool enabled aggregates and would want to move them to different disks. As far as I can see, there's no good way to do that. Using vol moves will require to download all the data from our S3 backend and re-upload it on the destination aggregate then. So say we want to "upgrade" our fabric pool aggregate from 12x1TB SSD to 12x4TB SSD and let's further assume that we have 20 volumes on the FP-enabled aggregate with a total of 1PB of storage in the S3 backend, do we really have to go through downloading and re-uploading all the data or is there an easier way (besides adding the new disks to the aggregate in the first step as spare disks and manually failing disk by disk) and afterwards increase the capacity in the raidgroup by addiding additional partitions? Best, Alexander Griesser Head of Systems Operations ANEXIA Internetdienstleistungs GmbH E-Mail: AGriesser(a)anexia-it.com<mailto:AGriesser@anexia-it.com> Web: http://www.anexia-it.com<http://www.anexia-it.com/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601

2 1

← Newer
1
...
8
9
10
11
12
13
14
...
1352
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters