toasters March 2021

toasters@lists.teaparty.net

24 participants
14 discussions

2node system, migration to a new box?
by Sami Kapanen 21 May '21

21 May '21

We have 2node FAS2520 (no switches). How to migrate SVMs to a new cluster? Want to preserve CIFS server names and IP addresses. Last time we used 7MTT to do the migrations.. is there similar tool for cluster mode migrations? br, sk

4 5

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

Remote backup hosting with ONTAP...
by Heino Walther 09 Mar '21

09 Mar '21

Hi there I need some suggestions to a setup where we will be hosing remote backup of other ONTAP clusters on one ONTAP FAS cluster. The clusters that needs to be backed up should be backed up are on separate VLANS and should be kept separate. So the plan is to create a SVM per system in its own VLAN. Trouble is that in order to setup a SVM-Peer you first need to setup a cluster peer… So I guess there are several ways to do this… 1. Create several cluster mgmt. LIFs (one for each VLAN) * Could this cause issues? 2. Create a firewall rule for the existing cluster mgmt. LIF so is can be reached from all the VLANs. I will also be using ipspaces for this because of security, and because there are two systems using the same IP range… Suggestions are welcome.. I somewhat lean towards option 1. /Heino Heino Walther<https://www.linkedin.com/in/heinowalther/> Beardmann ApS<http://beardmann.dk/> Jellingvej 9 - 7100 Vejle<https://goo.gl/maps/xQVPFMHXpXu> D: 7199 9060 M: 2075 7501 --

5 6

Re:
by tmac 04 Mar '21

04 Mar '21

That is correct and wrong at the same time ;) You still cannot go directly from 9.3 to 9.7. I have heard you can load up 9.5 and 9.7 images and it will progress through: 9.3 -> 9.5 and then 9.5 -> 9.7 Still a double hop, but somewhat automated. Personally, I would still do the 9.3 -> 9.5, wait at least 24 hours and then do the 9.5 -> 9.7 --tmac *Tim McCarthy, **Principal Consultant* *Proud Member of the #NetAppATeam <https://twitter.com/NetAppATeam>* *I Blog at TMACsRack <https://tmacsrack.wordpress.com/>* On Wed, Mar 3, 2021 at 8:44 PM Brian Parent via Toasters < toasters(a)teaparty.net> wrote: > > > > ---------- Forwarded message ---------- > From: Brian Parent <bparent(a)ucsd.edu> > To: Heino Walther <hw(a)beardmann.dk> > Cc: "toasters(a)teaparty.net" <toasters(a)teaparty.net> > Bcc: > Date: Wed, 3 Mar 2021 17:41:48 -0800 > Subject: Re: > I'm hearing conflicting information about being limited to 2 versions per > hop. > Our support team says once we're at 9.3, we should be able to go directly > to 9.7. > > I guess we'll find out, and I can update this list on how it goes. > The update is scheduled for late March. > > Re: > > From: Heino Walther <hw(a)beardmann.dk> > > Date: Fri, 5 Feb 2021 19:08:57 +0000 > > Subject: SV: > > To: Brian Parent <bparent(a)ucsd.edu>, "toasters(a)teaparty.net" > > <toasters(a)teaparty.net> > > > > I would recommend 9.7p10 > > > > Be aware that you can only jump 2 versions… so it’s 9.1 to 9.3 to 9.5 to > 9.7 😉 > > I would recommend a pause between each version, and get the latest > patched version also… > > I seem to remember one specific version where the cluster certificate in > the downloaded ontap version was expired, which resulted in an upgrade of > the cluster never completing… no downtime or anything, but NetApp support > had to fix the cluster database… 😊 > > > > I have been working with NetApp for 15 years and never done a real > production revert in my life 😉. But maybe I’m just lucky? > > > > /Heino > > > > > > I've got an AFF8020 running 9.1P10 and wondering whether I should target > 9.6P10 or 9.7P10 as the destination version. > > I'm aware that such an upgrade will require a hop to 9.3 first. > > I'm also aware that reversion can not skip a release, which is a little > scary. I *really* hope I don't have to revert > > from 9.6 or 9.7 all the way back to 9.3 (3 or 4 hops). > > > > What I'm looking for is advice on 9.6p10 vs. 9.7p10. Any horror stories > of either release? > > I'm leaning towards 9.7p10, just because it's more recent. > > > > > > Fra: Toasters <toasters-bounces(a)teaparty.net> på vegne af Brian Parent > via Toasters <toasters(a)teaparty.net> > > Dato: fredag, 5. februar 2021 kl. 19.22 > > Til: toasters(a)teaparty.net <toasters(a)teaparty.net> > > Emne: <no subject> > > _______________________________________________ > > Toasters mailing list > > Toasters(a)teaparty.net > > > https://urldefense.com/v3/__https://www.teaparty.net/mailman/listinfo/toast… > > -- > 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 > > > > ---------- Forwarded message ---------- > From: Brian Parent via Toasters <toasters(a)teaparty.net> > To: Heino Walther <hw(a)beardmann.dk> > Cc: "toasters(a)teaparty.net" <toasters(a)teaparty.net> > Bcc: > Date: Thu, 4 Mar 2021 01:42:00 GMT > Subject: > _______________________________________________ > Toasters mailing list > Toasters(a)teaparty.net > https://www.teaparty.net/mailman/listinfo/toasters

2 1

(no subject)
by Iluhes 04 Mar '21

04 Mar '21

I heard 9.7p11 is the way to go On Wednesday, March 3, 2021, 07:52:07 PM CST, Brian Parent via Toasters <toasters(a)teaparty.net> wrote: I'm hearing conflicting information about being limited to 2 versions per hop. Our support team says once we're at 9.3, we should be able to go directly to 9.7. I guess we'll find out, and I can update this list on how it goes. The update is scheduled for late March. Re: > From: Heino Walther <hw(a)beardmann.dk> > Date: Fri, 5 Feb 2021 19:08:57 +0000 > Subject: SV: > To: Brian Parent <bparent(a)ucsd.edu>, "toasters(a)teaparty.net" > <toasters(a)teaparty.net> > > I would recommend 9.7p10 > > Be aware that you can only jump 2 versions… so it’s 9.1 to 9.3 to 9.5 to 9.7 😉 > I would recommend a pause between each version, and get the latest patched version also… > I seem to remember one specific version where the cluster certificate in the downloaded ontap version was expired, which resulted in an upgrade of the cluster never completing… no downtime or anything, but NetApp support had to fix the cluster database… 😊 > > I have been working with NetApp for 15 years and never done a real production revert in my life 😉. But maybe I’m just lucky? > > /Heino > > > I've got an AFF8020 running 9.1P10 and wondering whether I should target 9.6P10 or 9.7P10 as the destination version. > I'm aware that such an upgrade will require a hop to 9.3 first. > I'm also aware that reversion can not skip a release, which is a little scary. I *really* hope I don't have to revert > from 9.6 or 9.7 all the way back to 9.3 (3 or 4 hops). > > What I'm looking for is advice on 9.6p10 vs. 9.7p10. Any horror stories of either release? > I'm leaning towards 9.7p10, just because it's more recent. > > > Fra: Toasters <toasters-bounces(a)teaparty.net> på vegne af Brian Parent via Toasters <toasters(a)teaparty.net> > Dato: fredag, 5. februar 2021 kl. 19.22 > Til: toasters(a)teaparty.net <toasters(a)teaparty.net> > Emne: <no subject> > _______________________________________________ > Toasters mailing list > Toasters(a)teaparty.net > https://urldefense.com/v3/__https://www.teaparty.net/mailman/listinfo/toast… -- 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 _______________________________________________ Toasters mailing list Toasters(a)teaparty.net https://www.teaparty.net/mailman/listinfo/toasters

1 0

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 March 2021