toasters August 2022

toasters@lists.teaparty.net

20 participants
8 discussions

Does NetApp still sell physical file servers?
by Turritopsis Dohrnii Teo En Ming 19 Aug '22

19 Aug '22

Subject: Does NetApp still sell physical file servers? Good day from Singapore, Does NetApp still sell physical file servers? I read from Wikipedia that NetApp, Inc. is an American hybrid cloud data services and data management company headquartered in San Jose, California. It has ranked in the Fortune 500 from 2012–2021. Founded in 1992 with an IPO in 1995, NetApp offers cloud data services for management of applications and data both online and physically. I see that NetApp has ventured into the cloud. Thank you. Regards, Mr. Turritopsis Dohrnii Teo En Ming Targeted Individual in Singapore 12 Aug 2022 Fri Blogs: https://tdtemcerts.blogspot.com https://tdtemcerts.wordpress.com

11 13

Cluster Node Distance
by Heino Walther 09 Aug '22

09 Aug '22

Hi there We are in the process of migrating from one DC to another DC which is about 2KM appart. We have a standard HA-Two node A300 which we would like to migrate out to the new DC without any down time. We have several 10G lines between the two sites with Cisco networking equipment. I seem to recall that officielly NetApp only supports about 3-400M between the cluster nodes, but I cannot see why it should not be possible to extend this distance to the 2KM we need? I doubt the systems will notice. This is of cause only as a part of the migration process, so we setup another A300 cluster with disks at the new DC and add them to the existing cluster over the 2KM distance, then move over the volumes and interfaces nice an easy. When we have moved everything, we remove the two (now empty) nodes to the new DC, join them to the cluster and move the data back again. Has anyone tried this over this kind of distances? We have official CN1610 but after looking at the Cisco RFC’s from NetApp I think we might as well just create this network on our own Cisco equipment. If this is a no-go, are there any other options, short of converting it into a MetroCluster? 😉 /Heino

5 8

CVO on Standard/SATA PD
by Jeff Mohler 06 Aug '22

06 Aug '22

Does anyone here have experience on a standard PD layer using CVO? If so, I am looking for -throughput- maximus, limitations, lessons learned, on per volume write workloads. We are in a mode where we believe we were told there are 100-200MB/sec write "edges" of throughput to CVO on standard PD storage...just we're not quite in shape here to test that yet ourselves. _________________________________ Jeff Mohler <jmohler(a)yahooinc.com> Sr. Storage Architect & Principal Yahoo Storage Engineering & Operations (831)454-6712 It puts the bundles in the buckets, or else it gets the hose again.

2 1

Re:
by André M. Clark 06 Aug '22

06 Aug '22

Check to see if your volumes are thin provisioned on the destination side. By default, they would not be thin provisioned. Regards, André M. Clark On August 5, 2022 at 11:03:13, Timothy Naple via Toasters ( toasters(a)teaparty.net) wrote: My first guess would be that you might somehow be retaining more snapshots on the destination, perhaps at the volume level? snap list for all the volumes is identical on source and destination? Does df -h on both sides show which volumes, if any, have a size discrepancy? Maybe a 2nd guess would be that the space guarantee settings are different on the destination volumes. ------------------------------ *From:* Toasters <toasters-bounces(a)teaparty.net> on behalf of Heino Walther <hw(a)beardmann.dk> *Sent:* Friday, August 5, 2022 7:31 AM *To:* toasters(a)teaparty.net <toasters(a)teaparty.net> *Subject:* Space issues on older NetApp... Hi there We have two systems that mirror eachothers volumes via snapmirror. We are talking 7mode ONTAP 8.1.4 The two systems have the same controller: FAS3240 They have the same disks and aggregate configuration (70TB aggregates) On the source side we use volumes that are thin-provisioned with LUNs that have space reservation enabled, the LUNs are mostly close to 16TB (which is max) All volumes are snapmirrored to volumes on the destination system with the same size and placed on the same aggregates that mirror the source aggregates in size… The aggregates on the source are all below 95% used. Yet.. we are now at the situation where a few destination aggregates are 100% full, while the source aggregates are still under 95% used… I have checked almost everything, like aggregate snapshot reserves etc.. but they should be the same… Can anyone explain why this can happen? We are of cause at a “deadlock” now.. I don’t think we can add anymore disks to the aggregates as they are max size… The only think I can think of is either delete a volume from the affected aggregates, and re-sync the volume and hope it doesn’t fill up again… Another way would be to add disks and build a new aggregate, and move some of the volumes… Is there something I have missed? 😊 /Heino _______________________________________________ Toasters mailing list Toasters(a)teaparty.net https://www.teaparty.net/mailman/listinfo/toasters

2 2

Space issues on older NetApp...
by Heino Walther 06 Aug '22

06 Aug '22

Hi there We have two systems that mirror eachothers volumes via snapmirror. We are talking 7mode ONTAP 8.1.4 The two systems have the same controller: FAS3240 They have the same disks and aggregate configuration (70TB aggregates) On the source side we use volumes that are thin-provisioned with LUNs that have space reservation enabled, the LUNs are mostly close to 16TB (which is max) All volumes are snapmirrored to volumes on the destination system with the same size and placed on the same aggregates that mirror the source aggregates in size… The aggregates on the source are all below 95% used. Yet.. we are now at the situation where a few destination aggregates are 100% full, while the source aggregates are still under 95% used… I have checked almost everything, like aggregate snapshot reserves etc.. but they should be the same… Can anyone explain why this can happen? We are of cause at a “deadlock” now.. I don’t think we can add anymore disks to the aggregates as they are max size… The only think I can think of is either delete a volume from the affected aggregates, and re-sync the volume and hope it doesn’t fill up again… Another way would be to add disks and build a new aggregate, and move some of the volumes… Is there something I have missed? 😊 /Heino

6 11

experience with standard PD layer using CVO?
by John Stoffel 06 Aug '22

06 Aug '22

>>>>> "Jeff" == Jeff Mohler via Toasters <toasters(a)teaparty.net> writes: Jeff> From: Jeff Mohler <jmohler(a)yahooinc.com> Jeff> To: Toasters <toasters(a)teaparty.net> Jeff> Date: Fri, 5 Aug 2022 10:15:13 -0700 Jeff> Subject: CVO on Standard/SATA PD Jeff> Does anyone here have experience on a standard PD layer using CVO? Jeff, I know you're deep in the weeds, but for those of us who aren't... can you spell out those acronyms please? PD = Purple Dinosaur? CVO = Cheery Victory Orange? *grin* Jeff> If so, I am looking for -throughput- maximus, limitations, Jeff> lessons learned, on per volume write workloads. Jeff> We are in a mode where we believe we were told there are Jeff> 100-200MB/sec write "edges" of throughput to CVO on standard PD Jeff> storage...just we're not quite in shape here to test that yet Jeff> ourselves. Yeah, you need to translate this section as well please. Or at least give a little more context so people have a hope of understanding. Or better yet, learning from your experiences! Cheers, John

4 4

Re: [E] Re: [EXTERNAL] experience with standard PD layer using CVO?
by Michael Bergman 06 Aug '22

06 Aug '22

There are such barriers w.r.t. (total) throughput from a VM to PD (and corresponding in AWS e.g.). But I very much doubt you'll hit those max throughput with CVO instances in any case. The problem would be 7.2K rpm back end: so very slow. No matter if you have a FAS appliance or a CVO, really. Then also CVO is a dog for WRITE pretty much in any case, always has been. Probably always will be, compared to real on-prem HW (FAS appliances). That's my $0.05 worth I *think* that what you're asking is in essence if anyone has been able to "max out" a theoretical 7.2K rpm based back in in a Hyperscalar Cloud with CVO. Can one do that with anything at all..? Some micro benchmark (fio, iozone)? I don't know. Never thought of that really, had no reason to ponder over it /M On 2022-08-05 23:10, Jeff Mohler via Toasters wrote: > Looking to see if any MB/sec on writes to sata physical disk backends have > been touched. > > Hearing that there may be (in some clouds) a write throughout barrier that > CVO may run into.

1 0

Re:
by Jeff Mohler 05 Aug '22

05 Aug '22

Whoops. PD=Persistent disk. Not physical. We don’t yet have vapor disks. :) On Fri, Aug 5, 2022 at 14:16 Jeff Mohler via Toasters <toasters(a)teaparty.net> wrote: > > > > ---------- Forwarded message ---------- > From: Jeff Mohler <jmohler(a)yahooinc.com> > To: Scott Miller <scott.miller(a)dreamworks.com> > Cc: John Stoffel <john(a)stoffel.org>, Toasters <toasters(a)teaparty.net> > Bcc: > Date: Fri, 5 Aug 2022 14:10:15 -0700 > Subject: Re: [E] Re: [EXTERNAL] experience with standard PD layer using > CVO? > Cloud volumes ontap > Physical disk > > :) > > Looking to see if any MB/sec on writes to sata physical disk backends have > been touched. > > Hearing that there may be (in some clouds) a write throughout barrier that > CVO may run into. > > On Fri, Aug 5, 2022 at 13:46 Scott Miller <scott.miller(a)dreamworks.com> > wrote: > >> >> CVO == Cloud Volumes Ontap, the software version of OnTap that you can >> run in the >> various CSP's (Cloud Service Providers) aka GCP, AWS, Azure >> >> https://cloud.netapp.com/ontap-cloud >> <https://urldefense.com/v3/__https://cloud.netapp.com/ontap-cloud__;!!Op6efl…> >> >> PD == Persistent Disk, what GCP (Google Compute Platform) calls its, uh, >> persistent disks. >> >> https://cloud.google.com/compute/docs/disks >> <https://urldefense.com/v3/__https://cloud.google.com/compute/docs/disks__;!…> >> >> And to answer Jeffs question, we've not done any CVO testing in GCP, so I >> can't >> answer the throughput questions. >> >> We have found that, in general, the VM Instance type that hosts the CVO >> instances matters, >> since the Instance type also determines networking connectivity. >> >> Disk type (in this case GCP PDs) matters, too; I'd test with Local SSD to >> see where >> that gets you. >> >> Lastly, Throughput is dependent on workload, which determines the IO >> pattern. >> Maybe create an FIO config that generates traffic close to what you >> expect, and >> use that to drive various CVO configurations. >> >> https://github.com/axboe/fio >> <https://urldefense.com/v3/__https://github.com/axboe/fio__;!!Op6eflyXZCqGR5…> >> >> -Skottie >> >> >> >> >> On Fri, Aug 5, 2022 at 1:26 PM John Stoffel <john(a)stoffel.org> wrote: >> >>> >>>>> "Jeff" == Jeff Mohler via Toasters <toasters(a)teaparty.net> writes: >>> >>> Jeff> From: Jeff Mohler <jmohler(a)yahooinc.com> >>> Jeff> To: Toasters <toasters(a)teaparty.net> >>> Jeff> Date: Fri, 5 Aug 2022 10:15:13 -0700 >>> Jeff> Subject: CVO on Standard/SATA PD >>> >>> Jeff> Does anyone here have experience on a standard PD layer using CVO? >>> >>> Jeff, I know you're deep in the weeds, but for those of us who >>> aren't... can you spell out those acronyms please? >>> >>> PD = Purple Dinosaur? >>> CVO = Cheery Victory Orange? >>> >>> *grin* >>> >>> >>> Jeff> If so, I am looking for -throughput- maximus, limitations, >>> Jeff> lessons learned, on per volume write workloads. >>> >>> Jeff> We are in a mode where we believe we were told there are >>> Jeff> 100-200MB/sec write "edges" of throughput to CVO on standard PD >>> Jeff> storage...just we're not quite in shape here to test that yet >>> Jeff> ourselves. >>> >>> Yeah, you need to translate this section as well please. Or at least >>> give a little more context so people have a hope of understanding. Or >>> better yet, learning from your experiences! >>> >>> Cheers, >>> John >>> >>> _______________________________________________ >>> Toasters mailing list >>> Toasters(a)teaparty.net >>> https://www.teaparty.net/mailman/listinfo/toasters >>> <https://urldefense.com/v3/__https://www.teaparty.net/mailman/listinfo/toast…> >>> >> -- > If it were me flying, Goose would be alive today. > > > > ---------- Forwarded message ---------- > From: Jeff Mohler via Toasters <toasters(a)teaparty.net> > To: Scott Miller <scott.miller(a)dreamworks.com> > Cc: Toasters <toasters(a)teaparty.net> > Bcc: > Date: Fri, 5 Aug 2022 22:10:33 +0100 > Subject: > _______________________________________________ > Toasters mailing list > Toasters(a)teaparty.net > > https://urldefense.com/v3/__https://www.teaparty.net/mailman/listinfo/toast… -- If it were me flying, Goose would be alive today.

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 August 2022