toasters

toasters@lists.teaparty.net

13531 discussions

options weirdness
by Peter D. Gray 13 Aug '13

13 Aug '13

Can someone who knows tell me whats going on here? We recently discovered a security issue with our netapp filers when using NFS. The netapps where allowing NFS mounts and operations from non-privilaged client ports. Investigation found the options nfs.nfs_rootonly and nfs.mount_rootonly options. nfs.mount_rootonly was true, but nfs.nfs_rootonly was false. I was kind of surprized by this, since I had no recollection of ever altering either of these options yet the settings were identical across all our filers. The environment is: 3170A running 8.2 3240 running 8.2 3240 running 8.1 (reverted from 8.2) Ok, so I asked another nearby site to check there filers for me and they tell me their filers have no such setting, all on 8.1 So, I guess my questions are: 1) has ONTAP always allowed NFS from non-privilaged ports? 2) was nfs.nfs_rootonly introduced in 8.2 and why is the default off? 3) why does this setting stay around after revert to 8.1? It would seem to me that allowing NFS from non-privilages ports is kind of bad. Any help appreciated. Regards, pdg

4 7

vFiler DR and Oncommand Core (DFM)
by Martin 09 Aug '13

09 Aug '13

We've been using vFiler DR for some time now without any issues. Recently we noticed that some of our qtrees on some new vFilers were not being discovered in Oncommand Core (5.1). After speaking with Netapp support they highlighted that there is a known issue (no bug or KB article AFAIK) where the passive DR vFilers cause exactly the issue we are seeing. On their advice we deleted the DR vFilers from Oncommand Core via the DFM CLI. As soon as we did this the new qtrees on the vFilers were detected. However when we flipped our vFilers from one site to another using vFiler DR: 1. Stop the active vFiler. 2. Resync the previously active to the DR vFiler. 3. Activate the DR vFiler 4. Cleanup snapmirrors and reverse snapmirror. We found that none of the volumes on the vFilers that were activated could be seen. On checking the vFilers on DFM we could see the active vFiler and several deleted instances of the DR vFilers. When I attempted to delete the vFilers so that I could readd them I got an SQL error as they were references by views and other objects in the DFM database. Netapp support then pointed me to this article: https://kb.netapp.com/support/index?page=content&id=3012455 which mentions: “With the ONTAP vFiler DR feature, it is possible to have two instances of the exact same path; one path on the active vFiler and one path on the standby vFiler.”. The article relates to Protection Manager but the root cause is the same. Support then mentioned that Vfiler DR appears to be incompatible with Oncommand Core. Has anyone else experienced this or have a workaround for it? Currently we would have to flip the vFilers, delete the vFilers, purge the database (after taking backup) and then restart the database so that it can rediscover the vFilers. This isn't workable but really don't want to go back to manually destroying and creating vFilers as it's time consuming and error prone. -- View this message in context: http://network-appliance-toasters.10978.n7.nabble.com/vFiler-DR-and-Oncomma… Sent from the Network Appliance - Toasters mailing list archive at Nabble.com.

1 0

7.3.4 10gb question
by David Byte 08 Aug '13

08 Aug '13

Hey folks, I have a lab box that is a 3020 with 7.3.4.(I know, I should go 7.3.7) and I have an x1005a-r5 (chelsio 10gb) in it. The network config is a 2 layer vif with leg1 being a multi of 4-1gb ports and leg2 being the 10gb nic. The 2 legs are in a single mode vif favoring the 10gb. All ports are connected to a stack of extreme networks summit 400-48t switches. When running over the 10Gb leg, I cannot get any jumbo frame packets to show crossing the switch. However, if I flip to leg1 (the multi of 1gb) jumbo frames are flying across. It's tough for me to admit as a former Netapp SE, but I am stumped. Any thoughts or suggestions? Sent from my iPhone. Typos are Apple's fault.

3 3

RPC error during ndmpcopy
by Jeff Cleverley 08 Aug '13

08 Aug '13

Greetings, I'm getting the following error when I run a ndmpcopy from one filer to another. Both are 8.1.2P4. Command: ndmpcopy -sa ndmpuser:QIxwKig5bnpwdAyB -d -l 0 source:/vol/old_sd28/Gold /vol/restore_test/Gold Error: Ndmpcopy: source: Log: DATA: Backup terminated: backup configuration failure (2) on RPC error: ERROR: BAD ENV OP (for /vol/old_sd28/Gold) I've looked in the messages, ndmpd, ems, and other logs and can't find anything useful. I thought using the -d was turning on debugging so I'm not sure what else to try. I assume it is telling me there is a bad environmental option but I don't know what it is. Ultimately I'm trying to find the fastest way to get data from a secondary filer that is a snapvault destination back to the primary filers. We don't currently have the SnapRestore licenses on these heads. Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

1 0

Snapmirror throttle not working
by Jeff Cleverley 06 Aug '13

06 Aug '13

Greetings, I'm running 8.1.2P4, 7-mode on some 6290s. I need to do some snapmirrors to re-balance some disk space. The -k option to throttle the transfer doesn't seem to be having any effect. I've tried modifying the placement of the -k but it doesn't seem to matter. I also tried to modify it after it was running and it doesn't seem to help either. Here is the command I'm running: snapmirror initialize -S sm15_3 -k 10000 new_sm15_3 If I'm understanding correctly, this should be allowing 10MB/s. The source and destination are on the same file system. Here is a cut of a sysstat 3 after starting it: CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache in out read write read write age 73% 598 0 0 2091 349080 338016 271 0 0 7 71% 1019 0 0 2046 319892 324019 13114 0 0 0s 71% 2800 0 0 3880 330527 343528 17379 0 0 7 69% 1440 0 0 3405 330279 392647 22343 0 0 0s 87% 1614 0 0 2128 320151 607753 168553 0 0 0s 87% 827 0 0 5652 244701 584436 371689 0 0 0s 91% 897 0 0 4242 344072 680454 386373 0 0 0s As you can see, the disk read/write counts go way up. This is causing some noticeable latency in the nfs access for clients. While I really like the new hardware can pump data around, I need to be able to control it. What am I doing wrong? Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

7 13

Re: updating disk qual files on many controllers
by Jack Lyons 05 Aug '13

05 Aug '13

You can use unc's and not worry drive letters Jack ------Original Message------ From: Scott Eno To: Jack Lyons Cc: Jordan Slingerland Cc: toasters-bounces(a)teaparty.net Cc: tmac Cc: Toasters Subject: Re: updating disk qual files on many controllers Sent: Aug 5, 2013 11:03 AM Correct. Not enough letters in the alphabet and my dfm host is win2008, although I cheated a bit by installing Cygwin and associated sshd service. ---- Scott Eno s.eno(a)me.com On Aug 5, 2013, at 11:00 AM, Jack Lyons <jack1729(a)gmail.com> wrote: > You can do this similarly with cifs and unc's. I assume the significance of 30+ is due to drive mappings? > Sent from my Verizon Wireless BlackBerry > > -----Original Message----- > From: Jordan Slingerland <Jordan.Slingerland(a)independenthealth.com> > Sender: toasters-bounces(a)teaparty.net > Date: Mon, 5 Aug 2013 14:53:24 > To: tmac<tmacmd(a)gmail.com>; Scott Eno<s.eno(a)me.com> > Cc: Toasters<Toasters(a)teaparty.net> > Subject: RE: updating disk qual files on many controllers > > _______________________________________________ > Toasters mailing list > Toasters(a)teaparty.net > http://www.teaparty.net/mailman/listinfo/toasters > > Sent from my Verizon Wireless BlackBerry

4 3

updating disk qual files on many controllers
by Scott Eno 05 Aug '13

05 Aug '13

Hi, I've got an environment of 30+ controllers, some with NFS licensed and some without, but all with CIFS. I'm trying to figure out a way to push drive qual files (and eventually the firmware files) to all the controllers without going one-at-a-time (batch vs. by hand). 30+ means I can't mount them all via CIFS at the same time, and with half of them without NFS licensed I can't mount them via NFS to, say, a linux host. I was looking into using DFM to push the files to all the controllers, but it doesn't seem to work with anything but the config files it can pull off the controllers first. Anyone have a favorite method?

6 7

Recommendations for adding new bigger disk shelves to existing system
by Alexander Griesser 03 Aug '13

03 Aug '13

Hey there, I just got two new DS4243 450GB 15k SAS Shelves and am about to add them to our FAS3240HA. Right now, 4 DS4243 300GB 15k SAS disks are attached to this HA pair, 2 shelves for each controller. Since 300GB disks cannot be purchased anymore, we got the 450GB shelves and I'm reluctant to just "throw away" about 5TB of usable SAS space by downsizing the 450GB disks to 300GB when adding them to the existing aggregates, or does this only apply to RAID groups? So if I create new RAID groups in the existing aggregates, will I then be able to use the "full" [0] size of the 450GB disks? Thanks, [0] After rightsizing and all other nice things NetApp does to the full capacity of the disks :) Alexander Griesser System-Administrator ANEXIA Internetdienstleistungs GmbH Telefon: +43-463-208501-320 Telefax: +43-463-208501-500 E-Mail: ag(a)anexia.at<mailto:ag@anexia.at> Web: http://www.anexia.at Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601

3 5

Oncommand Core(DFM) thresholds for thin provisioning
by Martin 02 Aug '13

02 Aug '13

Does anyone know of Netapp best practise docs or other references on configuring monitoring on aggregates that have thin provisioned volumes: Specifically: Aggregate Overcommitted Threshold (%) Aggregate Nearly Overcommitted Threshold (%) Aggregate Nearly Full Threshold (%) I realise this depends on the rate of change of the data i.e allowing enough time to provision more storage or reduce usage. I am not clear on some of the detail though. Any pointers would be appreciated. Martin -- View this message in context: http://network-appliance-toasters.10978.n7.nabble.com/Oncommand-Core-DFM-th… Sent from the Network Appliance - Toasters mailing list archive at Nabble.com.

3 3

3270 Powersupply degraded, then OK ?
by Fletcher Cocquyt 30 Jul '13

30 Jul '13

Anyone seen this before? Status is OK now, but if this is a sign its about to fail should I proactively replace this PS? (8.1.2) Jul 29 12:03:34 na02 Jul 29 12:03:34 [na02:monitor.chassisPowerSupplies.ok:info]: Chassis power supplies OK Jul 29 12:03:30 na02 Jul 29 12:03:30 [na02:monitor.chassisPowerSupply.ok:info]: Chassis power supply 1 is OK Jul 29 12:03:30 na02 Jul 29 12:03:30 [na02:callhome.chassis.ps.ok:info]: Call home for CHASSIS POWER SUPPLY OK: PS 1 Jul 29 12:03:24 na02 Jul 29 12:03:24 [na02:callhome.chassis.power:error]: Call home for CHASSIS POWER DEGRADED: Jul 29 12:03:24 na02 Jul 29 12:03:24 [na02:monitor.chassisPower.degraded:notice]: Chassis power is degraded: Jul 29 12:03:23 na02 Jul 29 12:03:23 [na02:cmds.sysconf.validDebug:debug]: sysconfig: Validating configuration. Jul 29 12:03:19 na02 Jul 29 12:03:19 [na02:callhome.chassis.ps.degraded:error]: Call home for CHASSIS POWER SUPPLY DEGRADED: PS 1 Jul 29 12:03:19 na02 Jul 29 12:03:19 [na02:monitor.chassisPowerSupply.degraded:notice]: Chassis power supply 1 is degraded: PSU1 Fan2 Speed is Unreadable thanks

2 1

EXN3000/DS4243 PSU replacement clearance issues
by Ray Van Dolson 29 Jul '13

29 Jul '13

We have EXN3000 (IBM rebranded DS4243's) disk shelves and recently had a PSU go out on one. In attempting to perform the hot swap, we discovered we didn't have enough clearance to pull the PSU out without hitting our vertically mounted CDU's (PDU -- our datacenter team prefers to call them cabinet distribution units). We had to actually lift the CDU off its mount and hold it away from the cabinet to gain the necessary clearance to pull the PSU free. While everything was still running. Definitely precarious! Turns out our racks are 42" deep and I *believe* the recommendation from NetApp is to have 48" racks -- probably exactly for this reason. Can anyone confirm? Other ways to potentially deal with this? I suppose using horizontal CDU's could work as well. Thanks, Ray

5 6

Ataching windows server to netapp
by Iluhes 23 Jul '13

23 Jul '13

Can you help me with the order of the installation of new windows server software: 1. Install Host Unilties? (Does that enable MPIO) 2. Install netapp DSM. Is there a short document on what steps to take Thanks

2 2

How non-disruptive is ifgrp favor <standby interface>
by Fletcher Cocquyt 20 Jul '13

20 Jul '13

Happy Friday We're looking to make a change to the production cluster running 300 VM's including Oracle RAC ahead of the weekend full backups (which are bottlenecking on an edge HP switch - 1000 discards/sec at the peak on this 10Gb port) So we want to run ifgrp favor <standby interface> to balance the traffic out and avoid the discards Do NFS clients (all our VMware datastores are presented via NFS vFilers) notice any issues when the standby interface (e1b) is favored over the active one (e1a)? thanks

2 2

FAS2040 to FAS2220 head swap
by James Andrewartha 17 Jul '13

17 Jul '13

Hi all, I have an (empty) single-controller FAS2040 with a DS4243 full of 2TB disks, which is purely used as a SnapVault destination. I need a bit more space in it, and somehow a new FAS2220 with 12 2TB disks is cheaper than 12 2TB disks for the 2040. So this means I need to do a head swap, which I haven't done before. I've had a read of https://library.netapp.com/ecm/ecm_download_file/ECMP1140588 and it seems fairly straightforward, but I thought I'd ask if there's any tips or gotchas I should know of. Also, what are the platform aggregate limits for the 2220? I know the 2040 had a limit of 22 disks (18 data), but can I add the new internal disks to the existing aggregate (which is 2x9+2 at the moment)? Obviously I'll need to do a reallocate after doing this. Thanks, -- James Andrewartha Network & Projects Engineer Christ Church Grammar School Claremont, Western Australia Ph. (08) 9442 1757 Mob. 0424 160 877

1 0

AD/LDAP Auth
by Mike B. 16 Jul '13

16 Jul '13

Disclaimer - I'm not Windows savvy so feel free to set me straight if/when/where needed.. I've searched around in the communities site and have read loads of posts about auth'ing admin users against AD. I've also searched the web for hours looking for similar content. I don't feel that all of what I've read fully addressed what I seem to be looking for though. Before I ramble on and on (my apologies in advance), here are some key things that need to be factored in: 1. There is no cDOT - everything is 7G/7-Mode 2. The use of vFilers is **not** an option 3. A 2-way trust **cannot** be established Reading through documentation all the way back to 7.2.5.1 (I didn't go any further), the docs say the following requirements exist to use AD: ======================== A valid CIFS license Your storage system joined to an Active Directory domain A two-way trust relationship established between your storage system’s domain and your LDAP server’s domain, if they are different ======================== >From what I've read, all one needs to do is run through the initial CIFS setup and that should suffice - but the above seems to contradict that. So do you actually /have to have/ a CIFS license AND be "..joined to an Active Directory domain" ?? IIRC you can auth against AD via the LDAP options without making the storage controllers actually be part of any AD. What if they *are* part of an existing AD; can you point them at a different set of AD servers all together for **just** admin auth? That whole trust req above seems to point to no.. Thanks!

2 1

thinn volume how to?
by Iluhes 15 Jul '13

15 Jul '13

Folks, Please tell me if the way to create a volume that will contain thin luns is the best below vol create vol1 -s none aggr0 550g vol options vol1 fractional_reserve 0 snap reserve -V vol1 0 vol autosize vol1 off snap autodelete vol1 on Do I need to create the LUNs with any special consideration Thanks!!!~

1 0

FAS 3210 after upgrade from 8.0.2 to 8.1.3 - higher cache age.
by Rafał Radecki 15 Jul '13

15 Jul '13

Hi All. I've performed an upgrade and afterwards my cache age is higher. Before the upgrade it was ~3 minutes, now it is ~4.5 minutes. The amount of operations over NFS did not change. NFS latency also for all volumes is now 50% of the latency before the upgrade. Cache age with snmp from http://ipmsupport.solarwinds.com/mibs/NETWORK-APPLIANCE-MIB/item.aspx?id=mi… . It is "Age in minutes of the oldest read-only blocks in the buffer cache. This indicates how fast read operations are cycling through system memory; when the appliance is reading very large files (larger than the machine's memory size), buffer cache age will be very low." In the background there is still a rlw_upgrading process running, details in https://communities.netapp.com/thread/22676 . So overall I think that lower nfs latency is better, but am not certain about cache age. Is it not strange both changed so much after the upgrade? Have you had similar experiences? Best regards, Rafal Radecki.

1 0

Netapp ONTAP 8.0.2 - NFS takeover/giveback - should it be visible to clients?
by Rafał Radecki 12 Jul '13

12 Jul '13

Hi All. I have two FAS3210 in 7-mode Actvie/Active. I perform takeover/gieback and see that the outage is visible for clients for nfs mounts. From your experiences should it be so? I think that information about nfs sessions and IP address are moved to the takeover node so in theory it should not be visible. Best regards, Rafal.

10 15

SAS cabling question
by Klise, Steve 09 Jul '13

09 Jul '13

Is there such a thing as SAS patch panels? I may have to go beyond the standard cable lengths to another area of the datacenter.. Has anyone had to get a workaround for this? I think the latest SAS length is 15M or 20M. Thanks Steve

4 3

Snapvault snapshots lost after upgrade.
by Jeff Cleverley 09 Jul '13

09 Jul '13

Greetings, I upgraded heads on our filers. The new 6290s had their own disks and I have 8.1.2P4 on them. The old 6080s were 7.3.5.1P4. For the update I kept the 6290 root/vol0 drives and changed ownership of all the other drives. The aggregates were brought online and all the volumes came up OK. I'm now seeing none of my backups are running because the snapvault snapshot is missing. I still have the acs snapshot. I've tried the snapvault start -S command and it says it is already a snapvault relationship. When I try to tell it to do a snapvault update, it says the base snapshot no longer exists on the source. What happened to all my snapvault snapshots? Why did the snapvault snapshots get released? I thought those were in the volume and should have carried over in the hardware migration. I still have access to the old root volumes and backups of the old root volumes if there is anything that can be salvaged. I would rather not have to re-baseline 100+TB of source data. Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

3 6

Hooking a DS4243 to a "standard" SAS adapter
by Laurent CARON 30 Jun '13

30 Jun '13

Hi, I just decomissionned 2 DS4243. Since I'd rather use them now than let them take the dust, I wonder if it is possible to hook them to a standard SAS adapter (eg: LSI SAS 9207-8e) placed in a server to act as a JBOD enclosure. Physical cabling is ok and will cause no trouble, but will the disks be seen by the card. What about multipath ? Did anybody already do/try it ? Thanks Laurent

2 3

large amount of CIFS other ops from xp desktops after canceled search
by Jordan Slingerland 28 Jun '13

28 Jun '13

Hello toasters, I am having an issue where I periodically find that some xp desktop users are doing a large amount of other ops to a cifs share with a large amount of files for a sustained period. If I let them go, they will continue for days. In most cases, it seems the users do a search and the search takes too long, so they close the window, but it is running in some kind of stuck state in the background. Killing the session or closing open files via connection to the toaster via computer management does nothing; the session just starts back up and pounds the crap out of the filler. Rebooting the desktop has been my solution so far, though I suspect restarting explorer.exe may be sufficient. I am running 8.1.1P2 and this is an IBM rebranded FAS3070. Any ideas? Has anyone seen similar behavior? cifs top -n 10 ops/s reads(n, KB/s) writes(n, KB/s) suspect/s IP Name 1859 | 0 0 | 0 0 | 0 | 10.10.100.223 domain\someuser2 1418 | 0 0 | 0 0 | 0 | 10.10.100.227 domain\someuser1 1673 | 0 0 | 0 0 | 0 | 10.10.180.230 domain\someuser3 qtree status share1 vol1 ntfs enabled normal vol status -v share1 Volume State Status Options Share1 online raid_dp, flex nosnap=off, nosnapdir=off, minra=off, sis no_atime_update=off, nvfail=off, 64-bit ignore_inconsistent=off, snapmirrored=off, create_ucode=on, convert_ucode=off, maxdirsize=73400, schedsnapname=ordinal, fs_size_fixed=on, guarantee=none, svo_enable=off, svo_checksum=off, svo_allow_rman=off, svo_reject_errors=off, no_i2p=off, fractional_reserve=0, extent=off, try_first=volume_grow, read_realloc=off, snapshot_clone_dependency=off, dlog_hole_reserve=off, nbu_archival_snap=off Volume UUID: 0fbdb127-a774-11e2-ab63-123478563412 Containing aggregate: 'sataaggr0' Plex /sataaggr0/plex0: online, normal, active RAID group /sataaggr0/plex0/rg0: normal, block checksums RAID group /sataaggr0/plex0/rg1: normal, block checksums RAID group /sataaggr0/plex0/rg2: normal, block checksums RAID group /sataaggr0/plex0/rg3: normal, block checksums Snapshot autodelete settings for share1: state=off commitment=try trigger=volume target_free_space=20% delete_order=oldest_first defer_delete=user_created prefix=(not specified) destroy_list=none Volume autosize settings: state=off Hybrid Cache: Eligibility=read-write For the fourth consecutive year, Independent Health is the highest ranked health insurance plan in the New York/New Jersey region by J.D. Power and Associates 2013 Member Health Plan StudySM. For the sixth consecutive year, Independent Health was named one of the best companies to work for in New York State. 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.

2 5

Hardware migration process
by Jeff Cleverley 27 Jun '13

27 Jun '13

Greetings, I think I have a good handle on what I need to do but wanted to check for additional gotchas. We've got some 6290s that will replace some 6080. The 6290s are already built with their own root volume/disks and are running 8.1.2P4. The 6080s are running 7.3.5.1P4. I'm not planning on using the 6080 root disks so I'm not planning on upgrading the 6080s before hand. The plan is basically shut everything down, pull the 7.3.5 root disks, reassign all the other disks to the new system IDs and boot everything up. It is my understanding that there is no 64 bit upgrade to aggregates/volumes that gets done automatically so I'm not seeing and issue there. I have functional /etc/rc, exports, hosts, and username.map files in place on the new hardware. I'm also going to have a copy of the old /etc directories on the new filers so I'll have everything there if I need to go back to them. All needed hardware is in the new heads and supported. The new hardware has newer disk and shelf firmware but I'm going to let it do that during boot up. We have some MOOS drives that have been problematic so I don't want to do those before hand. We have no VM, very little cifs, and 1 SAN/FCP client. Other than that it is all nfs clients. I know I may or may not have to re-join the domain for the cifs piece but I can deal with that. It seems this should be relatively simple and easy, but that's usually when I get into trouble :-) Any obvious holes/flaws in my plans that I should be aware of? Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

3 11

FWD: Any reason not to buy a FAS3220 over a FAS2240-2?
by Matt Kilham 27 Jun '13

27 Jun '13

Thanks to all those who replied - very helpful. We're going to go for te 3220 with FlashPool. Cheers, Matt Original Message: ---------- Forwarded Message ---------- FROM: Matt Kilham (stratton) TO: Mailing Lists DATE: Thu, 27 Jun 2013 12:32:42 +1000 RE: Any reason not to buy a FAS3220 over a FAS2240-2? Hi everyone, We are getting towards the pointy end of the procurement phase of an infrastructure refresh at our SME. Part of this is replacing an aging (badly) FAS2050 + a couple of DS14Mk4 shelves with a new NetApp filer. The FAS2240-2 was an obvious candidate, as even an entry-level system such as this provides more than enough performance and capacity for our current and planned needs for the next few years. However, a key part of our upgrade plan was implementing FlashPool, and we only discovered late in the piece that the FAS2xxx series has a 400GB usable FlashPool limit per HA pair. This is/was cause for concern around whether this would be sufficient for us given a pending VDI deployment, and led to a flurry of research into alternatives, quotes on alternative systems, etc. Where we have "landed" is that we have been quoted on two options - the original FAS2240-2 and a FAS3220*, with otherwise identical disks/software licenses/etc. The interesting part is that the FAS3220 quote is only fractionally more expensive than the 2240-2 (think 5% or so). I am being told that the reason for the close-to-price-parity is a sharp deal on the FAS3220, whereas the FAS2240-2 is close to being a fixed price / fixed margin unit. While I have not yet verified the truth of the above (that the reason for price parity is a cheap 3220 quote, not an expensive 2240-2 quote), if we take it at face value that this is true and hence can buy a FAS3220 for essentially the same price as a FAS2240-2, can anyone think of any good reasons why we wouldn't take the FAS3220 (remembering it is overkill for us in every way except the FlashPool limit)? All I can think of is that ongoing support / warranty will be a bit more expensive, and any additional software licenses purchased later will cost more (can anyone quantify approx how much?). I think the former is worth the cost, and the latter isn't too much of a concern as we barely use any separately licensed features now and I don't envision this changing in the future (except perhaps for SnapMirror - we may implement that). Thoughts from the gurus? And thanks, in advance. * we actually have two FAS3220 quotes, which are essentially the same price and spec except that one provides 7 x 200GB SSDs for 400GB + 200GB of FlashPool, and the other provides 2 x 512GB FlashCache cards. Secondary question here: for a small environment like ours that will only have one aggregate per controller, I'm thinking the FlashPool options makes more sense, but would like sanity check here - thoughts? Cheers, Matt

1 0

Any reason not to buy a FAS3220 over a FAS2240-2?
by Matt Kilham 27 Jun '13

27 Jun '13

Hi everyone, We are getting towards the pointy end of the procurement phase of an infrastructure refresh at our SME. Part of this is replacing an aging (badly) FAS2050 + a couple of DS14Mk4 shelves with a new NetApp filer. The FAS2240-2 was an obvious candidate, as even an entry-level system such as this provides more than enough performance and capacity for our current and planned needs for the next few years. However, a key part of our upgrade plan was implementing FlashPool, and we only discovered late in the piece that the FAS2xxx series has a 400GB usable FlashPool limit per HA pair. This is/was cause for concern around whether this would be sufficient for us given a pending VDI deployment, and led to a flurry of research into alternatives, quotes on alternative systems, etc. Where we have "landed" is that we have been quoted on two options - the original FAS2240-2 and a FAS3220*, with otherwise identical disks/software licenses/etc. The interesting part is that the FAS3220 quote is only fractionally more expensive than the 2240-2 (think 5% or so). I am being told that the reason for the close-to-price-parity is a sharp deal on the FAS3220, whereas the FAS2240-2 is close to being a fixed price / fixed margin unit. While I have not yet verified the truth of the above (that the reason for price parity is a cheap 3220 quote, not an expensive 2240-2 quote), if we take it at face value that this is true and hence can buy a FAS3220 for essentially the same price as a FAS2240-2, can anyone think of any good reasons why we wouldn't take the FAS3220 (remembering it is overkill for us in every way except the FlashPool limit)? All I can think of is that ongoing support / warranty will be a bit more expensive, and any additional software licenses purchased later will cost more (can anyone quantify approx how much?). I think the former is worth the cost, and the latter isn't too much of a concern as we barely use any separately licensed features now and I don't envision this changing in the future (except perhaps for SnapMirror - we may implement that). Thoughts from the gurus? And thanks, in advance. * we actually have two FAS3220 quotes, which are essentially the same price and spec except that one provides 7 x 200GB SSDs for 400GB + 200GB of FlashPool, and the other provides 2 x 512GB FlashCache cards. Secondary question here: for a small environment like ours that will only have one aggregate per controller, I'm thinking the FlashPool options makes more sense, but would like sanity check here - thoughts? Cheers, Matt

1 0

Jump to page:

2026

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