toasters June 2013

toasters@lists.teaparty.net

21 participants
13 discussions

[OT] General Purpose Storage Discussion
by Ray Van Dolson 29 Nov '13

29 Nov '13

This list is great -- but obviously is NetApp centric. I'm wondering if any of you can recommend a mailing list or forum, with a similar level of participation and expertise, that's a bit more "general purpose" but still storage-centric. Something similar to the SAGE/LISA mailing lists for Systems Administration. Perhaps SNIA offers such discussion groups to its members? Thanks, Ray

7 15

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

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

6290 10G ports vs PCI 10G cards
by Jeff Cleverley 19 Jun '13

19 Jun '13

Greetings, I haven't been able to find out anything about the bus speeds for the ports on a 6290. The hardware universe shows an order to fill slots for things like 10G or SAS controllers. This makes me believe all the ports are not the same. I was thinking of using the on-board 10G ports instead of a dual port PCI card that would be in one of the top 5 priority slots. We do have the IOX module on the arrays. Does anyone know where I can find this information? Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

1 0

roles an capabilities in ONTAP: the missing link in security-*
by Michael Bergman 10 Jun '13

10 Jun '13

Greeting, does anyone know what the secret (undocumented) security-* capability which allows you to elevate to diag level and see those CLI commands, i.e. running 'priv set diag', is called? The man pages for 8.1.x tells us this: - - - The security-* type currently only has a few elements (5): security-passwd-change-others which is used specifically to control if a user can change another user's password without knowing their previous password. By default, only root and members of the Administrators group have this capability. security-priv-advanced which is necessary to run advanced commands that are not used for normal administration. Please talk to a NetApp Inc representative before using advanced commands. By default, only root and members of the Administrators group have this capability. security-api-vfiler Normally a client will send ONTAP APIs directly to a vfiler if it wishes the API to be executed on the vfiler. The security-apivfiler capability is necessary to send ONTAP APIs to the physical node which are to be forwarded to a vfiler for execution. By default, only root and members of the Administrators group have this capability. security-load-lclgroups which is necessary to run the useradmin domainuser load command. This command changes all group membership. By default, only root and members of the Administrators group have this capability. security-complete-user-control which is used to allow an admin to add, modify, and delete users, groups and roles with more capabilities than himself. These users typically only have access to the cli-useradmin* and associated commands, though they can give themselves greater permissions. By default, only root and members of the Administrators group have this capability. - - - So that's fine and setting security-priv-advanced gives the results I expected. But it will not allow you to execute 'priv set diag'. However, setting up a role with capability security-* will, so there's at least one more in there with a name unknown to me. I tried a few educated guesses like security-priv-diag security-priv-diagnostics security-diag security-diagnostics security-priv-* security-priv-advdiag security-priv-advanced-diag plus a some more along those lines, to no avail. I'm assuming there is something in there that has a specific name -- this assumption could be wrong of course /M -- Michael Bergman Sr Systems Analyst / Storage Architect michael.bergman a ericsson d com Engineering Hub Stockholm Phone +46 10 715xxxx Service Delivery, Engineering EMEA N SMS/MMS +46 70 548xxxx Ericsson Torshamnsg 33, 16480 Sthlm, Sweden

1 1

clustered ontap monitoring snapmirror lag?
by Blake Golliher 10 Jun '13

10 Jun '13

Hi! It looks as if the concept of a 'snapmirror lag' has been scrubbed from Clustered OnTap, at least on 8.1.2P3. 'snapmirror show' will show the state, and transferring, but there's no lag, or a start time that I can extrapolate. There's a progress, which reports what's been transferred, but no lag. The api documentation doesn't have any mention of snapmirror lag in snapmirror-info attributes-list. There's no support of snapmirror in snmp (at least so far). So does DFM (or whatever they call it now) monitor this? And how would it? Thanks for any and all suggestions. If I find where this data is, I'll put the script on github. -Blake

3 7

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 June 2013