toasters September 2013

toasters@lists.teaparty.net

28 participants
18 discussions

Expected performance difference between two configurations
by Ray Van Dolson 17 Sep '13

17 Sep '13

Hi all; Am trying to understand what sort of performance difference I might see between two different configurations: 1) IBM N6240 E21 (FAS3240C) w/ 120x600GB 15K 3.5" SAS and 512GB of flash cache 2) IBM N6250 E26 w/ 80x900GB 10K 2.5" SAS and 512GB of flash cache. Sorry, on the latter I don't know the equivalent FAS. Probably FAS3250C? We have fewer spindles, but newer, beefier controllers. Our workload is primarily VMware via NFS. Lotsa random reads and writes (more on the read side) with I'd say the bulk of the IO requests in the 64KB+ range. Will I regret going with fewer spindles? Ray

8 8

The future of OnCommand Core for 7-mode
by Scott Eno 17 Sep '13

17 Sep '13

Hi, With the release of OnCommand Core 6.0 For Clustered OnTap (phew!), has anyone heard about the future of OnCommand Core (henceforth "DFM") for 7-Mode? Are we seeing the end of feature releases for DFM, and 5.2 is now in perpetual maintenance mode like OnTap 7.3.7? I have no plans or reason to transition to cluster mode and certainly won't do it just to get the latest version of DFM, but I do use the crap out of DFM (GUI, CLI, Protection and Provisioning managers, SnapMirror/Vault management, RBAC, controller user account management, controller config management, etc) and, like any other application, it's always got bugs and issues. I hate to think that DFM is at the end of the road unless I go to cluster mode. Kinda like Microsoft releasing a new version of Office but it only works on a Surface tablet. --- Scott Eno

1 0

CF Giveback options
by Eric Peng 16 Sep '13

16 Sep '13

Am running ONTAP 8.1.2 P4.. Currently have the following settings enabled: cf.takeover.on_network_interface_failure on cf.takeover.on_network_interface_failure.policy any_nic Want to enable setting: cf.giveback.auto.enable on Was informed by IBM there is a bug where enabling auto CF giveback can conflict with cf.takeover.on_network_interface_failure, and result in the system getting into an infinite loop of takeovers and givebacks. Do most of you leave auto CF giveback disabled as general policy? Eric

4 3

high CP_reads/writes ratio
by Fred Grieco 13 Sep '13

13 Sep '13

What kinds of things can cause a high CP_reads/writes ratio? And what can I do about it? I have an Oracle system with ASM that doing writes in a 15 minute cycle, and it causes a lot of writes with CP reads. The oracle ASM luns are basically the only thing on the aggregate. Disk alignment is fine (which is the first thing I thought of). I've attached a screenshot of the Oncommand graph for one of the disks in this aggr-- not sure if that will come through. The filer is a 3220 with 8.1.2 7-mode. Fibre channel attach, and there are about 20 luns under two volumes. TIA, Fred

6 12

RE: high CP_reads/writes ratioA
by Jordan Slingerland 13 Sep '13

13 Sep '13

At 87% it is certainly possible that wafl doesn't have enough contagious free space to write a chain in a single pass...slowing things down substantially. Are your volumes thick provisioned? That could cloud the equation. --JMS From: Fred Grieco [mailto:fredgrieco@yahoo.com] Sent: Friday, September 13, 2013 3:16 PM To: Jordan Slingerland; Jeff Mohler Cc: Toasters(a)teaparty.net Subject: Re: high CP_reads/writes ratio The prod one is 87%. The replication one is 71% but it was much higher during disaster recovery testing. I'll run a reallocate measure on the DR side this weekend and see what comes back. ________________________________ From: Jordan Slingerland <Jordan.Slingerland(a)independenthealth.com<mailto:Jordan.Slingerland@independenthealth.com>> To: Jeff Mohler <speedtoys.racing(a)gmail.com<mailto:speedtoys.racing@gmail.com>>; Fred Grieco <fredgrieco(a)yahoo.com<mailto:fredgrieco@yahoo.com>> Cc: "Toasters(a)teaparty.net<mailto:Toasters@teaparty.net>" <Toasters(a)teaparty.net<mailto:Toasters@teaparty.net>> Sent: Friday, September 13, 2013 3:12 PM Subject: RE: high CP_reads/writes ratio Are your aggregates more then 65-85% full? From: toasters-bounces(a)teaparty.net<mailto:toasters-bounces@teaparty.net> [mailto:toasters-bounces@teaparty.net] On Behalf Of Jeff Mohler Sent: Friday, September 13, 2013 3:09 PM To: Fred Grieco Cc: Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> Subject: Re: high CP_reads/writes ratio Wafl free space fragmentation most likely. Reallocate all volumes in the affected aggregate. Do not reallocate the _aggregate_. Sent from my iPhone On Sep 13, 2013, at 1:06 PM, Fred Grieco <fredgrieco(a)yahoo.com<mailto:fredgrieco@yahoo.com>> wrote: What kinds of things can cause a high CP_reads/writes ratio? And what can I do about it? I have an Oracle system with ASM that doing writes in a 15 minute cycle, and it causes a lot of writes with CP reads. The oracle ASM luns are basically the only thing on the aggregate. Disk alignment is fine (which is the first thing I thought of). I've attached a screenshot of the Oncommand graph for one of the disks in this aggr-- not sure if that will come through. The filer is a 3220 with 8.1.2 7-mode. Fibre channel attach, and there are about 20 luns under two volumes. TIA, Fred _______________________________________________ Toasters mailing list Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> http://www.teaparty.net/mailman/listinfo/toasters

3 2

was there a thread after this ?
by Uddhav Regmi 12 Sep '13

12 Sep '13

Was there a thread after this ? Just making sure, I'm getting them all.. From: toasters-bounces(a)teaparty.net [mailto:toasters-bounces@teaparty.net] On Behalf Of Jeff Cleverley Sent: Tuesday, September 03, 2013 5:29 PM To: tmac Cc: &lt,Toasters(a)teaparty.net&gtt&gt, Subject: Re: Third party LR 10G networking. It sounds like someone else may be working that already. If they get one approved we may be able to piggy back on that. Thanks, Jeff On Tue, Sep 3, 2013 at 2:04 PM, tmac <tmacmd(a)gmail.com <mailto:tmacmd@gmail.com> > wrote: Ask for a PVR from NetApp. The card with removable optics should support the LRs, NetApp would just needed to support the driver. On Sep 3, 2013 3:56 PM, "Jeff Cleverley" <jeff.cleverley(a)avagotech.com <mailto:jeff.cleverley@avagotech.com> > wrote: We're looking at having multiple connections per filer going to a switch in each room to remove the switch as a single point of failure. The single mode cabling will most likely be more compatible with 40G and 100G networking in the next few years. Jeff On Tue, Sep 3, 2013 at 1:05 PM, tmac <tmacmd(a)gmail.com <mailto:tmacmd@gmail.com> > wrote: Why not use SR optics from the NetApp to the switch (supported by netapp) And then LR optics between switches? --tmac Tim McCarthy Principal Consultant <http://dl.dropbox.com/u/6874230/na_cert_dma_2c.jpg> <http://dl.dropbox.com/u/6874230/rhce.jpeg> <http://dl.dropbox.com/u/6874230/na_cert_ie-san_2c.jpg> Clustered ONTAP Clustered ONTAP NCDA ID: XK7R3GEKC1QQ2LVD RHCE6 110-107-141 <https://www.redhat.com/wapps/training/certification/verify.html?certNumber= 110-107-141&isSearch=False&verify=Verify> NCSIE ID: C14QPHE21FR4YWD4 Expires: 08 November 2014 Current until Aug 02, 2016 Expires: 08 November 2014 On Tue, Sep 3, 2013 at 2:57 PM, Jeff Cleverley <jeff.cleverley(a)avagotech.com <mailto:jeff.cleverley@avagotech.com> > wrote: Greetings, We are looking at upgrading and changing our network architecture. One option would be to use LR GBIC with single mode fibre. This appears to require non-NetApp GBIC which I'm sure makes it unsupported. Is anyone doing this, and if so, which parts are you using and what issues have you run into? Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611 <tel:970-288-4611> _______________________________________________ Toasters mailing list Toasters(a)teaparty.net <mailto:Toasters@teaparty.net> http://www.teaparty.net/mailman/listinfo/toasters -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611 <tel:970-288-4611> -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

2 2

NetApp Simulator - Using different interfaces for each node of cluster?
by Paul Letta 11 Sep '13

11 Sep '13

Hello, I've got a 3240 HA pair running 7.3.5. We are looking to enable cluster failover for network interface failures. I want to test this with 7.3.5 line of the simulator. I can create a HA cluster with the simulator. The problem is that both nodes of the sim use the same interface on the underlying OS. I would like node1 of the sim to use eth1 and node2 of the sim to use eth2. That way I can simulate a network failure of just one of the nodes of the sim. I have not found a way to achieve this yet. I tried creating 2 separate sim nodes (without clustering) on different interfaces, but after they are both up, the cluster interconnect is missing -- so enabling cf does nothing. Any ideas on how I can do this with the sim? I tried just doing an "ifconfig ns0 down" on one of the nodes of the cluster-- that disconnected the network from that node, but it didn't trigger the cluster failover, so I'm guessing that an ifdown is not the same as a nic failure (at least as far as triggering a takeover is concerned). Thanks. Paul

3 4

Third party LR 10G networking.
by Jeff Cleverley 03 Sep '13

03 Sep '13

Greetings, We are looking at upgrading and changing our network architecture. One option would be to use LR GBIC with single mode fibre. This appears to require non-NetApp GBIC which I'm sure makes it unsupported. Is anyone doing this, and if so, which parts are you using and what issues have you run into? Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

2 4

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 September 2013