toasters January 2007

toasters@lists.teaparty.net

62 participants
76 discussions

SNMP Question
by Carl Howell 25 Jan '07

25 Jan '07

What would be the correct way to calculate the total size of an aggregate with the following SNMP information: dfHighTotalKBytes = 2 dfLowTotalKBytes = -1834563852 I thought it would be 2*4294967295 + 1834563852 = 10,424,498,442 But this comes out 625,839,594 short of the df -rk output: Aggregate total = 11,050,338,036 The MIB is at version 1.17.3. Thanks, --Carl

2 1

RE: SNMP Question
by Flank, Joshua 25 Jan '07

25 Jan '07

Carl, There's a public report for a similar issue in now: http://now.netapp.com/NOW/cgi-bin/bol?Type=Detail <http://now.netapp.com/NOW/cgi-bin/bol?Type=Detail&Display=80268> &Display=80268 The equation to combine these is: if (Low >= 0) x = High * 2^32 + Low if (Low < 0) x = (High + 1) * 2^32 + Low In this instance: High = 2 Low = -1834563852 since Low < 0, use the second equation: x = (2+1) * 2^32 + (-1834563852) = 12884901888 - 1834563852 = 11050338036 Good luck, -Joshua -----Original Message----- From: Carl Howell [mailto:chowell@uwf.edu] Sent: Thursday, January 25, 2007 1:20 PM To: toasters(a)mathworks.com Subject: SNMP Question What would be the correct way to calculate the total size of an aggregate with the following SNMP information: dfHighTotalKBytes = 2 dfLowTotalKBytes = -1834563852 I thought it would be 2*4294967295 + 1834563852 = 10,424,498,442 But this comes out 625,839,594 short of the df -rk output: Aggregate total = 11,050,338,036 The MIB is at version 1.17.3. Thanks, --Carl

1 0

RE: 1 disk shelf with 1 head....using just aggr0
by Fox, Adam 25 Jan '07

25 Jan '07

Yes, that is one of the rare cases where it could be useful to have a dedicated root volume/aggr. However, running wafl_iron should be a very rare event, and in the case of a single shelf system, it's my opinion that not consuming the space outweighs this possibility. But each site needs to make their own determination on this. -- Adam Fox adamfox(a)netapp.com -----Original Message----- From: Eino Tuominen [mailto:eino@utu.fi] Sent: Wednesday, January 24, 2007 11:58 PM To: Glenn Walker Cc: Fox, Adam; Linux Admin; NetApp Toasters List Subject: Re: 1 disk shelf with 1 head....using just aggr0 Glenn Walker wrote: > > Between the two above arguments, there aren't a lot of reasons not to > put it all together. One other reason I came up with: you cannot run wafl iron on rootvol or on an aggregate containing the rootvol on a live system. I once had to run wafl iron on a 4 shelf single aggr R200, and it was a _long_ downtime. -- Eino Tuominen

1 0

RE: 1 disk shelf with 1 head....using just aggr0
by Glenn Walker 25 Jan '07

25 Jan '07

Good advice - forgot about that one... To make matters worse, the wafl check in question (when root vol is involved) must first run on the underlying aggr, THEN on each volume contained within it. That is indeed serious down time, as the point of an aggr is large number of disks (which means lots of empty space to fill with data). Glenn -----Original Message----- From: Eino Tuominen [mailto:eino@utu.fi] Sent: Wednesday, January 24, 2007 11:58 PM To: Glenn Walker Cc: Fox, Adam; Linux Admin; NetApp Toasters List Subject: Re: 1 disk shelf with 1 head....using just aggr0 Glenn Walker wrote: > > Between the two above arguments, there aren't a lot of reasons not to > put it all together. One other reason I came up with: you cannot run wafl iron on rootvol or on an aggregate containing the rootvol on a live system. I once had to run wafl iron on a 4 shelf single aggr R200, and it was a _long_ downtime. -- Eino Tuominen

1 0

RE: 1 disk shelf with 1 head....using just aggr0
by Glenn Walker 25 Jan '07

25 Jan '07

We currently have both sides of this particular scenario: systems with only one massive aggr with the root vol contained within, and systems with a separate aggr\root vol and other aggrs dedicated to data. It depends on the system: FAS250, or FAS270 (especially the single-shelf variety), it is a WASTE of a system to have a separate aggr for root. With some of the larger sized systems, it typically doesn't matter if you suddenly lose 3 disks out of 50+. The traditional reason to keep the data sets separate is recovery: losing both the root vol and data would be a terrible shame. However, the chance of losing 3 disks across a single RG is almost NIL. Even then, I'd rather lose my root than my data. The other reason is performance: having a heavy I/O load on data access could cause problems for filer performance if the root vol were to be co-located within the same physical disks. Again, the chance of having that type of issue with an extremely large aggregate isn't very high. Between the two above arguments, there aren't a lot of reasons not to put it all together. Glenn ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Fox, Adam Sent: Wednesday, January 24, 2007 3:34 PM To: Linux Admin; NetApp Toasters List Subject: RE: 1 disk shelf with 1 head....using just aggr0 I would just grow out aggr0 and save the space. There are only a few corner cases where having a dedicated root aggr makes any sense and they are rare. So while it may be nice to have, I wouldn't bother on a system of that size. It would, indeed, be a horrible waster of space. With one shelf of disks, 99% of customers should set up a single RAID-DP aggregate with a single spare. There are exceptions, but without knowing anything else about your environment or future growth plans, that's what I'd do. -- Adam Fox adamfox(a)netapp.com ________________________________ From: Linux Admin [mailto:sysadmin.linux@gmail.com] Sent: Wednesday, January 24, 2007 2:39 PM To: NetApp Toasters List Subject: 1 disk shelf with 1 head....using just aggr0 Would anyone recommend not using/growing aggr0 if I only have 1 disk shelf Currently I have 1 vol0 in agr0 (1 data disk + 2 paratiy) I would think that creating a second aggr would be just enormous wast of 300GB disks What is the best practice? Should I just grow the aggr0? Thanks

2 1

RE: 1 disk shelf with 1 head....using just aggr0
by Fox, Adam 24 Jan '07

24 Jan '07

I would just grow out aggr0 and save the space. There are only a few corner cases where having a dedicated root aggr makes any sense and they are rare. So while it may be nice to have, I wouldn't bother on a system of that size. It would, indeed, be a horrible waster of space. With one shelf of disks, 99% of customers should set up a single RAID-DP aggregate with a single spare. There are exceptions, but without knowing anything else about your environment or future growth plans, that's what I'd do. -- Adam Fox adamfox(a)netapp.com _____ From: Linux Admin [mailto:sysadmin.linux@gmail.com] Sent: Wednesday, January 24, 2007 2:39 PM To: NetApp Toasters List Subject: 1 disk shelf with 1 head....using just aggr0 Would anyone recommend not using/growing aggr0 if I only have 1 disk shelf Currently I have 1 vol0 in agr0 (1 data disk + 2 paratiy) I would think that creating a second aggr would be just enormous wast of 300GB disks What is the best practice? Should I just grow the aggr0? Thanks

1 0

1 disk shelf with 1 head....using just aggr0
by Linux Admin 24 Jan '07

24 Jan '07

Would anyone recommend not using/growing aggr0 if I only have 1 disk shelf Currently I have 1 vol0 in agr0 (1 data disk + 2 paratiy) I would think that creating a second aggr would be just enormous wast of 300GB disks What is the best practice? Should I just grow the aggr0? Thanks

1 0

NFS equivalent of "cifs top"?
by Jeff Mery 24 Jan '07

24 Jan '07

Anyone know if this exists? "nfs stat -l" kind of give me what I want, but I'd really like it in the format of cifs top if possible. When the ops for CIFS look out of whack for some reason, it's really easy to find the culprit via cifs top. Would be great to have for nfs too. Jeff Mery - MCSE, MCP National Instruments ------------------------------------------------------------------------- "Allow me to extol the virtues of the Net Fairy, and of all the fantastic dorks that make the nice packets go from here to there. Amen." TB - Penny Arcade -------------------------------------------------------------------------

3 2

RE: netapp certified expert
by Herbert.Langenbach＠shell.com 23 Jan '07

23 Jan '07

if this day to day administration includes OSSV, MetroCluster, SnapLock and some other nice NetApp stuff your chances are better than zero. But if you get your hands on the official training material your chances are excellent. Herbert NACE -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com]On Behalf Of Leeds, Daniel Sent: dinsdag 23 januari 2007 18:51 To: toasters(a)mathworks.com Subject: netapp certified expert has anyone completed the netapp certified expert exams? coming from a day to day administration level on filers for many years im interested in what sort of topics they cover and if they are geared towards the real world or more of a training course regurgitation. in so many other vendor/system certs the exams focus on trivial and unrealistic aspects of adminstration--for example, sun certs focusing on adding printers and users via the gui---puhlease. mostly im just trying to figure out if someone could reasonably take the tests on experience alone or do i need to shell out 15K plus on all the little courses to get the courseware and then cram that. thanks -- Daniel Leeds Senior Systems Administrator Edmunds.com

1 0

RE: Aggregate expansion
by Glenn Walker 23 Jan '07

23 Jan '07

There's also the 'schedule-able' command 'wafl reallocate' that does the same thing. It's very useful, though there is some debate on exactly what happens when snapshots are taken into consideration. Generally speaking, if there is adequate free space, and the data in question is deemed 'fragmented' by the scanner, then it will move it. Not that it much matters, but adding disks to an existing RG vs adding a new RG is about the same: when data is written within an RG, it is striped across the disks in such a fashion that each disk gets x number of contiguous blocks, then it writes to the next disk (64 or 128 depending on the version of ONTAP and disk type). Parity is calculated per stripe. When writing across multiple RGs, WAFL decides how much data per RG to write (tetris, but not the game) and breaks up the writes per RG in this manner. I believe this is in 64MB chunks. Either way, the data is written to the location where the best example of contiguous free space exists. Glenn ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Glenn Dekhayser Sent: Monday, January 22, 2007 3:22 PM To: toasters(a)mathworks.com Subject: RE: Aggregate expansion All; It was my understanding that you can alleviate the pain caused by an uneven expansion of disks as mentioned below by performing a wafl scan reallocate, anyone have info to contrary? Glenn ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Fox, Adam Sent: Monday, January 22, 2007 2:30 PM To: Suresh Rajagopalan; toasters(a)mathworks.com Subject: RE: Aggregate expansion That depends greatly on the values of N and n, as well as the RAID group size of the aggregate. Done according to best practices, there will be almost no difference. Done poorly, it can make all the difference in the world. My personal view (I'm in no position to speak officially for NetApp) on this to add disks to an aggregate in one of 2 multiples: 1) a whole RG at a time 2) half a RG at a time. This typically allows for a sufficient number of free disks such that you should not expect any noticable performance difference. I realize that not all sites can implement this, but let's look at a worst practice: N = n+1 Fill up the aggregate, then add 1 disk. Ouch! This hurts! So you've seen what I consider to be the best case for adding space, and you've seen the worst case. So how close you are to these extremes should give you can idea of what to expect. I know this isn't a simple answer, but I believe it to be accruate. -- Adam Fox adamfox(a)netapp.com ________________________________ From: Suresh Rajagopalan [mailto:SRajagopalan@williamoneil.com] Sent: Monday, January 22, 2007 12:46 PM To: toasters(a)mathworks.com Subject: Aggregate expansion Is there any difference between creating an aggregate on a certain number of disks (say n) , and then later expanding the aggregate to N disks, as opposed to creating the initial aggregate on N disks? Suresh

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 January 2007