toasters

toasters@lists.teaparty.net

1 participants
13523 discussions

7.3 and MacOSX
by Todd C. Merrill 17 Feb '09

17 Feb '09

After a recent upgrade to 7.3, we have noticed all/many of our MacOSX clients banging the crap out of the filer that hosts their home directories (over NFS). (The ops are 75% NFS v3 lookups and 25% access calls.) Anybody else seen anything similar? It's definitely SpotLight that is the cause, but other than the filer OS change, the Mac clients have not changed one bit. It's almost as if the Macs think they have an entirely new volume/filesystem to catalog. Until next time... The MathWorks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

1 0

7.2 -> 7.3 performance improvements: did you measure?
by Todd C. Merrill 17 Feb '09

17 Feb '09

Folks, We beat our 6000's pretty hard. We were hoping the upgrade to 7.3 would really help the throughput of these group of filers due to the threading improvements, but our results were less than spectacular. If anyone has characterized their usage before and after an upgrade to 7.3, I'd like to hear about it and share some thoughts/charts/graphs/etc. Until next time... The MathWorks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

1 0

slow NIS push with filer as NIS slave
by David N. Blank-Edelman 13 Feb '09

13 Feb '09

Hi- I was thinking of contacting support, but before I do I was curious if anyone else has encountered the following situation: We're still (sigh, but this is changing as we migrate off of it) using NIS. We thought it made sense to use the built-in functionality to make the filer (a 3020) act as one of our NIS slaves for redundancy and performance reasons (so it wouldn't have to query off box). That seems to work fine, but for some reason the yppush of the data from our master takes a much, much longer time to return from the filer than any other of our nis slaves. Has anyone else encountered this issue and dealt with it? I realize that we could go the vaudeville route ("Patient: Doctor, my arm hurts when I move it this way. Doctor: so don't move it that way." but I was wondering if there are other solutions (options, etc) that we should try first? Thanks! -- dNb

1 0

Weird perms issue
by David McWilliams 13 Feb '09

13 Feb '09

One of our Windows admins tried to 'fix' what he thought was a perms problem on one of our mixed qtrees. I've spent most of the day working on it and it looks to be solved, apart from an error we get while using a particular CAD application - Ideas from Siemens running on Solaris 10. We can save files, etc. I did a truss -f on the script that starts the process and i get the following output. I appreciate any help anyone can give; 1740: read(0, " / u g s / n x i 5 / g e".., 8192) = 25 1740: read(0, 0x00023238, 8192) = 0 1740: close(0) = 0 1740: _exit(1) 1736: read(0, 0x000243EC, 5120) = 0 1736: ioctl(1, TCGETA, 0xFFBFBA9C) Err#22 EINVAL 1736: fstat64(1, 0xFFBFBB10) = 0 1736: fstat64(1, 0xFFBFB9B8) = 0 1736: llseek(0, 0, SEEK_CUR) Err#29 ESPIPE 1736: close(0) = 0 1736: write(1, " 0\n", 9) = 9 1728: read(10, " 0\n", 1024) = 9 1728: sigaction(SIGCLD, 0xFFBFAED8, 0xFFBFAF78) = 0 1728: waitid(P_ALL, 0, 0xFFBFAED8, WEXITED|WTRAPPED|WSTOPPED|WNOHANG) = 0 1728: sigaction(SIGCLD, 0xFFBFAED8, 0xFFBFAF78) = 0 1736: _exit(0) 1728: Received signal #18, SIGCLD, in read() [caught] 1728: siginfo: SIGCLD CLD_EXITED pid=1736 status=0x0000 1728: read(10, 0xFFBFB788, 1024) Err#4 EINTR 1728: lwp_sigmask(SIG_SETMASK, 0x00020000, 0x00000000) = 0xFFBFFEFF [0x0000FFFF] 1728: setcontext(0xFFBFAA88) 1728: read(10, 0xFFBFB788, 1024) = 0 1728: ioctl(10, TCGETA, 0xFFBFB04C) Err#22 EINVAL 1728: sigaction(SIGCLD, 0xFFBFB4B0, 0xFFBFB550) = 0 1728: waitid(P_ALL, 0, 0xFFBFB4B0, WEXITED|WTRAPPED|WSTOPPED) = 0 1728: sigaction(SIGCLD, 0xFFBFB4B0, 0xFFBFB550) = 0 1728: close(10) = 0 1728: fork1() = 1742 1742: fork1() (returning as child ...) = 1728 1742: getpid() = 1742 [1728] 1742: lwp_self() = 1 1742: lwp_sigmask(SIG_SETMASK, 0x00000000, 0x00000000) = 0xFFBFFEFF [0x0000FFFF] 1742: getpid() = 1742 [1728] 1742: sigaction(SIGXFSZ, 0xFFBFBF10, 0xFFBFBFB0) = 0 1742: sigaction(SIGQUIT, 0xFFBFBF10, 0xFFBFBFB0) = 0 1728: lwp_sigmask(SIG_SETMASK, 0x00000000, 0x00000000) = 0xFFBFFEFF [0x0000FFFF] 1728: sigaction(SIGCLD, 0xFFBFBF28, 0xFFBFBFC8) = 0 1742: execve("/ugs/nxi5/geo/geomod.exe", 0x0005CE78, 0x0005CE80) 1742: *** cannot trace across exec() of /ugs/nxi5/geo/geomod.exe *** 1666: waitid(P_ALL, 0, 0xFFBFE378, WEXITED|WTRAPPED|WSTOPPED) (sleeping...) 1726: waitid(P_PID, 1728, 0xFFBFC6A8, WEXITED|WTRAPPED|WNOWAIT) (sleeping...) 1728: waitid(P_ALL, 0, 0xFFBFBF28, WEXITED|WTRAPPED|WSTOPPED) (sleeping...) *** I-DEAS VMI Diagnostic (fiinq_c) *** ---------------------------------------------------------------- *** Operating System Function - *** stat() on directory name failed : Permission denied (errno=13) *** ---------------------------------------------------------------- The executable that it refers to /ugs/nxi5/geo/geomod.exe is on a different qtree that was messed with, the perms on the file are 4755, which are correct. Slainte, *David* "Build a man a fire, and he'll be warm for a day. Set a man on fire, and he'll be warm for the rest of his life." ~ *Terry Pratchett <http://en.wikipedia.org/wiki/Terry_pratchett>, author, Discworld <http://en.wikipedia.org/wiki/Discworld> series*

1 0

Re: NFS and VMFS on a single ESX host
by John Sellens 13 Feb '09

13 Feb '09

| Looking to run NFS and VMFS ESX data stores on a single ESX host. | | Anybody else doing it? | | Gotchas? We do it (total of 15 or so ESX hosts, 4 filers), though most of our ESX datastores are VMFS on LUNs from the filers. Works just as expected. I haven't looked closely at relative performance, but we have a number of low use VMs where disk performance is almost irrelevant, and tend to put them on NFS datastores. I like using NFS datastores: - snapshots let us mount old .vmdk files for "easy" file restores - doing it with a lun seems more of a pain - easy to adjust the size of a datastore on the filer - easier to create a machine - we lean towards LUN per VM, which means creating a new VM is more annoying One inconvenience of NFS datastores is that they need to be added separately on each ESX host. I don't know if adding a datastore can be done from the service console shell, but we use the VI client, which is tedious. With LUNs, we configure them once on the filer, and then our ESX rescan scripts result in the storage being visible everywhere. Hope that helps - cheers! John

3 2

RE: SSD. What's about?
by Kennedy, Jeffrey 13 Feb '09

13 Feb '09

Right, but the question was what's it all about. The SSD stuff is fairly robust, at least in the last couple of years. Burn-out is a concern but that's why you raid it; run mirror'd SSD Lun's. It's expensive for sure but if performance is everything you can get it in spades. For several years now you could get this kind of performance elsewhere; Sun comes to mind immediately. So if availability was less of a concern and you were a speed-freak, Sun owned Netapp by very large margins. And with other vendors pushing the availability envelop and creeping into Netapp's sweetspot, along with all the other features they've been using to entice buyers, Netapp had to do something. I think it's a great move. The capability is probably needed by only a small portion of the population but that portion has a lot of money to burn on performance. Jeff Kennedy Qualcomm, Incorporated QCT Engineering Compute 858-651-6592 From: Jeff Mohler [mailto:speedtoys.racing@gmail.com] Sent: Thursday, February 12, 2009 3:10 PM To: Kennedy, Jeffrey Cc: Milazzo Giacomo; toasters(a)mathworks.com Subject: Re: SSD. What's about? ..until you burn out the SSD. Although, thats getting better, but at much greater cost as well. Spinning will still always be cheaper, its a simple $ per IO question On Thu, Feb 12, 2009 at 12:35 PM, Kennedy, Jeffrey <jkennedy(a)qualcomm.com<mailto:jkennedy@qualcomm.com>> wrote: What's it all about? Performance. Blazing fast performance. Think about it like this. An SSD volume can provide ~1M IOPS and come in sizes of up to 1TB, maybe larger since I last looked. Add in Flexshare and you can basically stop time. Jeff Kennedy Qualcomm, Incorporated QCT Engineering Compute 858-651-6592 From: owner-toasters(a)mathworks.com<mailto:owner-toasters@mathworks.com> [mailto:owner-toasters@mathworks.com<mailto:owner-toasters@mathworks.com>] On Behalf Of Milazzo Giacomo Sent: Thursday, February 12, 2009 1:08 AM To: toasters(a)mathworks.com<mailto:toasters@mathworks.com> Subject: SSD. What's about? Hi all, there are a lot of rumors about Solid State Drives...or SS Disks...alfo if there are no 'disks' at all inside :) I've read some docs from NetApp and other SSD vendors. NetApp says they support them in vFiler (so that, as native disks also? considering that vFiler nowadays can work in mixed mode) but there aren't tech specs, availability, configurator does not provide them... So, what's about all this? Regards,

2 1

SSD. What's about?
by Milazzo Giacomo 13 Feb '09

13 Feb '09

Hi all, there are a lot of rumors about Solid State Drives...or SS Disks...alfo if there are no 'disks' at all inside :) I've read some docs from NetApp and other SSD vendors. NetApp says they support them in vFiler (so that, as native disks also? considering that vFiler nowadays can work in mixed mode) but there aren't tech specs, availability, configurator does not provide them... So, what's about all this? Regards,

4 4

NFS and VMFS on a single ESX host
by Gerald Villabroza 10 Feb '09

10 Feb '09

Looking to run NFS and VMFS ESX data stores on a single ESX host. Anybody else doing it? Gotchas? -=--=- gerald villabroza <geraldv at stanford.edu> technical lead, its storage, stanford university

1 0

Re: RE: Avamar VS SMVI
by hollandwl＠gmail.com 10 Feb '09

10 Feb '09

Snapvault is not EOL. The next version of SMVI will be including snapvault support. In the interim, Matt Robinson of NetApp has written SV SMVI (Snapvault for Snapmanager for Virtual Infrastructure). I've tried it in our development lab and it seems to work really well. Take a look at it at http://communities.netapp.com/docs/DOC-1868;jsessionid=A4775C7EF68BD0E628BE… The S550 is being discontinued and therefore snapvault primary for it is EOL. However, snapvault for the FAS series is not EOL. On Feb 9, 2009 11:40am, "Klise, Steve" <klises(a)pamf.org> wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > I think snapvault is EOL but is supported > > through 7.3. For some reason, the backup guys don’t want to do it, not > > sure about the specifics, but our VAR (who is great btw), has been trying to > > push us in that direction. > > > > > > > > > > IF it is EOL, that is a nail in that > > coffin. > > > > > > > > > > > > > > > > > > > > > > > > > > > From: > > owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Bill Holland > > > Sent: Sunday, February 08, 2009 > > 4:00 AM > > > To: Klise, > > Steve; toasters(a)mathworks.com > > > Subject: Re: Avamar VS SMVI > > > > > > > > > > > > > > > > > > > Have you looked at Snapvault from NetApp? > > > > > > > > > > > > > > > > > > ----- Original Message ----- > > > > > > > > > > > > > From: steve > > klise > > > > > > > > > > > > > > To: toasters(a)mathworks.com > > > > > > > > > > > > > > > > Sent: Saturday, February > > 07, 2009 2:16 PM > > > > > > > > > > > > > > Subject: Avamar VS SMVI > > > > > > > > > > > > > > > > > > > > > > > > We are starting an eval of SMVI and Avamar for backup/recovery of VM's. > > We are planning on having snapmirrors to another filer for backup (NDMP) and or > > DR purposes, but are doing the sm to a local filer for now. Avamar has some > > good selling point, but it seems to be a waste to buy yet another storage > > platform. We are heavily invested in netapp now with a VTL (1400), and > > NetBackup. Does anyone have any feed back? It seems to make sense to to SMVI > > but with Avamar I get longer retention. I was only going to keep a weeks worth > > of SMVI snapshots. With Avamar, it seem indefinite. The downside of the avamar > > is you seem to need two for going to tape. Avamar can do the single file > > restores, which SMVI 2.0 is supposed to do. Anyhow, any feedback is > > appreciated. > > > > > > > > > > > > > > > > > > View this message in context: Avamar VS > > SMVI > > > Sent from the Network Appliance > > - Toasters mailing list archive at Nabble.com. > > > > > > > > > > > > > > > > > > > >

2 1

How to balance volume priority (some NFS vs. CIFS)
by Adam McDougall 09 Feb '09

09 Feb '09

I have not opened a case with Netapp yet but probably will if no one has any good ideas; I just like to pick people's brains before going official. Thanks for any input. A few months ago we moved a file share off a Windows server onto our FAS3040 Netapp running 7.2.4 and shared it out via CIFS. It contains software install files and scripts and depending on scheduled jobs, it can get hit pretty hard and pushes out approximately 1 Gbit/sec, which has been drastically affecting the service times for our other shares on that filer, and its namely response-sensitive NFS shares we care about that are affected the most such as mail and web files. It doesn't really seem to be a disk bottleneck because the disk read/sec in sysstat is usually only half of what the filer is pushing out to the network, so I assume its reading some data from cache. The CIFS software install share can either get hit by 1-60+ CIFS clients where each client reads files on and off for hours at a time, or sometimes we have hundreds of clients hitting the share at once for a smaller set of files (such as to update one software package across a large set of PCs). I've been able to reproduce the slowdown with just 4 CIFS clients on gigabit downloading a large file from the share. Sometimes it only causes a modest slowdown in the NFS response time but sometimes email messages being moved between folders will stall for 8 seconds or much more, which is pretty much unacceptable. I don't think its a bottleneck in my core network because I've done tests where the slow nfs client is on the same switch as the filer, which is connected via two gig links using LACP. Also, in the normal situation where the slowdown is encountered, mail (NFS) traffic is flowing through a different gig uplink than the hungry CIFS clients. Goal: reduce the impact of greedy clients (primarily known ones, but hopefully unexpected ones too) on the response time of the rest of the filer's clients. I don't care if the CIFS software share must accept slower data rates, and I'd rather not run away from the problem by avoiding it but rather learn what I can do to prevent my filer from being held hostage by greedy clients. I do have another 3040 I could move the share to, but that filer also has volumes that would be affected negatively in the same way, and I'd rather not concede defeat and go back to hosting the share on a dedicated windows server. I can try different code versions in a test environment if I need to, but I'd like to think this kind of situation would have come up already and have a solution at hand. I've played around with na_priority trying to set the mail and website volumes to high or veryhigh priority and the software share to low or verylow but that isn't making a measurable impact. I'm not really sure what to tweak or check next. Here is an example from sysstat when I am simulating the slowdown condition with 4 CIFS clients on gigabit fetching the same file. CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache in out read write read write age 6% 2058 167 0 751 1543 2196 0 0 0 11 6% 2590 164 0 699 2238 2904 32 0 0 11 10% 2183 223 0 1241 4471 5072 17872 0 0 11 11% 3299 799 0 1577 22194 4935 1183 0 0 11 22% 3298 3072 0 3005 107869 9128 24 0 0 11 18% 2532 1986 0 2270 87651 2078 0 0 0 11 18% 2198 2200 0 1696 105941 8032 8 0 0 11 16% 3597 1650 0 1890 84691 3528 24 0 0 11 23% 4946 2216 0 2604 112741 14664 0 0 0 11 22% 4075 2041 0 2324 100380 21568 0 0 0 11 CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache in out read write read write age 21% 3272 2246 0 2862 115380 4688 24 0 0 11 21% 4117 2092 0 2686 109165 3864 8 0 0 11 26% 4188 2136 0 3436 115081 21900 0 0 0 11 ......(skip) 30% 7487 1773 0 4261 93385 10156 3328 0 0 6 25% 4566 1900 0 3339 96655 13764 9808 0 0 7 24% 2965 2202 0 2477 111493 11772 5475 0 0 8 23% 5256 1986 0 3093 102409 10508 24 0 0 8 19% 2979 2068 0 1810 102282 9926 0 0 0 8 20% 3164 2323 0 2301 111209 1560 8 0 0 8 23% 7082 2165 0 2322 103816 2292 24 0 0 8 22% 11780 1158 0 2763 55501 1760 0 0 0 8 20% 12032 675 0 3820 36504 2452 0 0 0 8 CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache in out read write read write age 23% 16269 1122 0 3914 54034 4460 24 0 0 6 18% 8991 1030 0 2739 48400 4568 8 0 0 6 10% 3903 237 0 1346 4494 3828 0 0 0 6 11% 3912 219 0 1623 4301 3808 6508 0 0 6 8% 2402 224 0 868 2027 2744 8712 0 0 6

5 5

Jump to page:

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