toasters

toasters@lists.teaparty.net

1 participants
13532 discussions

RE: nfs performance problems
by Leonard, Roger 17 May '02

17 May '02

actually i meant nfs tcp, not version 3. sorry. -----Original Message----- From: Leonard, Roger [mailto:Roger.Leonard@marconi.com] Sent: Friday, May 17, 2002 3:43 PM To: 'Kelvin Edwards' Cc: 'toasters(a)mathworks.com' Subject: RE: nfs performance problems what seemed to help my situation was turning on nfs V3 on the filer along with putting 5.37r3 down. the cp-from-cp went down, but not away and the client timeouts stopped. my filer load is still way high but i think the client not having to resend all those 32k blocks cut down a lot on the net load. -----Original Message----- From: Kelvin Edwards [mailto:kelvin@jlab.org] Sent: Friday, May 17, 2002 2:36 PM Cc: 'toasters(a)mathworks.com' Subject: Re: nfs performance problems We've been chasing some NFS timeout issues here as well. The systat 1 command on the filer doesn't look very bad, however, I did the wafl_susp -z followed by a wafl_susp -w after a few minutes(approx 5). The value at that time was >= 1 and it continued to increase over time. The recommendation below was to wait 30 seconds before checking the value, but does it matter. Does a value other than 0 in this parameter mean the filer is denying write requests to users ? Kelvin Edwards System Admin Jefferson Lab Brian Long wrote: > >The other command you can use to gather meaninful data is wafl_susp. Run >wafl_susp -z to reset the stats. Wait 30 seconds or more and then run >wafl_susp -w (you may want to do this via rsh and save the output to a >file). > >In the output, there is a field called "cp_from_cp". If this is called at >all, your NVRAM is overflowing and denying write requests to users. >cp_from_cp means you're in the middle of checkpointing the first half of >NVRAM and yet the second half already filled up and needs to checkpoint as >well. This is VERY bad. > >

1 0

nfs performance problems
by Leonard, Roger 17 May '02

17 May '02

I have a 740 with ~600GB of data. it is running 5.36r2. it has 512mb of ram and is running oc-3 (155mb) atm. periodically i see the good old "nfs server suchandsuch not responding" on various mid-range solaris boxes. and users complain about bad performance. general network performance is good for file transfers etc. there is a Ontap bug 12898 for 5.36r2 relating to these indications but in a TCP nfs environment. most of my solaris NFS mounts are V3 UDP. i am going to do an upgrade anyway but I was wondering if anyone else has seen this type of indication. also, see below the output of my sysstat. this goes on most of the day. it looks to me as if my filer is getting hammered. what do you guys think? thanks roger CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache 51% 4074 1 0 2854 3059 2040 908 0 0 1 100% 4146 0 0 4803 5471 6737 10593 0 0 1 67% 3228 1 0 4511 5770 3899 3587 0 0 1 81% 1894 0 0 3431 3363 8564 10631 0 0 1 71% 2308 0 0 5999 6482 3671 2843 0 0 1 57% 2108 1 0 2645 3359 4482 7641 0 0 1 65% 2882 1 0 5364 8072 3736 0 0 0 1 99% 2466 0 0 8339 7281 5808 6465 0 0 1 96% 2264 2 0 2869 3556 10404 12876 0 0 1 84% 4252 0 0 4003 2174 4090 12652 0 0 1 70% 3063 0 0 2939 4694 4820 5616 0 0 1 67% 3098 1 0 3097 4024 2878 10606 0 0 1 66% 4958 0 0 5352 2693 2137 0 0 0 1

4 3

ORA-07353 on NetApp F760 at 100MBit
by Joseph Bishop 17 May '02

17 May '02

Hello folks, Has anyone seen a message like this while trying to create an Oracle datafile from SUN Solaris under heavy Network load? ERROR at line 1: ORA-01119: error in creating database file '/blah/blah/blah.dbf' ORA-07353: sfccf: close error, unable to close file. SVR4 Error: 11: Resource temporarily unavailable There isn't much noted on the Oracle site. I am just curious if anyone has seen it and if it is an NFS timeout problem? I have checked /var/adm/messages but I don't seen anything. Joe

1 0

Antivirus Scanning in DOT 6.1.3
by Allen, Pat 17 May '02

17 May '02

Hi everyone, I'm taking a look at the antivirus scanning capabilities in DOT 6.1.3. I've looked at the white paper that NetApp put together (http://www.netapp.com/tech_library/3107.html) and have some questions for those of you who might have implemented it: 1. What kind of performance impacts do you see in production environments? 2. What type of network interconnects are you using? In particular, would a VLAN on top of gigabit Ethernet be OK? 3. Are there any recommendations about the vendors? We use Trend ScanMail for our Exchange servers and have been happy with that product. We also use NAV for our desktops. 4. Are there any horror/satisfaction stories or questions I missed? THANKS! --- Pat Allen (pat(a)mbari.org) Monterey Bay Aquarium Research Institute (MBARI) 7700 Sandholdt Rd, Moss Landing, CA 95039 (voice) 831-775-1724; (fax) 831-775-1620

1 0

lock manager in 6.2 is broken....fyi
by Jeff Kennedy 17 May '02

17 May '02

There is now a bug id regarding the lock manager in 6.2. Using ReflectionX to nfs mount a volume under 6.2 causes failed writes/copies with a 'delayed write error' message. After a bit of testing with support they have concluded that it's definitely a bug in OT. Apparently the lock manager was completely rewritten for 6.2. The bug id is 71703. It may still be internal. -- ===================== Jeff Kennedy Unix Administrator AMCC jlkennedy(a)amcc.com

1 0

Fastlane DM Administrator
by Don.Hickey＠alcatel.com 17 May '02

17 May '02

I was wondering if anyone uses this product against their filers? I have an 840 cluster running 6.1.2r2 and would like to have an operations group administer local cifs groups and directory permissions. We use this on our NT side and it works well but we've run into a few bumps with the filer. I don't want this group to have root permission but I don't want to be bothered updating permissions for every directory on the thing. Is there another product that is commonly used? Thanks Don

1 0

Re: Do file locks affect snapshots?
by Don.Hickey＠alcatel.com 17 May '02

17 May '02

Steve One thing to note is that if you are running in a multiprotocol environment you may find some snapshot files inaccessable through windows. The reason: If you access a file via NFS only and later try to view/recover that file from a snapshot directory with windows the result will be permission denied. This is because the first time you access a file with windows it converts the filename to unicode format. It can't do this inside a snapshot since it is a read only filesystem. There are cifs and wafl options you can set to avoid this. You noted that later you could access the file via w2k.... was it the same snapshot or a later snap? I suspect if you were doing a filer wide virus scan then the file would have been touched (in the live directory) thus converting it to unicode.... Just a thought.... Don Steve Losen <scl(a)sasha.acc.virginia.edu> Sent by: owner-toasters(a)mathworks.com 2002-05-09 09:59 AM To: toasters(a)mathworks.com cc: Subject: Do file locks affect snapshots? We are running ONTAP 6.1.1R2. I thought I read somewhere that there are no file locks in a snapshot. The snapshot is read only, but no files are locked, so you can open any file if you have the necessary permissions. However, I was running a virus scan (Norton) on a snapshot and it failed to open a few files, according to the log. I'm pretty sure it was not a permissions problem. We use unix passwd authentication for CIFS and therefore all files have unix permissions. The CIFS user was root (uid 0) and Norton successfully scanned about 180,000 files, many of which are readable only by owner (and superuser). Norton failed to open 16 files. Here is an error from the log: Scan could not open file F:\weekly.0\a\af\afc3f\private\Disparity.doc [00000012] (We made a read only share out of the .snapshot directory and attached it to drive F) So that got me wondering about file locks. If a file is locked when a snapshot is created, is the lock included in the snapshot? And if so, does it lock the file in the snapshot? Or else, if a file in the active filesystem is currently locked, does that affect access to the corresponding file in a snapshot? Later on, using the same share as above, I was able to copy the file to the local desktop, so W2000 was able to open it. Maybe it's a Norton thing. I'll try and look up that error code [00000012]. Is this a CIFS error perhaps? By the way, Norton failed to open some other files with error code [00000003] but the ones I checked were all symlinks that the filer could not follow, so that makes perfect sense. Steve Losen scl(a)virginia.edu phone: 434-924-0640 University of Virginia ITC Unix Support

1 0

CIFS Virus Scanning
by Strange, John 16 May '02

16 May '02

Hey Gang, I need to create a cifs share that I can virus scan monthly, I would like to set this up with CIFS, but the underlying unix permissions seem to override and I get permission denied on the files. Is there anyway I can setup a CIFS share that does not use the underlying unix permissions. I had the share set to full permissions and it does get some of the things but a few locked down unix directories seem to get permission denied on. Unless someone knows a good virus scanner that will run on linux/solaris. Thanks, - John

3 2

Re: CIFS Virus Scanning
by Joseph Bishop 16 May '02

16 May '02

John, Our email group has just gone through research on anti-virus for Solaris. They suggest looking into Trend Micro VirusWall . Cheers, Joe At 11:29 AM 5/16/02 -0400, Strange, John wrote: >Hey Gang, > >I need to create a cifs share that I can virus scan monthly, I would like to >set this up with CIFS, but the underlying unix permissions seem to override >and I get permission denied on the files. Is there anyway I can setup a >CIFS share that does not use the underlying unix permissions. > >I had the share set to full permissions and it does get some of the things >but a few locked down unix directories seem to get permission denied on. > >Unless someone knows a good virus scanner that will run on linux/solaris. > >Thanks, > >- John

1 0

Oracle 8.1 vs. netapp
by Chris Moll 16 May '02

16 May '02

Anybody seen any performance changes with going to oracle 8.1.7 (from 8.0.5) with the databases on netapp storage? (host is sun enterprise 5000, solaris 2.6). Many batch jobs ran dramatically slower on a DB that was upgraded from 8.0 to 8.1. Fingers are getting pointed to the netapp, partly 'cause nobody wants to contemplate down-revving the database. The fact that the jobs ran fine on the netapp before the upgrade would seem to exonerate it but it's a convenient target... A few efforts at tuning the DB seemed to reclaim much of the loss; I'm no oracle expert but it sounds to me like the DB is far from optimal, for example it's set to a 2k block size which seems questionable to me. Anyway - anyone encountered anything similar? Chris Moll Lawrence Berkeley Laboratory

2 3

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