toasters May 1999

toasters@lists.teaparty.net

90 participants
122 discussions

Re: 100% Cpu using quotas under 5.x
by Ketil Kirkerud Elgethun 14 May '99

14 May '99

Following up on my own posting: It looks a _lot_ like SNMP is the culprit - we have a script, run daily, that extracts information (via SNMP) from the filer about qtrees. Running this scipt by hand made the filer go bananas (CPU pegged at 100%, until i did a 'quota off;quota on'. The interesting bit of the script is as follows : for $host (ListService("netapp-server", "ALL")) { mkdir($host, 02775) unless (-d $host); # netapp.netapp1.quota.quotaState.0 ($state) = snmpget ($host, $cmty, '1.3.6.1.4.1.789.1.4.1.0'); if ($state != 2) { warn "$host does not have quotas initialized (state = $state)"; next; } $id=0; while (1) { $id++; # netapp.netapp1.quota.qrTable.qrEntry.qrIndex # netapp.netapp1.quota.qrTable.qrEntry.qrType @ret = snmpget ($host, $cmty, '1.3.6.1.4.1.789.1.4.3.1.1.'.$id, '1.3.6.1.4.1.789.1.4.3.1.2.'.$id, '1.3.6.1.4.1.789.1.4.3.1.8.'.$id, ); last if $ret[0] < 0; next if $ret[1] != 3; # Only qrTypeTree $ret[2] =~ y/A-Z/a-z/; ($fs = $ret[2]) =~ s%/%-%g; print CFG <<EOF; << print to mrtg.cfg-file >> EOF; } } So. This works fine on 4.x - and it looks like it works fine on a F330 running 5.1pre (but this box does not have 160K users on it). Anyone have any ideas ? -- ---Ketil Kirkerud Elgethun, SOL System

1 0

100% Cpu using quotas under 5.x
by Ketil Kirkerud Elgethun 14 May '99

14 May '99

I have an interesting problem with one of my F540s (maxed out on memory and disk controllers, approx 40 disks) : It used to run 4.3.1D3 - on recommendation from my vendor (there was a bug in the release we were running that bit us), I upgraded to 5.2.1P2D4 - the NetApp is serving the home directories of approx. 160.000 users (largest UID approx 161000)- and we use default quotas. A few (less than 500) of the users have specific quotas. This worked very well under 4.x - it still works, but after having been up and running for 20 hours or so, the NetApp becomes _very_ sluggish, with 100% CPU consumed - it still serves nfs requests, but very slowly. Doing 'quota off; quota on' gets the box going again - but this is hardly ideal. I can imagine some mechanisms that would cause this: 1) Bugs have crept into the quota software in OnTap 2) Bugs have crept into syslog - we run with the default syslog config, and we get a _lot_ of messages about exceeded quotas. The messages differ slightly : Thu May 13 21:26:12 MES [wafl_hipri]: uid 75355 tid 15: disk quota exceeded on volume vol0 Thu May 13 21:27:31 MES [de1]: uid 75355 tid 15: disk quota exceeded on volume vol0 Thu May 13 22:26:44 MES [de0]: uid 18002 tid 16: disk quota exceeded on volume vol0 Thu May 13 22:26:44 MES [wafl_lopri]: uid 18002 tid 16: disk quota exceeded on voume vol0 (these are just excerpts, we get a _lot_ of them as well as quite a few "Thu May 13 21:40:59 MES last message repeated 7 times" related to the aforementioned messages) It's the [de0|de1|hipri|lopri]-bit that puzzles me a bit - can anyone explain ? I couldn't care less about users exceeding their quotas - and wouldn't mind being able to turn off the reporting. (I would still like to be able to know about qtrees exceeding their quotas, however). My gut feeling is that this is the root of the problem. (too many 'quota exceeded' messages overloading syslogd somehow, that is) 3) The quota mechanisms can't take the load (but then, why did it work so well in 4.x ?) Has anyone else seen tha same problem ? (It's being escalated through our vendor - my experience with them and NetApp is good, but I'd like to find out if anyone else has seen anything like this problem) -- ---Ketil Kirkerud Elgethun, SOL System

1 0

RE: Netapp filers & MS Exchange
by Jonas Jonsson 13 May '99

13 May '99

I've tried to do some "brain work" on this and come to the following conclution: 1. How can I create a "Exchange consistent" snapshots to provide for the backup of Exchange (snapshot) with of the shelf software like Legato Networker (no Exchange API - just files). Snapshtos will do for this. Just remember Exchange might be running for a while in a non-consistant mode before it decides to do a halt. 3. In case of a need for partial data recovery - is there a Filer local copy, the convert a part of a snapshot from RO to RW mode besides a through the host copy can can always be done but consumes a lot of bandwidth, time and resources. Would appreciate any thoughts/suggestions/enlightnings/pointers to the above subjects. Partial recovery I think might be a problem which ever solution you chose. Exchenge, as far as I know, is working on a information blob. The best solution is to use snapshots and recover to a stable point in time, preferably some hours back in time. If anyone has information on the workings/non-workings of Exchange and how it stores/not stores information. Pleas comment or I'll go further on my darkend path ;) /// Regards, Jonas

1 0

Re: Netapp filers & MS Exchange
by sirbruce＠ix.netcom.com 13 May '99

13 May '99

On 05/13/99 07:30:54 you wrote: > >I am willing to implement Exchange on NetApp. I have two exchange >servers with each holding about 20GB of data. I have seen the document >on NOW. It is imperative to have close to 24*7 service from the Exchange >servers. You can't, because you've already chosen Exchange. There's nothing wrong with Netapp in this case, but Exchange is a crappy piece of software that may give you 24*6 service at best. Bruce

1 0

Re: Syslog message
by sirbruce＠ix.netcom.com 13 May '99

13 May '99

On 05/13/99 10:24:30 you wrote: > >Jay Orr wrote: >> >> One last question on this topic -- I had removed the syslog.conf file from >> the toasters yesterday to stop the messages, but it kept logging anyway. >> If it was a unix box, I'd just send a HUP signal to the syslogd, but how >> does a toaster know the file has changed and needs to reconfigure (or that >> the file is missing)? > >It checks the file every so often (every 30 seconds, IIRC), and updates >automagically. And, just the clarify, if you remove the file entirely I think it just falls back to the "default" configuration. It would take an empty syslog file to eliminate most messages (but I think some would still get logged anyway). Bruce

1 0

change notification handles
by Todd C. Merrill 13 May '99

13 May '99

Some users here have recently reported significant slowdowns with some testing suites. They claim the addition of a few directories to a certain path triggered the performance degradation and are asking me about change notification handles for the CIFS service. Searching the NOW site, CD docs, and the toaster mailing list archives just gave me hits on "Yup, it's there," but nothing else. Can anyone reference a pointer on more information? What the ONTAP limit is? If there are any tunable parameters? NetApp Release 5.2.1P2: Tue Mar 16 10:41:03 PST 1999 Until next time... Todd C. Merrill The Mathworks, Inc. 508-647-7792 24 Prime Park Way, Natick, MA 01760-1500 508-647-7012 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

2 1

Syslog message
by Jay Orr 13 May '99

13 May '99

Anyone seen this or know what it means? I just setup the syslog to send to another system (it didn't have a syslog.conf file before). I set it up like the example so it's sending *.err and kern.* to another system. Now I'm getting : May 12 11:57:31 hardrock Wed May 12 11:57:32 CDT [portmapper]: portmap: prog (300214,1,201) not registered. May 12 11:57:31 obriens Wed May 12 11:57:32 CDT [portmapper]: portmap: prog (300214,1,201) not registered. in the syslog (hardrock and obriens are the two toasters). It's repeating every half hour or so... Thanks in advance for the help... ----------- Fujitsu - Nexion, St. Louis, MO Jay Orr (314) 579-6517

2 5

Netapp filers & MS Exchange
by Itzik (Itzhak) Meirson DSc 13 May '99

13 May '99

I am willing to implement Exchange on NetApp. I have two exchange servers with each holding about 20GB of data. I have seen the document on NOW. It is imperative to have close to 24*7 service from the Exchange servers. My questions are related to backups of the Exchange on NetApp: 1. How can I create a "Exchange consistent" snapshots to provide for the backup of Exchange (snapshot) with of the shelf software like Legato Networker (no Exchange API - just files). 2. As circular logging should be disabled - Is there any API/procedure that will clean up the transaction logs as it is done by "Exchange aware" backup clients but without actually doing the backup to a physical device. 3. In case of a need for partial data recovery - is there a Filer local copy, the convert a part of a snapshot from RO to RW mode besides a through the host copy can can always be done but consumes a lot of bandwidth, time and resources. Would appreciate any thoughts/suggestions/enlightnings/pointers to the above subjects. -- Itzik Meirson <mailto:imeirson@xpert.com> Xpert Integrated Systems Systems & Storage Integration Specialist

1 0

Re: Dual power on F760
by sirbruce＠ix.netcom.com 13 May '99

13 May '99

On 05/11/99 10:39:39 you wrote: > >On Mon, 10 May 1999, Timothy Moore wrote: > >> The machine 'hitz' is a F760 in my lab running 5.3. At the back of the machine I >> turned one power supply off then back on, repeat with the other. The messages were >> generated to the console, system logger and as event notification email. > >The machines I was referring to, are NetCaches C760 running 3.3R2D12. We >did the same with switching the power supplies on and off and the Ok >message never came, only the failed one did. Sounds like a bug. Contact support. :) Bruce

1 0

Re: Solaris/NIS/shadow-passwords
by sirbruce＠ix.netcom.com 13 May '99

13 May '99

On 05/11/99 08:42:47 you wrote: >As always, your results may vary - but it'll be a long time before >I bother attempting to use NetApp NIS again. I have the files >all copied locally and can't see any reason to go back..... Yep. When I tested the NIS myself and eventually went back to using and recommending that configuration. Making the copying of files part of the ypmake isn't all that difficult. As with any new Netapp feature, the initial implementation is done quickly and leaves much to be desired, especially in the areas of reliability and ease-of-use. After a few iterations they get it right; this is not unlike the way new features appear in other OSes, really. You've seen the same thing with CIFS, NFS, and so on. I'm sure the current NIS code is much better than it was when they first released it. I've found many sysadmins who use NIS don't like to make important servers NIS clients for various reliability reasons. They run local YP masters on them; with the filer this means local files. Bruce

1 0

← Newer
1
2
3
4
5
6
7
8
...
13
Older →

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 May 1999