toasters

toasters@lists.teaparty.net

13513 discussions

Fixed: NFS processes using mount point question...
by julie.stephen＠vf.vodafone.co.uk 06 Aug '98

06 Aug '98

Hi, We found out that if you specify the hostname:/mountpoint with fuser -k you can kill off dead/hung process. We were hitting the problem due to typing fuser -k /localmountpoint. Thanks anyway! Julie

1 0

CIFS and NT ACLs experiences
by Garrett Burke 06 Aug '98

06 Aug '98

Hi All, A word of warning - OnTap 5.1 with its full NT ACLs has some problems. From bitter and ongoing experience I wouldn't advise using it on a production filer. If you are running this I'd advise getting onto netapp to get the latest bug fixes. I have been advised to run whack. Still haven't dared switch quotas back on. I used to think that those giving off about the quality of netapp SW were being a bit harsh, but now I'm not so sure. I know I'm running an early release, but never again. The troops will just have to wait for the new fetures. GB -- Garrett Burke, Network Admin,Managed Solutions Corp. 32 Upper Mount St., Dublin, Ireland. burkeg(a)msc.ie +353-1-661-4840 A jug of wine, a loaf of bread and thou.

4 6

NFS processes using mount point question...
by julie.stephen＠vf.vodafone.co.uk 06 Aug '98

06 Aug '98

Hi, We are using two NetApp F630's which replicate each other so that in the event of failure of one, the clients using that one can be switched over. We are using HP clients so would like to run fuser -ck to kill the processes hanging on the dead mount points. Unfortunately this does not work since fuser first attempts to stat the mount point and then seems to hang indefinately. We can kill the processes manually, but putting the list of "dead" processes together is going to be difficult without fuser. We thought of running fuser at regular points throughout the day while both servers are available and using this list of processes, but this is not a guaranteed way of capturing everything. Does anyone have any ideas ? Does lsof behave better on hung nfs mount points ? Many thanks Julie Stephen Vodafone Ltd. UK.

1 0

Re: Renaming quota trees
by Daniel Quinlan 01 Aug '98

01 Aug '98

Brian Tao <taob(a)risc.org> writes: > The magic that "quota qtree" performs isn't needed for the rename > then? I thought there was something in /etc/db/quota.db had to be > updated to stay consistent. In any case, I'll give that a try to > see how it works. Thanks. I could be wrong. I'd suggest trying a test case. I know that after the "mv", "quota qtree" does show the new name, so I assume what I described will work. - Dan

1 0

Re: Renaming quota trees
by Daniel Quinlan 31 Jul '98

31 Jul '98

Brian Tao <taob(a)risc.org> writes: > Is this possible? We have some transition requirements that would > be greatly eased if we could fire up a quota tree under, say, > /devtree at first, and then rename it to /tree once we roll the new > filesystem into production. Yes, it is. Just shut off quotas, "mv" the tree, change /etc/quotas, and restart quotas. You may have some stale file handle issues on clients afterwards if there was an old tree with similarly named files in active use. Dan

2 1

Moving a secondary volume to primary
by marc_merlin＠magic.metawire.com 30 Jul '98

30 Jul '98

Hi, I have a filer with two volumes, and I'd like to remove the first volume and keep the second one only. Is it as easy as doing: - copy of /etc from the first one to the second one - remove the firt volume - boot on floppy - download - reboot. ? Also, not that I have any reason to do that, but if I wanted to put the first volume back, how does the filer know which volume to boot on? Marc -- Home page: http://magic.metawire.com/~merlin/ (browser friendly) Finger merlin(a)magic.metawire.com for PGP key and other contact information

6 6

None
by Blake Cooper 29 Jul '98

29 Jul '98

subscribe toasters enteract-lists-toasters(a)news.enteract.com

1 0

Subscribe Toasters
by Poh G. K. Desmond, OI/SAE 29 Jul '98

29 Jul '98

1 0

midnight disk failure
by Daniel Quinlan 28 Jul '98

28 Jul '98

Does something about the timing of this disk failure strike anyone as odd or is it just me? I'm curious to know whether some checks only done once an hour, once a day, or if something else is at play. This happened on an 18 day old F630/FiberChannel running 5.0.1D1. Dan ------- start of cut text -------------- Mon Jul 27 23:00:00 PDT [statd]: 11:00pm up 17 days, 18 mins, 319161196 NFS ops, 0 CIFS ops, 0 HTTP ops Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008e2330): request failed after retry #20 Mon Jul 27 23:59:59 PDT [raid_stripe_owner]: Read on data disk 8.29 in volume vol0, RAID group 0, failed; reverting to degraded mode. Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008de230): opcode=0x2a sector 0 not ready (2 4, 1) Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008de230): request failed after retry #20 Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008da610): opcode=0x2a sector 0 not ready (2 4, 1) Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008da610): request failed after retry #20 Mon Jul 27 23:59:59 PDT [raid_stripe_owner]: Read on data disk 8.29 in volume vol0, RAID group 0, failed; entered degraded mode. Mon Jul 27 23:59:59 PDT [raid_stripe_owner]: One disk is missing from volume vol0, RAID group 0. A "hot spare" disk is available and the missing disk will be reconstructed on the spare disk. Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008d9910): opcode=0x28 sector 0 not ready (2 4, 1) Mon Jul 27 23:59:59 PDT [isp2100_main]: Disk 8.29(0xfffffc00008d9910): request failed after retry #20 Tue Jul 28 00:00:00 PDT [statd]: 12:00am up 17 days, 1:18 319612667 NFS ops, 0 CIFS ops, 0 HTTP ops Tue Jul 28 00:00:00 PDT [asup_main]: System Notification mail sent ------- end ----------------------------

2 1

remove snapshot, or what?
by Dan Bethe 26 Jul '98

26 Jul '98

Hi there guys. I'm not the regular NetApp tech around here, so I have a quick weekend maintenance question. Our filer had about -60MB (less than zero) free when I last checked it. We're deleting plenty of space now, but it's of course moving the deleted data to the snapshot storage. Is it proper to delete the snapshot storage manually? Or will the filer give it up in favor of the normal NFS export? We have, for example, some '.snapshot' directories that contain redundant data. Is 'rm -rf' proper there, or what? Thanks a lot.

2 1

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