toasters

toasters@lists.teaparty.net

2 participants
13520 discussions

DataONTAP 5.0.1 IP cache bug
by dfinkels＠siac.com 06 Jul '98

06 Jul '98

-- Daniel Alex Finkelstein Infrastructure Enhancement Group Systems and Technologies Support phone 212.383.2951 pager 917.427.1630 Securities Industry Automation Corporation SIAC has found a bug with the 5.0.1 OS of DataONTAP. There appears to be no time-to-live on the cached IP addresses of hosts. Changes in the IP addresses of hosts in DNS or NIS are not recognized until the NetApp is either rebooted or "exportfs -a" is executed. To force re-reading of DNS and NIS data, we have set up cronjobs to reboot our 6 NetApp 520's weekly.

3 3

Re: Trouble with Unix groups running NIS
by sirbruce＠ix.netcom.com 06 Jul '98

06 Jul '98

On 07/06/98 11:54:37 you wrote: > >Hello, > >I'm trying to set up group quotas on my filer, but I keep getting the >following message back : > >tap> quota on >Mon Jul 6 11:38:07 BST [rc]: getgrp: Input buffer not large enough >quota: group name 'uk' not found (/etc/quotas line 7). >quota: quotas not turned on. >tap> > > >What is this getgrp thing and how can I enlarge the input buffer ? Well, my GUESS is that the error, essentially, is saying that your entry for the group "uk" is too long. Normally I'd say check the group file on the filer but you mention NIS in the subject line, so I'd say check the group file on the NIS master. The filer *should* be able to accept group lines as big as your UNIX server. Either your group line is actually too big on your UNIX server (in which case it probably was broken and you just didn't know it), or the filer's buffer needs to be increased. In the latter case, you'll have to wait for a patch or new release that has a bigger buffer. Bruce

2 1

Trouble with Unix groups running NIS
by Raymond Brennan 06 Jul '98

06 Jul '98

Hello, I'm trying to set up group quotas on my filer, but I keep getting the following message back : tap> quota on Mon Jul 6 11:38:07 BST [rc]: getgrp: Input buffer not large enough quota: group name 'uk' not found (/etc/quotas line 7). quota: quotas not turned on. tap> What is this getgrp thing and how can I enlarge the input buffer ? "I read a dictionary once; I thought it was a poem about everything" -Anon

1 0

SNMP support on filers
by agy 03 Jul '98

03 Jul '98

Hello All, NetApp is planning to enhance the SNMP support in the filer in an upcoming release of the OS. Due to some risk factors, I'd rather not promise that it will be the next major release (next year), but that is the current plan. Things we definitely plan to do when this support is added: (a) GETs for general CIFS-related filer parameters. (b) GETs for statistics on all SMB requests, both cumulative and per client. (c) Generic TRAP support, with settings and parameters for each available trap manageable through Filerview. I would welcome feedback on: (a) Additional GETs you think are necessary. (b) Trap support considerations. (b) Specific traps you think should be implemented. (d) Other things pertaining to SNMP that would be helpful. It would help us if you could indicate the relative importance-- to your company's future infrastructure plans--of things that you suggest. Thanks, Alan --------------------------------------------------------------- Alan G. Yoder agy(a)netapp.com Network Appliance, Inc. 2770 San Tomas Expressway voice 408-367-3031 Santa Clara, CA 95051 fax 408-367-3451 ---------------------------------------------------------------

3 2

bug fix for na_snapmon
by Daniel Quinlan 02 Jul '98

02 Jul '98

Hi everyone. I just fixed a minor bug in na_snapmon (it would try to delete dump snapshots all of the time instead of just the ones that were more than two days old). The copy that I put up on my NetApp web page doesn't have that problem: http://www.pathname.com/~quinlan/netapp/ That could actually cause some problems for older releases (before 4.3x) affected by bug #650 that would crash the filer if you tried deleting a snapshot that's in use. Now, na_snapmon makes the assumption that dumps older than two days are safe to delete, or at least you're running 4.3 or later. Dan

1 0

Re: SNMP support on filers
by Daniel Quinlan 02 Jul '98

02 Jul '98

Hi Alan. There are a few things: - Information located in the "Tree" field of the "quota report" is missing from the SNMP quota information. (this is bug 8704, and log 35271) - Putting CIFS and HTTP statistics into SNMP, starting with the same information that's in "sysstat", but also the other information you can get from the "cifs" command. - getting snapshot information via SNMP - per-client NFSop/CIFSop/HTTPop statistics Thanks. Dan

1 0

Re: SNMP support on filers
by sirbruce＠ix.netcom.com 02 Jul '98

02 Jul '98

On 07/02/98 11:01:03 you wrote: > >Hello All, > NetApp is planning to enhance the SNMP support in the filer >in an upcoming release of the OS. Due to some risk factors, >I'd rather not promise that it will be the next major release >(next year), but that is the current plan. > > Things we definitely plan to do when this support is added: >(a) GETs for general CIFS-related filer parameters. >(b) GETs for statistics on all SMB requests, both cumulative > and per client. >(c) Generic TRAP support, with settings and parameters for > each available trap manageable through Filerview. > > I would welcome feedback on: >(a) Additional GETs you think are necessary. >(b) Trap support considerations. >(b) Specific traps you think should be implemented. >(d) Other things pertaining to SNMP that would be helpful. My main advice would be to make sure anything obtainable through SNMP is also obtainable from the command line and from the remote GUI; i.e. if you're going to have per-client stats for CIFS in SNMP then it should be available via a "cifs stats" command or similar as well. Bruce

1 0

na_snapmon: snapshot monitor
by Daniel Quinlan 01 Jul '98

01 Jul '98

I finally got fed up with deleting snapshots by hand when they started using too much disk space, so I created a simple snapshot monitor that automatically deletes snapshots when they start using more than 100% of the space reserved for snapshot usage. Now, I configure each filer to schedule more snapshots than I really want and I use only "snap reserve" to limit how much disk usage is taken by the snapshots. If a filer's data turnover rises, it automatically scales the number of snapshots back. For example: Jun 26 10:30:36 sodium na_snapmon: rsh uranium snap delete hourly.3 Jun 26 10:33:03 sodium na_snapmon: rsh neptunium snap delete nightly.1 Jun 26 10:34:49 sodium na_snapmon: rsh thorium snap delete hourly.2 Jun 26 10:40:04 sodium na_snapmon: rsh neptunium snap delete hourly.3 It will also try to delete dump snapshots that are older than two days old, and it will also ignore snapshots that have been specially created or renamed. The best way to run this is as root cron job, every ten minutes or so. This runs on Linux now, but should also work on Solaris. Other Unix systems with very minimal modification. ------- start of cut text -------------- #!/usr/bin/perl # na_snapmon - monitor snapshot usage and keep it under control # Copyright (C) 1998 Daniel Quinlan # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. use Getopt::Std; require "find.pl"; sub usage; sub trim_snapshots; sub snap_delete; my $limit; my $prog; $prog = $0; $prog =~ s@.*/@@; getopts("hd"); if ($opt_h) { &usage; exit 0; } foreach $dir (@ARGV) { my $df_worked = 0; if (! -d $dir) { warn("$prog: $dir: no such snapshot mount point"); } open (DF, "df -k $dir |"); while (<DF>) { if (/[0-9]%/) { $df_worked = 1; ($fs, $blocks, $used, $avail, $capacity, $mount) = split; if ($capacity !~ /^[0-9]?[0-9]%/) { $fs =~ s/:.*//; trim_snapshots($fs, $dir); } } } close (DF); } sub trim_snapshots { my ($filer, $dir) = @_; my (@list, @snapshots, %atime); # read snapshot list opendir(DIR, $dir) || die "can't opendir $dir: $!"; @snapshots = grep { ! /^\.\.?$/ && -d "$dir/$_" } readdir(DIR); closedir DIR; # sort by atime foreach $entry (@snapshots) { $atime{$entry} = -A "$dir/$entry"; } @snapshots = sort {$atime{$a} <=> $atime{$b}} @snapshots; # maybe delete something while ($s = (pop @snapshots)) { # try to delete old dump snapshots, but since this might # fail, keep trying to delete after this one. if ($s =~ /^snapshot_for_dump\.[0-9]+$/ && time - $atime{$s} >172800) { snap_delete($filer, $s); next; } # skip manual snapshots next if ($s !~ /^(hourly|nightly|weekly)\.[0-9]+$/); # leave one snapshot at all times next if ($s =~ /^hourly\.0$/); snap_delete($filer, $s); last; } } sub snap_delete { my ($filer, $snapshot) = @_; if ($opt_d || ($> != 0)) { print "debug: rsh $filer snap delete $snapshot\n"; } else { system("logger -t na_snapmon \"rsh $filer snap delete $snapshot\""); system("rsh $filer snap delete $snapshot"); } } sub usage { print <<EOF; usage: $prog [-hd] [list of snapshot mount points] -h print this help -d debugging mode: don't do, just show EOF } ------- end ----------------------------

1 0

Quotas
by Raymond Brennan 01 Jul '98

01 Jul '98

I'm sort of new to all this, so here goes : I want to create a root directory called, say, dog. I then want to create a subdirectory called pup. I want to limit the size to which pup can grow to, say, 50 Mb. The thing is, the quota qtree command won't accept slashes in the qtree name. Also, if I create "dog" by doing # mkdir dog then do #cd dog # mkdir pup then go to the quotas file on /etc and do /dog/pup 10G 10K and run quota on I get "/dog/pup is not a quota tree. In short, how do I set quotas on file systems a number of steps down the tree ? Why can't I do /dog/pup 5G /dog/puppy 5G

1 0

OT 5.?
by Weeks, Thomas 01 Jul '98

01 Jul '98

Just curious... how is the OT 5.x software working for eveyone? Any problems? Stability probs? Backout (to 4.3.4) problems? What about quotas and the like? Just curious... Tom > -----Original Message----- > From: Mark D Simmons [SMTP:mds@gbnet.net] > Sent: Monday, June 29, 1998 1:48 AM > To: quinlan(a)transmeta.com; toasters(a)mathworks.com > Subject: Re: what data is being used? > > I recall asking for this in an RFE sense a year ago as it would be > incredibly > useful. Sun just put the capability to track the load on disks at the > partition > level in Solaris 2.6 - this sort of stat on a filer would be just as > useful IMHO > > Per quota tree seemed most natural to me as that is already a > distinctive > structure in filer housekeeping meta-data. Directory level gathering > might be > too much load to put on a filer which is meant to perform under high > demand... > > ...how about an admin command that flips a bit on a directory and > makes it a > watchpoint in the filesystem - accesses against anything beneath it > count on > that directory's stats. Even if the total number of such points were > limited > to put a boundary on overhead in the filer, it would still be a useful > analytical tool.

1 0

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