toasters May 2000

toasters@lists.teaparty.net

122 participants
148 discussions

Re: File-specific NDMP backups?
by Dave Toal 23 May '00

23 May '00

> Bill, Thanks for the assistance. > The netapp returns this: > > smith> dump 0uf rst0a /vol/brs1/argus/brs-data1/prod/alr2 > DUMP: creating "/vol/brs1/../snapshot_for_backup.98" snapshot. > DUMP: Using Partial Volume Dump > DUMP: Date of this level 0 dump: Tue May 23 22:24:22 2000. > DUMP: Date of last level 0 dump: the epoch. > DUMP: Dumping /vol/brs1/argus/brs-data1/prod/alr2 to rst0a > DUMP: mapping (Pass I)[regular files] > DUMP: mapping (Pass II)[directories] > DUMP: estimated 95116 tape blocks. > DUMP: dumping (Pass III) [directories] > DUMP: dumping (Pass IV) [regular files] > DUMP: dumping (Pass V) [ACLs] > DUMP: 97906 tape blocks > DUMP: DUMP IS DONE > DUMP: Deleting "/vol/brs1/../snapshot_for_backup.98" snapshot. > Tue May 23 22:25:04 GMT [smith: ndmp_session_0]: ../common/ndmpd/tape.c:788 ndmpdTapeRead: read error. > smith> > > NetBackup returned "media write error". > > <sigh> I've just discovered that the tape won't unload from the > netapp's DLT: it's rewound but the mechanism (storagetek 9714) is jammed. > Which means I get to drive through Boston to go play with it. > > I don't want to attribute this to the read error. Generally I > don't have problems with this sort of thing, though. Also, the netapp apparently wasn't happy with being poked at: > Subject: > Cluster Alert from smith > Date: > Tue, 23 May 2000 22:44:41 +0000 > From: > unixgroup(a)t-t.com > To: > unixgroup(a)t-t.com > > > > > REBOOT on smith at Tue May 23 22:44:41 GMT 2000 > Most annoying. No downtime, though. Dave > >

1 0

File-specific NDMP backups?
by Dave Toal 23 May '00

23 May '00

Folks, Another clueless-newbie question, as is my wont: I've got NetBackup 3.1.1GA running against a 740 filer with 5.2.1.P2. When I attempt an ndmp backup of a directory a few layers down in a /vol, it fails. Backups of the /vol itself work fine. /vol/brs1/prod/brs-data1/ger1 -- fails /vol/brs1 -- succeeds Is this an inherent limitation of ndmp at this rev level? I'd rather not set up nfs backups of these specific directories, but I'm not sure how else to do it. Dave -- Dave Toal Unix Dude Thomson & Thomson N. Quincy, MA

1 0

NFS problems between Tru64 UNIX and F760's (5.3.5)
by Cliff Nadler 22 May '00

22 May '00

We've been seeing alot of 'NFS3 server servername not responding still trying' followed by 'NFS3 server servername ok' messages. Alot of them. It's relaly only on our DEC hardware running Digital UNIX 4.0d. Has anyone seen this before and found a fix? It's starting to affect our performance, especially trying to put some database files on them. It's happening on multiple DEC boxes, against multiple filers (3 F760). -- -------------------------------------------------------------------------- Cliff Nadler Collective Technologies, Inc. (a Pencom Company) cnadler(a)colltech.com (617)-443-1144 -------------------------------------------------------------------------- To win, you must treat a pressure situation as an opportunity to succeed, not an opportunity to fail Gardner Dickinson

3 2

Interaction of /etc/hosts.equiv and telnet.hosts option
by Chris Thompson 22 May '00

22 May '00

I've just spent some time trying to work out why I couldn't rsh to our F740 from some newly installed machines even though I had added them to /etc/hosts.equiv. It turns out, apparently, that the option telnet.hosts list gets into the act: if the rsh'ing host isn't in the telnet.hosts list, then it gets a prompt rejection, and nothing is logged in /etc/messages; if it gets past that, then /etc/hosts.equiv is checked for host and (possibly implied) userid, and if *this* check fails then a "[rshd_0]: Permission denied to rsh request from ... at ..." is logged to /etc/messages. This is with 5.3.5. Has it always been like this? I could have sworn not, but the two lists have been in step here for a long time, so I could be wrong. There's nothing in the hosts.equiv(5), rshd(8), or options(1) man pages to suggest that telnet.hosts affects anything except telnet, or that rsh'ing is affected by anything except /etc/hosts.equiv. It also seems to me very counter-intuitive, and quite inconvenient because the telnet.hosts list is restricted to so few elements. Chris Thompson University of Cambridge Computing Service, Email: cet1(a)ucs.cam.ac.uk New Museums Site, Cambridge CB2 3QG, Phone: +44 1223 334715 United Kingdom.

4 3

Re: Omniback -Reply
by Timothy Cook 22 May '00

22 May '00

We also use OmniBack II to back up our F630 and F760, and I agree that it is inefficient across NFS and does not preserve ACL's. Netapp discusses several alternatives that use NDMP (Workstation Solutions "Quick Restore", Veritas and Legato). Though just as inefficient, an NT backup software would preserve the ACL's, though I don't know whether the restored data would have any unix permissions from a mixed qtree or not. ADSM has also been mentioned by some on this board. We are currently investigating alternative backup software, any recommendations or advice? >>> Philip Thomas <thomas(a)act.sps.mot.com> 05/22/00 11:30am >>> Yes, we use Omniback to backup our 540's & 330's across NFS for the last 5+ years. Of course, it's inefficient across NFS compared to having a "client" on filer like that for Legato and one would loose CIFS acls. -- Begin original message -- > From: "Uros Lampret" <uros.lampret(a)ourspace.si> > Date: Mon, 22 May 2000 09:03:22 +0200 > Subject: Omniback > To: <toasters(a)mathworks.com> > -- End original message -- ---philip thomas

1 0

RE: Interaction of /etc/hosts.equiv and telnet.hosts option
by tim.thompson＠netapp.com 22 May '00

22 May '00

> There's nothing in the hosts.equiv(5), rshd(8), or options(1) man pages to > suggest that telnet.hosts affects anything except telnet, or that rsh'ing If you search for "telnet.hosts" on the NOW site, you'll find several online bugs related to this change in behaviour. The documentation can be found likewise, in an online "Changes to the System Administrator's Guide" for things that didn't make it into the Systems Administrator's Guide. The justification for the change was to provide more consistent access control (and "trusted.hosts" is an alias for "telnet.hosts", to help transition to the new meaning of the option). As Bruce points out, per-protocol access control is desired, and we have someone actively working on that. ...Tim Thompson...Network Appliance, Inc...

1 0

RE: Omniback
by Anders Ljungberg S (QDT) 22 May '00

22 May '00

Hi all! Does omniback support nt clients? If, use one as frontend for NTFS data. Anders Ljungberg Ericsson Business Consulting -----Original Message----- From: Philip Thomas [mailto:thomas@act.sps.mot.com] Sent: Monday, May 22, 2000 5:30 PM To: toasters(a)mathworks.com Subject: Re: Omniback Yes, we use Omniback to backup our 540's & 330's across NFS for the last 5+ years. Of course, it's inefficient across NFS compared to having a "client" on filer like that for Legato and one would loose CIFS acls. -- Begin original message -- > From: "Uros Lampret" <uros.lampret(a)ourspace.si> > Date: Mon, 22 May 2000 09:03:22 +0200 > Subject: Omniback > To: <toasters(a)mathworks.com> > -- End original message -- ---philip thomas

1 0

Omniback
by Uros Lampret 22 May '00

22 May '00

Does any body know how to backup a filer with Omniback. We use Omniback and now testing an F740. Uros

2 1

RE: software distribution with NDMPcopy/SnapMirror over WANs?
by Aaron.Sims＠netapp.com 22 May '00

22 May '00

One thing that SnapMirror can do, that NDMPcopy and rsync won't do is, not show the changes until the entire SnapMirror operation is complete. With NDMPcopy and rsync, as each changed file is copied, the change is seen immediately on the destination, whereas with SnapMirror, the changes are all made at once. Depending on the type of data and the schedule of sync'ing, this may or may not be desirable. Aaron > -----Original Message----- > From: Eyal Traitel (r55789) [mailto:eyal.traitel@motorola.com] > Sent: Monday, May 22, 2000 8:45 AM > To: toasters(a)mathworks.com > Subject: Re: software distribution with NDMPcopy/SnapMirror over WANs? > > > I believe "rsync" utility can help you out.... > > > Eyal. > > > Aaron.Sims(a)netapp.com wrote: > > > > -----Original Message----- > > > From: Thaller Horst (MDCA Villach) > [mailto:Horst.Thaller@infineon.com] > > > Sent: Monday, May 22, > 2000 3:41 AM > > > To: 'toasters(a)mathworks.com' > > > Subject: software distribution with NDMPcopy/SnapMirror over WANs? > > > > > > > > > Who has experiences with software distribution/syncronisation > > > over WANs with > > > NDMPcopy or SnapMirror? > > > > > > We have a software pool at our master site, where we maintain > > > our software > > > (~ 500 GB) and then we distribute the changes to our client sites. > > > Our actual rcp/ssh-based solution has several drawbacks: > > > * No automatic syncronisation > > > * Transfer of whole sw-packages only > > > * Only Unix-packages supported (no NT packages) > > > > > > My idea is to do this job with NDMPcopy or SnapMirror, but I > > > have a lot of > > > questions: > > > Can I export (ro) the mirrored volumes? (NDMP, SnapMirror) > > > > SnapMirror, yes. The target volume of a SnapMirror > operation is always > > read-only. If you make it read-write, another Level 0 dump > is required. > > > > > It´s possible to tunnel NDMP/SnapMirror with ssh? > > > > SnapMirror, no. NDMPcopy, maybe... I never tried. > > > > > Can I mirror more than one site with SnapMirror? > > > > Yes, easily. > > > > > Cascaded distribution? > > > > SnapMirror, no. NDMPcopy, maybe. > > > > > Is there a commercial product? > > > > SnapMirror is a commercial product, contact your sales rep > for more info. > > > > Hope that helps. > > > > Aaron >

1 0

RE: software distribution with NDMPcopy/SnapMirror over WANs?
by Aaron.Sims＠netapp.com 22 May '00

22 May '00

> -----Original Message----- > From: Thaller Horst (MDCA Villach) [mailto:Horst.Thaller@infineon.com] > Sent: Monday, May 22, 2000 3:41 AM > To: 'toasters(a)mathworks.com' > Subject: software distribution with NDMPcopy/SnapMirror over WANs? > > > Who has experiences with software distribution/syncronisation > over WANs with > NDMPcopy or SnapMirror? > > We have a software pool at our master site, where we maintain > our software > (~ 500 GB) and then we distribute the changes to our client sites. > Our actual rcp/ssh-based solution has several drawbacks: > * No automatic syncronisation > * Transfer of whole sw-packages only > * Only Unix-packages supported (no NT packages) > > My idea is to do this job with NDMPcopy or SnapMirror, but I > have a lot of > questions: > Can I export (ro) the mirrored volumes? (NDMP, SnapMirror) SnapMirror, yes. The target volume of a SnapMirror operation is always read-only. If you make it read-write, another Level 0 dump is required. > It´s possible to tunnel NDMP/SnapMirror with ssh? SnapMirror, no. NDMPcopy, maybe... I never tried. > Can I mirror more than one site with SnapMirror? Yes, easily. > Cascaded distribution? SnapMirror, no. NDMPcopy, maybe. > Is there a commercial product? SnapMirror is a commercial product, contact your sales rep for more info. Hope that helps. Aaron

2 1

← Newer
1
...
4
5
6
7
8
9
10
...
15
Older →

Jump to page:

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 2000