TSM & VMWare - Vstorage Api
Hi all, Do you have any material / recomendation on how to config Vstorage Api for Vmware backups, how does it work, what's the impact for datastore (caused by the snapshot failure), or what's should be considered for this? Thanks in advance, Yudi Darmadi http://www.npp-asia.com Sent from XL BlackBerry®
Re: Re: snapdiff advice
Pete Tanenhaus wrote: > > The problem you will run into on Windows is that if the scheduler runs as a > service under the > default system account accessing the shares won't be possible, you will > need to configure > the service to log in under an account which can access the CIFS shares. That's a good point. I'll have to work out with our NetApp admins and Windows admins how they want to handle this. > If the NetApp filer is configured to be a trusted domain member (and the > account the scheduler service > runs under is a domain admin) you should be able to backup the shares > directly via the UNC names. > > If the filer isn't a trusted domain member it is a little more difficult as > you must supply credentials > in order to authenticate the shares with the filer, and as previously > suggested this can be done > with NET USE commands in a pre-schedule command. Except, in my initial manual testing, I still had to do NET USE to get the snapdiff backups to work at all, even when using the UNC name and running the client under a domain account with rights to the share. I didn't have to provide any additional credentials, but I did have to run "NET USE \\filername\volname", or the backup didn't recognize the share as a NetApp volume. Maybe I'm showing my knowledge gaps in the nuances of Windows authn/authz and share access. Thanks for your advice, Pete. (At least I have snapdiff working manually for CIFS. I'm having less luck with NFS...) =Dave -- Hello World.David Bronder - Systems Admin Segmentation Fault ITS-EI, Univ. of Iowa Core dumped, disk trashed, quota filled, soda warm. david-bron...@uiowa.edu
Re: snapdiff advice
The problem you will run into on Windows is that if the scheduler runs as a
service under the
default system account accessing the shares won't be possible, you will
need to configure
the service to log in under an account which can access the CIFS shares.
If the NetApp filer is configured to be a trusted domain member (and the
account the scheduler service
runs under is a domain admin) you should be able to backup the shares
directly via the UNC names.
If the filer isn't a trusted domain member it is a little more difficult as
you must supply credentials
in order to authenticate the shares with the filer, and as previously
suggested this can be done
with NET USE commands in a pre-schedule command.
Hope this helps .
Pete Tanenhaus
Tivoli Storage Manager Client Development
email: tanen...@us.ibm.com
tieline: 320.8778, external: 607.754.4213
"Those who refuse to challenge authority are condemned to conform to it"
|>
| From: |
|>
>--|
|David Bronder
|
>--|
|>
| To:|
|>
>--|
|ADSM-L@vm.marist.edu
|
>--|
|>
| Date: |
|>
>--|
|06/28/2011 06:52 PM
|
>--|
|>
| Subject: |
|>
>--|
|Re: Re: snapdiff advice
|
>--|
|>
| Sent by: |
|>
>--|
|"ADSM: Dist Stor Manager"
|
>--|
Thanks, Bill. I was coming to the conclusion that this would have to be
scripted, at least for the CIFS shares. Do you leave the various shares
mapped/used, or do you "net use /delete" them at the end of your script?
I might experiment with a PRESCHEDULECMD script to do the "net use" bits
and then add the shares to the DOMAIN (if that works -- I'm totally not
a Windows admin). Though I suppose it'd be wiser to script the entire
thing, as you did, so there's only one place to add or remove filespaces.
=Dave
Colwell, William F. wrote:
>
> I can't comment on your error messages, but you asked how I schedule
> snapdiff backups.
>
> The schedule invokes a command on the client. Here is a shortened
> version of the command file.
>
> echo on
> for /f "tokens=2-4 delims=/ " %%a in ('date /t') do (set
date=%%a-%%b-%%c)
> echo %date%
>
> net use
> ... 12 more net use statements ...
>
> dsmc i -snapdiff -optfile=dsm-unix1.opt >>
> c:\backuplogs\xxx\snapdiff-%date%.txt
>
> ... 12 more dsmc commands ...
>
> dsmc i c: -optfile=dsm-unix1.opt >>
> c:\backuplogs\vscan64\local-%date%.txt
>
>
> The last line backs up the local file system.
>
--
Hello World.David Bronder - Systems
Admin
Segmentation Fault ITS-EI, Univ. of
Iowa
Core dumped, disk trashed, quota filled, soda warm.
david-bron...@uiowa.edu
Re: Re: snapdiff advice
Thanks, Bill. I was coming to the conclusion that this would have to be
scripted, at least for the CIFS shares. Do you leave the various shares
mapped/used, or do you "net use /delete" them at the end of your script?
I might experiment with a PRESCHEDULECMD script to do the "net use" bits
and then add the shares to the DOMAIN (if that works -- I'm totally not
a Windows admin). Though I suppose it'd be wiser to script the entire
thing, as you did, so there's only one place to add or remove filespaces.
=Dave
Colwell, William F. wrote:
>
> I can't comment on your error messages, but you asked how I schedule
> snapdiff backups.
>
> The schedule invokes a command on the client. Here is a shortened
> version of the command file.
>
> echo on
> for /f "tokens=2-4 delims=/ " %%a in ('date /t') do (set date=%%a-%%b-%%c)
> echo %date%
>
> net use
> ... 12 more net use statements ...
>
> dsmc i -snapdiff -optfile=dsm-unix1.opt >>
> c:\backuplogs\xxx\snapdiff-%date%.txt
>
> ... 12 more dsmc commands ...
>
> dsmc i c: -optfile=dsm-unix1.opt >>
> c:\backuplogs\vscan64\local-%date%.txt
>
>
> The last line backs up the local file system.
>
--
Hello World.David Bronder - Systems Admin
Segmentation Fault ITS-EI, Univ. of Iowa
Core dumped, disk trashed, quota filled, soda warm. david-bron...@uiowa.edu
Re: snapdiff advice
Thanks, Alain! For the CIFS case, it looks like I was missing the "net use" step; doing that first made the backup work (we already had the NetApp configured with a user with correct capabilities and the password for that user was already set on the TSM client side). I'm still having no luck with snapdiff of NFS from a Linux client. The share is mounted (exported read/write with root access), I'm using the same NetApp account for the API connection (and the password is set at the client), but the backup fails with the same errors as I originally described. I was missing the TIVsm-BAhdw RPM at first, but correcting that made no difference. (My NFS testing is on a filer running 7.3.2, but I'm not worried about Unicode support at this point; I just want to get snapdiff to work at all. Neither the SNAPDIFFNAMEFILTEROFF nor SNAPDIFFONTAPFAP testflags had any effect on the attempts, either.) =Dave Alain Richard wrote: > > We are using snapdiff for almost a year. We use some trick's found in > the forum. > > First, mount your share before starting the backup " net use \\dffdg\fgdgj" > Second, If you talking to a NAS like a Netapp, you need to have http > access For TSM : httpd.admin.enableon, and use command tsm> "set > password -type=filer sdfggd $logname". > > Be sure to have at least tsm6.2.2 client and Ontap 7.3.3 if you want > the Unicode work! > Don't forget to do full scan at least once a month just in case it > misses some files. > > Alain > > -Message d'origine- > De : ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] De la part de > David Bronder > Envoyé : 27 juin 2011 16:24 > À : ADSM-L@VM.MARIST.EDU > Objet : [ADSM-L] snapdiff advice > > Hi folks. > > I'm trying to get snapdiff backups of our NetApp (OnTAP version 8.0.1P5) > working so I can move away from everybody's favorite NDMP backups... > > So far, I'm not having much luck. I don't know whether I'm just Doing > It Wrong (tm) or if something else is going on. In particular, on both > Windows 2008 R2 (6.2.3.0) and RHEL 5.6 (6.2.2.0), I'm getting failures > like the following, depending on the dsmc invocation: > > ANS1670E The file specification is not valid. Specify a valid Network >Appliance or N-Series NFS (AIX, Linux) or CIFS (Windows) volume. > > ANS2831E Incremental by snapshot difference cannot be performed on >'volume-name' as it is not a NetApp NFS or CIFS volume. > > (These are shares at the root of full volumes, not Q-trees. I'm using a > CIFS share for the Windows client, and an NFS share for the Linux client, > with the correct respective permission/security styles. TSM server is > still 5.5, but my understanding is that that should be OK.) > > For those of you who have snapdiff working, could you share any examples > of how you're actually doing it? E.g., your dsmc invocation, how you're > mounting the share (must a Windows share be mapped to a drive letter?), > or anything relevant in the dsm.opt or dsm.sys (other than the requisite > testflags if using an older OnTAP). Or anything else you think is useful > that the documentation left out. > > (Also of interest would be how you're scheduling your snapdiff backups, > and how you have that coexisting with local filesystems on the client > running the snapdiff backups.) > > Thanks, > =Dave > -- Hello World.David Bronder - Systems Admin Segmentation Fault ITS-EI, Univ. of Iowa Core dumped, disk trashed, quota filled, soda warm. david-bron...@uiowa.edu
Re: snapdiff advice
Hi Dave,
I can't comment on your error messages, but you asked how I schedule
snapdiff backups.
The schedule invokes a command on the client. Here is a shortened
version of the command file.
echo on
for /f "tokens=2-4 delims=/ " %%a in ('date /t') do (set
date=%%a-%%b-%%c)
echo %date%
net use
... 12 more net use statements ...
dsmc i -snapdiff -optfile=dsm-unix1.opt >>
c:\backuplogs\xxx\snapdiff-%date%.txt
... 12 more dsmc commands ...
dsmc i c: -optfile=dsm-unix1.opt >>
c:\backuplogs\vscan64\local-%date%.txt
The last line backs up the local file system.
Regards,
Bill Colwell
Draper Lab
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
David Bronder
Sent: Monday, June 27, 2011 4:24 PM
To: ADSM-L@VM.MARIST.EDU
Subject: snapdiff advice
Hi folks.
I'm trying to get snapdiff backups of our NetApp (OnTAP version 8.0.1P5)
working so I can move away from everybody's favorite NDMP backups...
So far, I'm not having much luck. I don't know whether I'm just Doing
It Wrong (tm) or if something else is going on. In particular, on both
Windows 2008 R2 (6.2.3.0) and RHEL 5.6 (6.2.2.0), I'm getting failures
like the following, depending on the dsmc invocation:
ANS1670E The file specification is not valid. Specify a valid Network
Appliance or N-Series NFS (AIX, Linux) or CIFS (Windows)
volume.
ANS2831E Incremental by snapshot difference cannot be performed on
'volume-name' as it is not a NetApp NFS or CIFS volume.
(These are shares at the root of full volumes, not Q-trees. I'm using a
CIFS share for the Windows client, and an NFS share for the Linux
client,
with the correct respective permission/security styles. TSM server is
still 5.5, but my understanding is that that should be OK.)
For those of you who have snapdiff working, could you share any examples
of how you're actually doing it? E.g., your dsmc invocation, how you're
mounting the share (must a Windows share be mapped to a drive letter?),
or anything relevant in the dsm.opt or dsm.sys (other than the requisite
testflags if using an older OnTAP). Or anything else you think is
useful
that the documentation left out.
(Also of interest would be how you're scheduling your snapdiff backups,
and how you have that coexisting with local filesystems on the client
running the snapdiff backups.)
Thanks,
=Dave
--
Hello World.David Bronder - Systems
Admin
Segmentation Fault ITS-EI, Univ. of
Iowa
Core dumped, disk trashed, quota filled, soda warm.
david-bron...@uiowa.edu
Users of Tivoli Storage Manager Administration Center
Hello, The TSM development team is seeking feedback and input from current users of the Administration Center. If you are interested and willing to provide feedback and participate in some telephone interviews about how you use it, please respond via email to: tvoli...@us.ibm.com For those that respond, please provide the following information: Email and other contact information for how we should contact you. A brief summary of your environment. (Number of servers, how long have you used the product, versions of the product currently in use) A brief description of your use of the administration center. How frequently do you use it (hourly? daily? )? What kinds of tasks you use it for? Thanks for your consideration and input. TSM Development
Will SUSPENDED EXPORTS be affected if TSM Servers are recycled?
Hi All, Quick question I am exporting clients from TSM server A to server B, If I suspend the exports and recycle both TSM server A and Server B the exports should be ok and pickup where they left off once I restart them correct? Thanks
SAS RPO and TSM
The company is developing a SAS RPO, has anyone had any experience using TSM as a solution for its backup requirements. The SAS docs recommend using the supplied utilities (wizard or %omabakup) to minimize down time and basically to backup all components simultaneously to assure a usable restore. Estimates are that once fully operational the data impact would be roughly 5+ tb of which most is EMC SAN. Finally, the SAS consultants that are assisting with the rollout have provided a very high level document requesting the full plus incremental approach. Having little to no knowledge of SAS I thought it best to ask has anyone had an experience backing up SAS using TSM and what is working for them? If it makes a difference it is SAS RPO. TSM Server level 5.5.5.2 (Windows) SAS on AIX All comments appreciated. Rick Adamson Jacksonville, FL.
