Gill,
Any groups you create on 1 server must also be defined
on the other server(s) in the cluster otherwise, the
cluster will not start between the servers this is if
you're using HACMP as your cluster application. After
defining each volume group on each server in the
cluster you have to "sync"
Whatever mgmt class you want to use should be made the
default mgmt class and then activate the policy set.
Thanks,
Angela
--- Manoel Braz <[EMAIL PROTECTED]> wrote:
> Hi *SMs,
>
> There are some management class in a domain. How can
> I specify a management
>
> class that it will be used by backu
Bert,
Will the ADSM server be part of the cluster? If so,
you'll have to install the server code on all nodes in
the cluster that you'd want to function as well as the
backup server and make those file systems and volume
groups part of the cluster so they'll fail over (they
must be identical in n
Yury US,
Yes, I've done it for IBM internally and externally
and it can be done w/TSM.
Thanks,
Angela
--- Yury Us <[EMAIL PROTECTED]> wrote:
> Hi everybody,
> Currently we use ADSM for back up only, but there
> are some other components,
> so I am trying to get information about those
> components
Does the P Drive exist anymore? I've seen this happen
when a drive was once there but no longer exists and
is part of the include stmts in the dsm.opt file.
Thanks,
Angela
--- "Amini, Mehdi" <[EMAIL PROTECTED]>
wrote:
> When upgrading Terminal Server TSM from 3.7.1.0 to
> 4.1, I got this error
>
Veritas did a comparison to TSM last yr. but it has
numerous errors in reference to what TSM can do that
they say it can't. I've highlighted the major issues
but haven't publicly sent out the entire comparision
as these are consulting svcs that I get paid for.
Thanks,
Angela
--- "PINNI, BALANAND
Unless Tivoli Support still has some copies on CD you
can't officially get it. I wouldn't run TSM 3.1 code
because of all of the problems w/it if I were in your
place. I'd first migrate AIX from 4.1.5 to 4.3 and
TSM to 4.2.X and stabilize. Then I'd migrate to AIX
V5 later. By doing it in a pha
You could send the Core Dump to IBM AIX Support to
analyze to see exactly what is causing the problem.
Thanks,
Angela
--- Demaerel Miguel <[EMAIL PROTECTED]>
wrote:
> Hoi
>
> Server
> AIX 4.3.2.0
> ADSM Server 3.7.2.0
>
> Client
> Solaris 2.8
> ADSM client 3.1.0.8
> ADSM API 3.1.0.8
>
> ADSM Oracl
If you had problems getting resolution from Tivoli
Support personnel all you have to do is ask to speak
to the Duty Manager. Explain your problem to them and
indicate what the person told you, that you need it
resolved, and ask to send it to a higher level of
support. The Duty Manager will do th
Irene,
Check with AIX Support to see if there's a patch
that's needed for TCPIP. I can't remember if it's
4.3.1 or 4.3.3 that this is a noted problem for but it
happened to me and the customer thought it was TSM
when it was AIX.
Thanks,
Angela
--- irene <[EMAIL PROTECTED]> wrote:
> We were runnin
go to the Tivoli web site www.tivoli.com and search on
adsm tuning guide or tuning guide. Support developed
this tuning guide to assist administrators in tuning
and understanding all of the elements involved awhile
back. It's very good and I've used it at IBM
previously. The down side is that t
execute q po and q mgmt. Depending upon what mgmt
class you're using for certain data it could be
different. Which you can find out how mnay versions
are associated with each mgmt class via q po and their
retentions.
Thanks,
Angela
--- Ofer Nachom <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I would
collocation is what will reduce your restore time but
also seems that your system could use some performance
tuning @ 1 GB per 3 hrs.
Thanks,
Angela
--- Ofer Nachom <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> Will I am restoring a big amount of data the restore
> take allot of time
> ( 1 GB per 3 hou
a scheduled backup starts depending upon what time the
backup schedule on the TSM server is defined. If
you're using sched mode prompted and the schedule for
example is defined as a weekd (weekday) schedule @
05:00 then the TSM server will prompt the client to
being a backup @ 05:00 Mon-Fri. as l
27;mainframe' S390 client.
>
> Bill
>
> -Original Message-
> From: ADSM: Dist Stor Manager
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> Angela Hughes
> Sent: Friday, June 15, 2001 2:22 PM
> To: [EMAIL PROTECTED]
> Subject: Re: About TSM on OS/390
>
>
> Pa
the mgmt. class that you've assigned the node to isn't
active because the policy set isn't active. You have
to activate the policy set that coincides with the new
mgmt class that that you've assigned the node to via
"activate po . This info. can be
obtained by q mgmt * and q po *
Thanks,
Angela
Paul,
Yes, TSM can run on OS/390.
Thanks,
Angela
--- Paul CC Lo <[EMAIL PROTECTED]> wrote:
> Hi,
>
> Does anyone knows that TSM Client can run on OS/390.
>
> Thanks!
>
> Paul Lo
> Direct Line: 886-2-27259874
> E-Mail [EMAIL PROTECTED]
> Address: No. 206, Sec. 1, Keelung Road, Taipei,
Taiwan
_
Dave,
Yes, via cron on AIX platforms or a cmd schedule
inside TSM.
Thanks,
Angela
--- "Pearson, Dave" <[EMAIL PROTECTED]> wrote:
> Hi...
> I'm still learning about TSM...
> I'm trying to have a script to run automatically
> after starting the TSM
> server.
> Is this possible?
>
> Thanks for your h
I've also had this happen with earlier versions of
code and had to wind up rebooting the machine. The
answer from support was that the process was hung in
memory and the only fix was to reboot. It did solve
it as well.
Thanks,
Angela
--- Jeff Bach <[EMAIL PROTECTED]> wrote:
> Try kill -11 4853
>
copy your dsmserv.opt, dsm.sys, dsm.opt, inclexcl and
device config file if they aren't defined in the
dsmserv.opt file.
Exporting the server will get all of the parameters
you listed and will not need to be recreated.
Thanks,
Angela
--- Jeff Bach <[EMAIL PROTECTED]> wrote:
> Create a file device
Sounds like it could be that you're spawning more than
1 archive session each time you're performing an
archive via maxsession parameter on the client. Also
depending upon what stgpool you're using for archives
do you have more that 1 resource for that stgpool (ie
does that stgpool contain only 1
also ensuring that you stop and restart the sched
daemon so that the changes take effect.
Thanks,
Angela
--- Daniel E Whicker <[EMAIL PROTECTED]>
wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> Jeannine,
>
> what you're probably looking for is:
> exclude /archive/.../*
> (the /.../ tr
you have to create mt devices as well.
--- "Warren, Matthew James"
<[EMAIL PROTECTED]> wrote:
> Hello TSM'ers,
>
> We are attempting to install TSM 4.1 on a Sun
> Solaris 8 platform.
>
> The Server, License and Device packages install
> fine, and we can start the
> TSM server succesfully.
>
> Our
It depends on how much data was on the old servers and
that now appears on the consolidated server as well as
how much capacity you have assigned for your db. My
experience as a consultant in dealing with this issue
is that there is either a lot of data on the new
server or there isn't a lot of c
Going on memory I don't think Windown 2000 is
supported in ADSM V3.1. I know Windows 95 and
Windows98 are. The EXABYTE library should be defined
as an 8mm tape device in order for it to establish
connectivity.
Thanks,
Angela
--- Diego Rufino
wrote:
> Hi
>
> I need to install ADSM V3.1 over Wind
There is a performance tuning guide off of Tivoli's
web site for ADSM/TSM.
--- James Thompson <[EMAIL PROTECTED]> wrote:
> I have been asked to perform some performance tuning
> on some WinNT, Win2k,
> Sun, AIX systems. We just received a 3583 with 2
> 3580 drives, and I need to
> show some perf
Everyone,
I've seen a lot of questions regarding service dates
and code being out of compliance. Any vendor is going
to want you to upgrade to the latest code when
problems arise because many times the fixes are
incorporated as well as enhancements. The fixes and
enhancements to TSM are always i
Raymond,
Did you grant access to all filesystems or to only a
few? This could be why you're not able to restore
them.
Thanks,
Angela
--- Raymond Chao <[EMAIL PROTECTED]>
wrote:
> Hi,
> I am restoring from one Aix client node to
> another.
> I am able to restore some filespaces and
IBM has a Redbook for DR that goes through the
procedure in depth that can accessed off the Tivoli
web site or www.redbooks.ibm.com.
Thanks,
Angela
--- Ken Sedlacek <[EMAIL PROTECTED]> wrote:
> TSM'ers:
>
> Our Management is finally getting where we should
> have been years ago with
> Disaster Rec
as been that it
> has never worked with prompted.
>
> At 08:46 AM 5/30/2001 -0700, Angela Hughes wrote:
> >In prior versions of TSM it has worked with both
> >schedmode prompted and polling.
> >Thanks,
> >Angela
> >--- "Rushforth, Tim" <[E
True but having worked in many large environments and
for IBM I've found that using prompting with the
TCPServername and TCPServeraddress has been very
reliable.
--- Steven P Roder <[EMAIL PROTECTED]>
wrote:
> > Steven,
> > The only problem with that is if the DNS server
> goes
> > down during you
Jon,
No problem. Just didn't want you to make this MAJOR
mistake as I did the first time and hadn't backed up
the volhist file. This was not a good thing!!!
Thanks,
Angela
--- Jon Milliren <[EMAIL PROTECTED]> wrote:
> Angela Hughes wrote:
> >
> > You mean de
hat's
> where randomization
> really pays off. (Ok, GigE is pretty darned cheap
> these days, and getting
> cheaper, but I don't know of anybody with much GigE
> to the desktop.)
>
> Thanks,
> Alex
>
> -Original Message-
> From: Angela Hughes [mai
Jeff,
This is where a good working relationship w/your CE
comes in handy because they share info. about new
releases of microcode and there outcomes. What I
usually do is keep all microcode up to date and
schedule these upgrades w/my CE as part of the Service
Level Agreement (SLA) if your company
Steven,
Sorry for the confusion I should been more direct. Set
servername cmd is done on the TSM server and the
TCPServername on the clients must match with the
server name set on the TSM server.
Thanks,
Angela
--- Steven P Roder <[EMAIL PROTECTED]>
wrote:
> This information is incorrect. The "se
A work around would be to put a new entry for the new
server name in the dsm.sys file for Unix clients and
dsm.opt for Windows. Previously, I've done testing
for different reasons and as long as there is an entry
in the config file and the node is registered on the
new server there isn't a proble
The command is "set servername (new servername)" The
key is that whatever you change it to must be the same
in the clients dsm.opt (for Windows) and dsm.sys files
(for Unix) otherwise you'll get the error invalid
TCP/IP servername. So you might have to change the
client config files as well.
Tha
Steven,
The only problem with that is if the DNS server goes
down during your backup window then your backups will
fail.
Thanks,
Angela
--- Steven P Roder <[EMAIL PROTECTED]>
wrote:
> Actually, this is made so much easier if it is
> turned into a DNS issue,
> rather than editing dsm.opt/dsm.sys fi
Jon,
You mean del volhist type=db otherwise you delete the
history of all the volumes.
Thanks,
Angela
--- Jon Milliren <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I was wondering that if you performed a del
> volhist -1 daily, would the
> db volume offsite be usable (in a dr situation),
> eve
In prior versions of TSM it has worked with both
schedmode prompted and polling.
Thanks,
Angela
--- "Rushforth, Tim" <[EMAIL PROTECTED]>
wrote:
> The "SET RANDOMIZE" is only used for schedmode
> polling, not for schedmode
> prompted.
>
> -Original M
I definitely agree with Richard that 1.1 GB of RAM is
inadequate. I've always started with a minimum of 2
GB of RAM. In large enviroments such as @ IBM we
rebooted our TSM servers every 30 days via cron.
Thanks,
Angela
--- Richard Sims <[EMAIL PROTECTED]> wrote:
> >But I did have another inquiry
Also execute "q license" to see exactly what license
TSM is complaining about.
Thanks,
Angela
--- Jeff Bach <[EMAIL PROTECTED]> wrote:
> I would try this first " REGISTER LICENSE file=*lic*
> "
>
> Jeff Bach
> Home Office Open Systems Engineering
> Wal-Mart Stores, Inc.
>
> WAL-MART CONFIDENTIAL
>
This problem can also be eliminated by performing a
full DB backup daily which I've always done and if the
environment is very large with a lot of activity you
can perform incremental DB backups with the DB trigger
option throughout the day.
Thanks,
Angela
--- David Longo <[EMAIL PROTECTED]> wrot
I'd move off of 3.1 ASAP as well due to performance
problems with the code for 3590 tape devices, not
supported by IBM anymore, restore problems (either
restoring more data than requested or not showing the
data backed up as available), DB problem, and the list
goes on and on not to mention the ad
Todd,
Have you tried the MAC client guide that provides
recommendations for includes and excludes? What error
are you getting about the scheduler not working?
Thanks,
Angela
--- Todd Zenker <[EMAIL PROTECTED]>
wrote:
> Does anybody have a default preferences for the Mac.
> I also cannot
> get th
The reason to reduce randomization to 0 is for large
scale environments where schedmode is set to prompted
and you want to limit the amount of time it takes for
the backups to complete. If you have the resources
such as CPU, memory, disk, network band width, and
tape devices this is not an issue
46 matches
Mail list logo
