have any sql to do such ?
Thanks, anyway !
Zoltan Forray
Virginia Commonwealth University - University Computing Center
e-mail: [EMAIL PROTECTED] - voice: 804-828-4807
Cook, Dwight E (SAIC
We have about 3,000 K tapes in service currently within two atl's (one with
16 drives the other with 8)
One environment takes in between 700 1,300 GB a day (ok, average about 1
TB) of compressed client data
the other isn't as heavily used (yet) and takes in about 200 GB daily
In the past 6
You will find that command line objects are specific at the file system
level...
and since the restore command gives you the ability to ~rename~
the file (aka restore to a different location)
you can only process one item (at most one file system) at a time
restore
check out on www.adsm.org and look for messages along the lines of subject:
Clean archdir commands and context
you are probably running into that...
and if you are seeing any
ANR0104E imarupd.c(548): Error 2 deleting row from table
Expiring.Objects.
there is a new TSM Serv for aix TSM
Double check the status of your DB !
I glitched the other day in that my log was doing odd things like going to
100.1% utilized, then would zero back out
(oh, check to see if you have logmode normal or rollforward ! )
with logmode normal you should rarely see it get much over a few %
utilized...
uhmmm I believe you need to set your target server up to itself...
the device class has to reference a server definition just like on
the server the export was performed on...
look through www.adsm.org for mail messages from [EMAIL PROTECTED] on virtual
volume or server to server
I gave a
How many different management classes are defined for use within your TDP/R3
configuration file ?
(are all the management classes the same one ? )
For example, we have a different management class used for the logs because
they are more important than the db files themselves... that is,
generally, normal data expiration allows for the tapes to ~roll~ scratch
again.
files will expire based on associated management class copy group settings.
(and having expire inventory run, expireinterval of 0 says don't
automatically run expiration and you need an admin schedule to run
Do as Kelly suggests...
When we moved a server out of one 3494 with 3590 B1A drives into a different
3494 with 3590 E1A drives,
basically the same sort of action...
we just
set up the new library
set the next pool of all our diskpools to nothing...
took the old storage pools and renamed them to
for really long term storage...
if you have any management class with long term retention you get screwed by
all your directory info being stuck in it and I've learned the error in my
ways with (10yrarch management classes ! )
Here is a best bet !
register a node with name blah_exp
have the
ANR2121W ATTENTION: More than megabytes changed MB of the database has
changed and the last database backup was more than hours name hours ago. Use
the BACKUP DB command to provide for database recovery.
ANR0362W Database usage exceeds utilization percentage % of its assigned
capacity.
are two
FYI #3 won't do anything to the data...
I talked with someone that had relabeled (by mistake) some tapes
(with valid data) on new 3590 E1A drives
they then put the labels back on with the 3590 B1A drives and experienced
only limited data loss. I believe just the first file on the tape(s).
Long ago in a 3.x version there was a (client) bug that went something
like...
there were 4 bytes of control info, the last byte indicated if a file was
compresses, routines always assumed the 4 bytes would be together within an
inbound record... when the randomness of data actually split the 4
what you will find is that dsmc doesn't like to process or
what you will need to do is use eval such as
eval dsmc incr -pass=blah /somepath/sched.log
or
eval nohup dsmc incr -pass=blah /somepath/sched.log
that should work for you...
later,
Dwight
-Original
Yep, works just fine...
need to make sure it is the same size as the others...
and all you have to do on the one you want to get rid of is delete logvol
blah
one does get paranoid the first few times doing a delete dbvol or delete
logvol ;-)
later,
D
-Original Message-
From: Kai
Well, right off I'd say
q auditocc
now besides that use
select sum (num_files), sum (physical_mb), sum (logical_mb), node_name from
adsm.occupancy group by node_name
Dwight
Unnamed[1]: 46132
Unnamed[2]: 1554.47
Unnamed[3]: 1523.37
NODE_NAME: node1
Unnamed[1]: 10992
Unnamed[2]: 1317.63
Well, I can fill you in on 3590 drives and that might be close to LTO's
for each drive on the chain, add half of the previous drives thruput
with 3590-B1A's a single drive on a scsi card all by itself is rated at 9
MB/sec
add another one on the chain, you will gain another 4.5 MB/sec (9/2) or
OK, just an off the wall, slightly related observation.
On an NT Domino server just using normal TSM archive, I hit a Dr. Watson
error and was dumped back to a C:\ prompt from my dsmc session.
What my problem was I was performing an archive, using a management
class that didn't have an
Yep, just define a devclass like and use it... (I set the est/max capacity
to the size of the filespace)
tsm: q dev flatfile f=d
Device Class Name: FLATFILE
Device Access Strategy: Sequential
Storage Pool Count: 0
Device Type: FILE
I would use backups (nightly incrementals) to cover op system, application
code (the installed oracle code, etc...)
and I would use archives to save all the oracle .dbf files...
This way (as you said) all the open/active oracle files are excluded from
nightly incremental processing.
OH, the TDP
Are you in NORMAL mode or ROLLFORWARD ?
I run NORMAL mode and I've noticed my 2 GB log fill now again (very rare)
but after a few failed actions (because the log was full) the hang breaks
free and all is OK again. (normally I never see the log get much over 1 or
2 %)
Never had the system die
AIX 4.3.3 TSM 4.1.4.0
In our test environment I had forgot to set the db copy and log copy file
system to mount automatically upon system boot.
Well, in testing something today I performed a reboot of this box.
When the TSM server started it issued absolutely NO message what so ever
about its db
I would look into other activity currently going on within TSM at that
time...
I see things like that when (say) expiration is running at the same time a
client is backing up...
another time might be when reclamation is running and a client is
backing up
and/or some combination of such
3590E normal tapes long tapes ? ? ?
are you referring to J tapes K tapes ? (that is all I know of...)
J tapes are 3590-B1A tapes which may be written to by 3590-E1A tape drives
(a reformat is performed)
and K tapes are the 3590-E1A tapes from the start...
If this is the case, yes, you can have
Bach
Home Office Open Systems Engineering
Wal-Mart Stores, Inc.
WAL-MART CONFIDENTIAL
-Original Message-
From: Cook, Dwight E (SAIC) [SMTP:[EMAIL PROTECTED]]
Sent: Wednesday, December 12, 2001 7:53 AM
To: [EMAIL PROTECTED]
Subject:Re
Have you ever had a finger check in an admin session and when you ment to
issue a q pr you only hit pr ?
PREPARE (Create a Recovery Plan File)
I have some environments that I've done that and it has put some hook into
the system where those environments have the DRM in use flag and I haven't
Na
(for the most part) it doesn't matter what pool data resides in
you could create two new tape pools, point things where they will be used
have absolutely nothing point at the old tape pool
and all would function just fine.
eventually most of the data would expire off the tapes in the
The big thing behind the point system is that you buy a bunch of Tivoli
points.
Now, you have them all sucked up in TSM, TWS (workload scheduler) and
software distribution.
You now expand your data center but a dozen machines but you've used up all
your points and you don't want to buy any more.
it is your versions data exists says keep only two copies...
Works like this...
Versions Data Exists : how many copies to keep as long as the file exists on
the client
only keep inactive versions for up to Retain Extra number of days
Versions Data Deleted : upon deletion of the file from
Title: Message
Well, just remember that if you want to cancel either a
session or a process, you can't have any commas in the number so you can't just
do a query cut the number out.
Here is an example of how I deal with process
numbers
#!/bin/kshfor PROCNUM in $(dsmadmc -id=someid
used for image backups ;-)
for backing up raw volumes (or any logical volumes...)
Say you have a file system with only 10 GB but it has a million files in
it... the overhead in backing up a million files is BAD but you could
perform an image backup of the logical volume on which the file system
do you have your dsmsnmp agent running ? ? ? (found in the server install
directory)
ps -ef | grep snmp
root 8072 28022 0 Sep 18 - 0:00 ./dsmsnmp
root 17118 26478 0 Nov 26 - 0:00 ./dsmsnmp
zdec23 23154 19598 1 09:27:06 pts/1 0:00 grep snmp
root 26478 8072
It is a pain (sort of) but do this...
Empty all disk volumes
Perform queries to ensure you know what disk volumes are in what storage
pools...
Delete all disk storage volumes
Halt TSM
Perform
dsmserv auditdb diskstorage fix=yes
that will clear any hanging pointers to that volume
Start TSM
IF your routing on the ATL's pc isn't right, you can see a message such as
that.
(that is if you are using tcp/ip communication method)
or if that specific host for some reason got wiped from your list of LAN
hosts
(that is if you are using tcp/ip communication method)
If you are using the RS232
-Original Message-
From: Cook, Dwight E (SAIC) [SMTP:[EMAIL PROTECTED]]
Sent: Tuesday, November 20, 2001 8:20 AM
To: [EMAIL PROTECTED]
Subject:Re: One server, two ip-addresses
We have one that uses three (3).
Two fast ethernet one
We have one that uses three (3).
Two fast ethernet one GB.
You have to remember that your standard TCP/IP routing is what determines
how these boxes talk !
Client traffic will go in which ever interface you point to and will return
out to the client based on standard routing in the system (but
Look at the format of the device class for the drives.
If you let it default to DRIVE it will use all available hardware
compression...
I always tell clients to use compression and I also have the drives set to
use compression.
tsm: TSMSRV10q dev f=d
Device Class Name: 3590E1A
I think you are getting the wrong idea on delete definitions !
Delete them from AIX, not TSM !
Just stop TSM
sure go ahead delete your rmt# lmcp# definitions.
I think the install will kill the /etc/lmcpd deamon
install your code
run cfgmgr
you will probably notice all devices come back
Uh, I don't remember where I saw it but isn't the old 2.x stuff actually
backint code where the newer 3.x stuff is the TDP and I don't think you
can restore any 2.x stuff with 3.x code
I've told all my users that as we upgrade from backint 2.4.x.6 code up to
tdp 3.2.0.x that none of their
you need your tivoli customer number
go to the getting support page http://www.tivoli.com/support/getting/
then there is a link to fill out a registration form
Dwight
-Original Message-
From: mrkirra2001 [mailto:[EMAIL PROTECTED]]
Sent: Thursday, November 15, 2001 5:45 AM
To: [EMAIL
FYI.
if you are running 3.1.x.x you will probably want to advance to
3.2.0.6
APAR IC31783 is an event that we ran into the other day when building a dev
environment from a prod backup
IC31783 is NOT the type of problem you would want to run into when
trying/having to restore a production
I have an admin schedule run them weekly, that way if at any given time I
happen to see something strange, I know it has happened within the last
week...
Dwight
-Original Message-
From: Robert Ouzen [mailto:[EMAIL PROTECTED]]
Sent: Sunday, November 11, 2001 3:01 AM
To: [EMAIL PROTECTED]
very easy
just set up the new environment to look like the old
restore the tsm data base to the new environment
stick all the tapes in the new atl
run inventory
run audit library
things should be fine...
look out on www.adsm.org in the archives for this list for mail messages
anyone else notice that with the server
Storage Management Server for AIX-RS/6000 - Version 4, Release 1, Level 4.0
even with log mode normal, the log doesn't clear for the ENTIRE duration of
expiration...
When expiration runs for 6-8+ hours a day and clients initiate a half
million sessions
use
mtlib -l/dev/lmcp# -C -s -tFF10 -V__
# is the proper number for your control point
the -s is the source category (use -qV -V to find out what its current
category is)
the -V is the volser
the FF10 is EJECT
-Original Message-
From: Shawn Bierman [mailto:[EMAIL
did you shut down HSM unmount the file system(s) which HSM runs against ?
sounds like the HSM code didn't install because it was running.
Dwight
-Original Message-
From: Jason Jackson [mailto:[EMAIL PROTECTED]]
Sent: Monday, October 22, 2001 11:02 AM
To: [EMAIL PROTECTED]
Subject:
or...
select count(volume_name) FROM LIBVolumes where libvolumes.status='Scratch'
-Original Message-
From: Lawrence Clark [mailto:[EMAIL PROTECTED]]
Sent: Thursday, October 18, 2001 7:28 AM
To: [EMAIL PROTECTED]
Subject: Re: How many scratch volumes in library?
SELECT
OK...
The new IBM Z series processors are being pushed as consolidation servers.
rum VM with a whole bunch of virtual LINUX servers...
The only TSM server for VM is functionally at 3.1.2
VM LINUX client is at 4.2 ( I don't know if 4.2 clients would talk to 3.1.2
server ?)
I do see
just add the
-pass=blah
to your dsmc incr and it shouldn't ask for the id...
dsmc incr -pass=blah somedir
now this would be with passwordaccess prompt which is what we run our unix
clients with...
Dwight
-Original Message-
From: Sean McNamara [mailto:[EMAIL PROTECTED]]
Without changing anything we went from 40 GB/hr (compressed data across fast
ethernet)
to 56 GB/hr (compressed data across Gb, aprox 224 GB/hr client SAP data base
filespace)
with an 7017-S70, that took the processor up to around 90+%
Dwight
-Original Message-
From: Leijnse, Finn F
correct
compression yes|no goes in the dsm.sys
compressalways yes|no goes in the dsm.opt
-Original Message-
From: Joshua S. Bassi [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, October 02, 2001 8:52 AM
To: [EMAIL PROTECTED]
Subject: Re: compressalways
Offhand I think it
you can set a scheduled event for that box that will run a command
dsmc restore -pitdate=mm/dd/ /etc/passwd
using a date (for mm/dd/) from which you know the password file is
good...
-Original Message-
From: Thida Chin [mailto:[EMAIL PROTECTED]]
Sent: Monday, October
Under the Unix environment there is the date command
zdec23@tsmutl01/home/zdec23/bin date +%a
Tue
zdec23@tsmutl01/home/zdec23/bin date +%A
Tuesday
zdec23@tsmutl01/home/zdec23/bin
I use that a lot in scripts...
Dwight
-Original Message-
From: Alan Davenport [mailto:[EMAIL PROTECTED]]
On the first part, nope, not that I've found...
if you elect to go directly to tape, you want to make sure you have enough
tape drives available when it starts...
This is why I like to go to a diskpool first (even though we are about to
roll out our first direct to tape environment)
On the BKI,
You aren't alone !
If you are like here, things seem to be code blue and/but I believe code
blue has been out there for a year.
(this is just some of the stuff I've heard)
You could probably use your sched.log to get a list of what was backed up,
then use it to build delete statements/commands
Well, in the following environment
Session established with server TSMSRV10: Solaris 2.6
Server Version 4, Release 1, Level 4.0
Server date/time: 09/17/01 14:27:09 Last access: 09/17/01 14:25:39
I've found that a show archive nodename filespacename of a node that
was renamed will kill
check http://www-5.ibm.com/de/entwicklung/tdp/matrix.html
-Original Message-
From: Walter Ridderhof [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 12, 2001 3:06 AM
To: [EMAIL PROTECTED]
Subject: TDP for SAP level ?
Fellow TSM'rs,
What level of TDP for SAP should I install on
With the 3494, if purchased and installed with RS232 host communication
method, you will need an MES to upgrade to TCP/IP communication method.
Need to make sure you have the drivers for the tape drives library...
see ftp site index.storsys.ibm.com in the /devdrvr directory or somewhere
down
Uhmmm I don't think you would want all that !
Could end up in an infinate loop ! ! !
what if you did a q act begind=-7 endd=today
if all the output were put in the log, you have just increased your activity
log (space occupied in your db) by the last 7 days. If you only keep 7
days, you've just
Just a FYI on something real strange we saw here...
Say you have CLIENTA and you have an archive /d105/oracle/blah.cntl
Now over on CLIENTB you set up a SErver entry to pretend to be CLIENTA and
issue a retrieve like...
dsmc retrieve -serv=myserv_clienta -pass=mypass
register admin some_admin_name some_admin_name_pswd
grant auth some_admin_name class=node auth=owner node=node_name
Dwight
-Original Message-
From: Diana Noble [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 06, 2001 9:31 AM
To: [EMAIL PROTECTED]
Subject: Client
Denier [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 06, 2001 10:45 AM
To: [EMAIL PROTECTED]
Subject: Re: IBM 3494
Quoting Cook, Dwight E [EMAIL PROTECTED]:
Are you on a ship ?
His telephone area code indicates that he is in California. I have
seen pictures of household water heaters
Are you on a ship ?
If this is to mainly keep the equipment from walking off (like a 3494 would)
I would just do the following...
Remember you have to be real careful not to screw up the alignment when you
attempt to bolt it down
I would pull a caster to get the bolt pattern for it on the atl
try performing an initial command line session from the nt server console
you will need to do something to force a connection, like a q sched
I believe you will then be forced to enter the initial assigned password set
for the client when registered on the tsm server, after that I believe the
If they doubt this, then they basically doubt all networking...
If someone wants verification, tell them to retrieve/restore the data to a
different named file and compare the two.
Dwight
-Original Message-
From: Matthew Large [mailto:[EMAIL PROTECTED]]
Sent: Thursday, August 23, 2001
Now I don't know about mvs anymore but under aix there has to be the dsmsnmp
subagent running...
there might be another required task started under mvs also...
might look for the mention of a/the subagent in the mvs tsm info...
Dwight
-Original Message-
From: Thierry ITTY [mailto:[EMAIL
I have both in use...
with JFS file systems you get the benefit of journaling...
with raw you get the benefit of having the entire device available for use
reduction of the overhead of journaling
and with raw you don't have to spend time doing dsmfmt against them...
Dwight
-Original
Does anyone currently use Sun Solaris 2.6, TSM 4.1, TDP/R3 3.2 and process
data base files greater than 2 GB ?
I have an environment with Sol 2.6, TSM 3.7 backint 2.4.25.6 (yes, so old
it has gray hair) and it doesn't like files over 2GB.
I'm seeing that
BKI0005I: Start of
Many years back I posted a note along the lines of a poor mans guide to
fixing a snapped 3590
I believe #2 pencil pair of round nose safety scissors was all that was
required.
I'm sure it is out at www.adsm.org still ;-)
Dwight
-Original Message-
From: Richard Sims [mailto:[EMAIL
I went out an pulled my note from the archives to provide a FYI...
DWight
Did it really eat the tape or just say pop of a foot or two from the
front ?
You wouldn't believe the amount of confetti that 1 ft. of 3590 can
make inside a drive !
Now, naturally IBM has a bunch
I don't know if this is available to 3.1 or if it came about in 3.7 but
DSMSERV EXTEND LOG (Emergency Log Extension)
Use this command to extend the size of the recovery log when you need
additional log
space to restart the server. After the server is running, you can do the
following:
v Back up
1) well, this is why I set up a domain just for remote servers, they have no
backup copy group and their archive copy group has a default of never
expiring archives. (I'm big on covering my a$$)
2) IT is always nice to easily see just what in the heck is going on
utilization wise with a remote
The device classes probably are not compatible ??? right ???
like 8mm tape DLT tape or 3590 tape ???
you would have to do
move data old_vol_ser stg=stg_pool_using_some_other_devclass
like
move data AAA001 stg=diskpool
hope this helps
later, Dwight
-Original Message-
ONLY DO A SINGLE LOG A SINGLE DB !
I ran into the same thing... I exceeded the max continuation of a EXEC card
!
Just to the single log single db vol
then do formats of the other db vols like normal and go into the running tsm
that you started with the single log single db vol
Where are you seeing the name dbbk.1 ?
if you use a device class that is a file out on a disk somewhere I believe
the names you will see under Unix will be things like 96577202.DBB
if you are going to a tape device, the label should not change.
tsm: TDCDSM03q volhist t=dbb
Date/Time:
change it in the device class with
upd devc blah mountr=#
where # is the number of minutes...
tsm: TSMSRV06q dev 3590devc f=d
Device Class Name: 3590DEVC
Device Access Strategy: Sequential
Storage Pool Count: 3
Device Type: 3590
My problem is the vast number of support groups across client machines...
getting them to do things :-/
Yes, I know looking at the support matrix we currently see things such as a
TSM 4.1 server works with TSM 3.7, 4.1, 4.2 clients.
I'm thinking about the making a statement that I can NOT
You will need to first put on 4.1.0.0 which will require the install CD.
If you have support for your product you can call tivoli support and they
will send you the 4.1 install pack.
I just did my first AIX 3.7 to 4.1.4.0, had to make the pit stop at 4.1.0.0
Dwight
-Original Message-
Not that I know of...
We have some around 32 GB
These are in an ESS storage unit...This server has about 1 TB of disk
storage pools
Volume Name Storage Device EstimatedPct Volume
Pool NameClass Name Capacity Util Status
On the failed client schedules...
You could set a postschedcmd to run a script that does something like a
grep -i fail sched.log | mail -s $(hostname)_bkup_failures
[EMAIL PROTECTED]
Here we have about 500 clients, mainly big unix servers... we have a fleet
of unix admins and we make it
remember that TDP is not going to your server as a backup it is going as
an archive look at where the archives are pointed to in the management
class you've specified in your initSID.utl file...
Dwight
-Original Message-
From: Gisbert Schneider [mailto:[EMAIL PROTECTED]]
Sent: Friday,
schedule
but its a success schedule.
-Original Message-
From: Cook, Dwight E [mailto:[EMAIL PROTECTED]]
Sent: Friday, July 27, 2001 11:11 AM
To: [EMAIL PROTECTED]
Subject: Re: Admin changes in last 2 years?
On the failed client schedules...
You could set a postschedcmd to run a script that does
mtlib -l/dev/lmcp# -qI somefile
-Original Message-
From: James healy [mailto:[EMAIL PROTECTED]]
Sent: Monday, July 23, 2001 8:12 AM
To: [EMAIL PROTECTED]
Subject: TSM tapes out of Sink with Library manager
I was wondering if anyone has encountered this situation.
Environment: two
IBM SCSI Tape Drive, Medium Changer, and Library Device Drivers has the
info...
0190 is probably from your initial install of TSM
019A 019C are your private scratch categories (in HEX)
Now notice how IBM has a problem converting from decimal to hex !
Even though your numbers are just one apart
sorry, that should have been OLDNODENAME not oldfilename...
My brain isn't what is used to be :-(
Dwight
-Original Message-
From: Cook, Dwight E
Sent: Friday, July 20, 2001 12:38 PM
To: 'ADSM: Dist Stor Manager'
Subject: RE: restore/retrieve in different server ** URGENT **
just make
just make sure you have the same level client on the other server, make sure
the other server is also NT 4, then use the virtualnodename OR just edit the
options file and update the node entry (or add it) with the name of the node
whose data you wish to restore.
NOTE don't let the box perform an
Uhmmm guess you didn't have your volumes mirrored via tsm...
if you only had one volume, did your operating system show any errors logged
to it ?
pretty much I'd say you are not going to get out of that one.
could be a good case to justify the cost of additional dasd to mirror your
DB LOG for
nope, actually I never use that one and get rid of it as soon as I see it
(security issue) but if you start the server in the foreground, you are
basically at a console and may register any admin you please.
-Original Message-
From: Joseph Dawes [mailto:[EMAIL PROTECTED]]
Sent: Tuesday,
Try
select volume_name,cast((pct_utilized * est_capacity_mb / 100) as
decimal ) as MB on Tape from adsm.volumes where devclass_name='3590DEVC'
or
select cast(sum(pct_utilized * est_capacity_mb / 100) as decimal )
as MB on Tape from adsm.volumes where devclass_name='3590DEVC'
ie.
Try putting that path in the default path for that win2000 box...
-Original Message-
From: Phillip Guan [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, July 11, 2001 12:53 PM
To: [EMAIL PROTECTED]
Subject: Script does not run - Full path name has space?
Hi,
We have setup a scheduler to
ALSO, forgot this a second ago...
REMEMBER
C:\WINNT\system32\command.com doesn't allow for spaces
C:\WINNT\system32\cmd.exe does allow spaces (when enclosed in double
quotes)
double check under which environment you are running
Dwight
-Original Message-
From: David Longo
Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED]] On Behalf Of
Cook, Dwight E
Sent: Wednesday, July 11, 2001 11:31 AM
To: [EMAIL PROTECTED]
Subject: Re: Script does not run - Full path name has space?
ALSO, forgot this a second ago...
REMEMBER
C:\WINNT\system32\command.com
Well, based on the library, even if you specify checklabel=yes it still
won't mount a volume.
Example, with an IBM3494 with an AIX tsm server will only verify things such
as if the category of the tape within the library manager is correct (and
exists within the atl naturally) it won't physically
Richard, YOU GOT IT ! ! ! !
If you have clients that were registered prior to the level that
automatically created admin id's and those clients still don't have ids...
you get the blah 53.
I just took a box that was causing problems... registered an admin, set the
auth, and tried a client
Can't remember this one being mentioned and I didn't find anything on a
search of adsm.org so I though I'd toss this to the group...
anyone seen this in their client dsmerror.log ?
06/15/01 15:41:32 cuSignOnResp: Server rejected session; result code: 53
AIX 4.3.3 server running TSM 3.7.2
What exactly are you looking for out of your query ?
You might be better off turning accounting on and just look at accounting
records, there will be one for each session a client initiates.
It is real easy to pull accounting records onto a PC and load them into
Excel (or some similar spread
Uh by backup set are you just making reference to a nightly incremental
schedule ?
From an administrative session try a
q sch * * node=MIS_10
for each schedule found perform a
delete assoc blah blahh MIS_10
That will clear any tsm server scheduled activity... (which might be
I believe as long as the DB is in excess of what it was, you can get the
restore to run...
(back with ADSM 2.x that wasn't the case but I believe it might be now, we
always have specs on our DB's)
This is why it is nice to have some automated task to do something simple
like mail you the
Now do you have your environment set to require a backup prior to HSM
migrating data off to TSM ?
If you do and you have in fact lost the HSM data, then you may restore the
client's data from backups (and then migrate it off again) long process but
works.
All HSM data here has to have a valid
This is where you might want to create an administrative id of queryid (or
something similar) with no authority, this will allow it to perform queries
and you won't have to worry about an id being in scripts with its password.
OK, I'll assume you have a Unix box somewhere that has access to the
201 - 300 of 419 matches
Mail list logo
