Pretty neat Tom..
just a little more food for thought since you made reference to
alignment...
some of the ATLs that I'm involved with were built with heavy angle iron
under the feet to spread the load and help keep the frames in proper
alignment in the event a weak spot were to develop in
going from the client to the server... standard routing is used...
to get the client to go out the gige... you need to add a route statement
that forces traffic out the gige when going to the ip address that is your
MVS server
try the man pages for "route" on the sun box... and remember, your rout
see
http://www-1.ibm.com/support/entdocview.wss?uid=swg21050781
for more information...
Disclaimer : I DO NOT ENCOURAGE MESSING WITH THE TSM DB, USE AT YOUR OWN
RISK !!
but this will get you close to what you desire...
Oh, to see all the things you can be VERY DANGEROUS WITH and use with the
You could
identify some known scratch tapes NOW and in the past when that data
existed
roll the environment back to when the deleted data existed
export that filespace
roll the environment back to current
import the filespace
whew...
but will work... (famous last wo
So all "active" versions of files stored on a TSM server is what you can
classify as ~the current restore set~.
They make up all the data as it existed the last time a ~backup~ was
performed (be it "incr" or "sel")
Now, what might be missing ? files that are found in "exclude"
statements, files t
0%
Elapsed processing time: 00:00:05
Thu Jun 12 20:02:57 AST 2003
if the files do not was updated, why TSM backed up all files in the
incremental by date?
Janeth
-Mensaje original-
De: Dwight Cook [mailto:[EMAIL PROTECTED]
Enviado el: Viernes, 13 de Junio de 20
"ADSM: Dist Stor
Manager"
yes, I checked with a query -incremental the last backup date and TSM add
one new backup. If I do not mofified anything file, why TSM backed up
again?
Janeth
-Mensaje original-
De: Dwight Cook [mailto:[EMAIL PROTECTED]
For your diskpool don't worry about the "Pct Logical"
Pct Logical
The logical occupancy of the storage pool as a percentage of the total
occupancy. Logical occupancy is space occupied by client files that
may
or may not be part of an aggregate. A Pct Logical value less than 100%
just simply register an admin but grant it no authority...
the only thing they can do are queries...
and they can't even perform "show" commands (but that depends on what level
of tsm server you are at)
Dwight
Joni Moyer
<[EMAIL PROTECTED]To
>select * from auditocc order by total_mb asc
;-)
"Gill, Geoffrey
L." To: [EMAIL PROTECTED]
Subject: Help Q auditocc or other query
Sent by: "ADSM:
sorry, one of these days I'll learn to read...
here is more along the lines of what you are looking for...
(not much data here because this is the test serve
tsm: TSMSRV08>select
auditocc.node_name,auditocc.total_mb,sum(occupancy.num_files) as file_cnt
from auditocc,occupancy where auditocc.n
Kind of off topic but it is protection of TSM servers...
We are still running with all SCSI attached 3590's both B1A's and E1A's
nice thing is you can use mtlib to mount a tape, then cut a "mksysb" to
that drive & tape
Two TSM environments require capacity increases and I'm wondering if that
occ
generally during DR you go back to like platforrms
ya know though... since the other box is a "new" server, give it a try...
on like systems, I've performed tsm db backups to flat files, ftp'ed them
over to a different tsm server, perormed a restore db and things ran just
fine
can't say what luc
I'm missing out on all the cool options by remaining at 4.2 :-(
"Rushforth, Tim"
<[EMAIL PROTECTED]To: [EMAIL PROTECTED]
PEG.CA> cc:
Sent by: "ADSM: Subject: Re: exp
with import you can't merge client data...
so when those boxes are down to just the last few archives and you want to
take those out of the test tsm env & move over to the prod tsm env you can
do one of...
1) import the node with "replacedef=no" which will rename the imported
filespaces to somethin
Try your command and specify a source category... -s and if it says
manually ejected that should be FFFA
OH, and if you want to remove it from reference in the library manager...
use target -t FFFB to purge the volume information...
Dwight
Mike Cantrell
/usr/bin/mtlib -l/dev/$LMCP -f/dev/$RMT -m -V$VOLSER
where $LMCP is your library, $RMT is the drive you wish to use and $VOLSER
is your tape volser...
so to mount volume ABC123 on drive rmt1 in atl lcmp1 use
mtlib -l/dev/lmcp1 -f/dev/rmt1 -m -VABC123
then -d is dismount...
it is always nice to
Rather than using "incremental" processing to push that data to the tsm
server, use "archive".
the exclude statements for incremental processing won't impact the archive
processing.
I tend to use incremental processing for everything OTHER than specific
application data, then have the application
but to get past this little quirk...
OK, you sould be able to start TSM as if it were a totally new
environment...
do that and since your device class you used to backup the db was a "file",
redefine that ~file_device_class~ in your bare/fresh tsm, halt tsm, and
then perform your recovery again.
Sh
All I can say is that the dsmfmt simply writes "Eric" over and over and
over again...
If things grind to a halt, it is probably, simply due to the I/O to the
drive/raid array/filesystem
Dwight
"Ochs, Duane"
<[EMAIL PROTECTED]To: [EMAIL
IBM Tivoli Storage Manager
Command Line Administrative Interface - Version 5, Release 2, Level 2.0
(c) Copyright by IBM Corporation and other(s) 1990, 2003. All Rights
Reserved.
working OK on AIX 5.2 with the 5.2.2.0 admin client
Dwight E. Cook
Systems Management Integration Professional, Ad
and things such as an admin sched of "BEGIN EVENTLOGGING TIVOLI" to run
every 24 hours can be very handy...
(helps recover from any connectivity problems with your TEC server)
Dwight E. Cook
Systems Management Integration Professional, Advanced
Integrated Storage Management
TSM Administration
(918
If folks have lots of archives... look into the "convert archive" if you
look at my output below, I bet you can tell on what day we did a "convert
archive" to all our nodes but we do have a lot of archives... more
archives than backups
tsm: TSMSRV5>select sum(BACKUP_MB)/1024 as "Bkup GB",sum(
If people notice what is being referred to as "tsm going to sleep" you
might want to check your internal TSM lock count.
The debug command "show locks" will list out all the internal locks,
generally this will be less than 25 even on the most busy system.
I've noticed that during exports I've seen
I've seen it now and again, on Solaris clients.
In the dsmerror.log file I see...
07/07/05 08:31:54 Trying port number 1502
07/07/05 08:31:54 Trying port number 1503
07/07/05 08:31:54 Trying port number 1504
07/07/05 08:31:54 Trying port number 1505
07/07/05 08:31:54 Trying port number 15
Also remember that on PENDING events that end up as MISSED because the TSM
server can't contact node _ on (or is it at) IP address
will trigger one of those messages right off the bat at the beginning of
its schedule window.
Depending on your amount of automation and monitoring
tsm client is a sun solaris 8 running tsm 4.2.2.1 with tdp/r3 3.2.0.10
(yes, I'm behind on a few things... )
Initially running with
MAX_BACK_SESSIONS 15
BRBACKUPMGTCLASS A35_PR5_A A35_PR5_B A35_PR5_C A35_PR5_D
I saw session 1 use mgmtclass A35_PR5_A
session 2 use mgmtclass A35_PR5_B
Morning Geoff,
-pitdate= & -pittime=should be the two options to get you where
you want to be.
The only thing I can think of that would result in more allocated space
being restored than the file system can handle would be if there was a lot
of HSM migrated data from the file system and
You might try keeping only one scheduler service running under node name
BOCOVS2 and just have two schedules for it.
One will be the standard "incr", the other being a "command" and make that
command be "incr -virtualnode=SYSBOCOVS2"
just a thought
Dwight
Yiannakis
the recovery cell is that ~upper left~ cell that you refer to
the CE tape cell is in that same column but the bottom cell or ~lower left~
you might say...
those are the only two I know of that you can't access for general use
Dwight
"Thomas Rupp,
V
Looks like there is probably some data that is lost on that tape, unless
you have a copy storage pool...
To get around that you will want to take the recl up to 100 to pause
reclamation, then
if you have a copy storage pool,
and this tape is in the primary pool
mark the volume destroy
you can use the undocumented debug command to look at things
use "show archive" with the node's name and the file space /tdpmux, so say
the node is SAPSRV1, try
show archive sapsrv1 /tdpmux > mytemp.out
from a dsmadmc session
Dwight
Rob Hefty
OH, yep, after seeing Wanda's note... it will depend on which TDP agent you
are using.
I'm so use to mainly dealing with TDP/R3, sorry...
the tdp/oracle and others use backups rather than archives to store the
data.
you may run a "q file " to look for the filespace that stands out
as what m
maybe someone is doing a "dsmc -virtualnode=blah" from the address listed
Gerhard Rentschler
<[EMAIL PROTECTED]To: [EMAIL PROTECTED]
UTTGART.DE> cc:
Sent by: "ADSM: Dis
You might try any tape utility that exists on the host system to perform
such a tape copy.
FYI... one thing I tried (which failed) was to use a host system utility
(tapeutil under AIX 4.3) to copy the ~files~ off a TSM DB backup tape to a
flatfile on disk so I could ftp that over to a system,
This might be old news but I just noticed a real neat event while clearing
a cached disk pool on an AIX TSM 5.1.7.0 server...
So I have a disk pool with some 19 volumes for about 550 GB of storage.
Two migrations were running but I fired off an additional six move data's
to occupy all the tape dr
Can anyone share a number (such as a GB/hr figure) from personal experience
for migration process rates using 3592 tape drives ?
at native 40 MB/sec, that is 144,000 MB/hr or 140 GB/hr and I'd expect to
see about 70% of that, or 98 GB/hr for a migration process (of data already
compressed by the c
Some old fashion ways would be to
A) export the backups for that file space, then you could import with
relative dates down the road...
B) rename the filespace BUT the inactive versions would still expire based
on "retain extra versions" so that might not buy you much...
but you could re
Personally, if you are really going to keep something forever...
to reduce the buildup in your TSM data base tracking all that stuff...
I'd back them up, export the node to tapes (to keep forever)
then let the data expire naturally...
you can always import the node later with relative dates
the
hard to say...
I'll speculate though...
now, after doing all the voodoo, did you push an incremental from each node
you desired to increase the retention of ???
there is the internal table "Expiring.Objects" and I ~think~ that as your
client runs its normal incremental and you see all of those
I'm going to make a wild guess that the "CONVERSION STATE" has to do with
the node using the "archive description" table.
Try running a "clean archdir showstats" to see what the
"conversion state" of the node is and that it actually reports either
"true" or "false".
I've been seeing odd errors
could it be that aggregates in the primary pool have gone through
reclamation processing and thus purged expired space out of the aggregate
thus reducing the data in the primary pool but not the copy pool ?
Try looking at the percent logical of each pool, doesn't the percent
logical in
If you have plenty of scratch volumes available you will be better off
if you have a spare server, you can do it with minimal impact to the
existing environment
in the past, I took an old data base backup tape and restored it on another
machine to restore an image of the environment as it
In the old way, with SCSI attached 3590's you only had to either manually
mount your mksysb tape in an attached drive OR use another host defined to
the atl to mount the mksysb tape to a drive attached to the host you were
wanting to restore. (I always have a ~utility server~ defined to all m
tcpclientaddress only sets the initial address for the server to come in
on (last I remember)
beyond that, standard system/network routing takes over.
I'm still at 4.2 & 5.1 (moving to 5.2 in the next couple of months) but we
currently put route statements on the client nodes to ensure the
was my initial concern. Can
anyone confirm exactly what one needs to do to ensure one traverses the
GIGE NIC for both inbound and outbound traffic?
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Behalf Of
Dwight Cook
Sent: Tuesday, March 23, 2004 10:50 AM
To:
we are currently on S70's and 3590-B1A's and backup (across various tsm
servers) between 500 GB and 1.7 TB of "client compressed data" so that is
between 1.5 & 5 TB of client filespace nightly per S70 (6 major ones...)
If you use client compression you will find that the network is still the
b
just go to the console and on the far left pull down select "manual mode"
once in manual mode, the drives should indicate what tape to mount and
where it currently resides
there is also a mount screen that goes up on the console but I generally
ignore it and go off the drives
Dwight E. Cook
S
Also, remember, (if you are running client compression) that the
preallocation performed against the disk will be 100% of how big the file
is on your host system.
ALSO if you are using TDP/R3 and it is greater than tdpr3 v3.2.0.11 it will
clear cache and preallocate at a rate of 110% unless you alt
Look for an NT box that had a session canceled due to a timeout value being
reached...
04/23/2004 23:37:24 ANR0481W Session 834 for node (WinNT)
terminated - client did not respond within 7800
seconds.
(SESSION: 834)
That sucker pinned my
have you tried to audit the volume to get things sort of reset ?
mtlib -l /blah/blah -a -Vxxx
also does a query of the library show anything such as a gripper not
available ?
tape might have been on its way out of the library and the gripper failed
to be able to release it...
or any
in the install directory there might be a "dsmserv.err" file that is the
default error logging file for the tsm server
start by looking in there
once you get the server restarted after any unexpected halt, you should
always check the activity log for any messages that might indicate what
happe
that is an OS type issue...
just create some form of network file system mounted up on your server,
create the device class of type FILE pointing to that area, then use that
device class to do your DB backup to...
Ugh... I'm not the windows type person but say you have a K: disk that is
really
one thought...
for clients that require the monthly 4 year archives...
register alternate nodes by the name of _exp
push any data required for long term retention (in the form of either
backups or archives or both) using the client _exp
export that node
remove from environment...
doesn't tak
Sure...
started a long time ago when directories were bound to the longest
retention management class that existed...
with a 10 year archive management class defined, I was keeping a lot of
stuff WAY to long :-(
made me realize I might not want such management classes existing in the
environme
going to be something wrong with your next pool in one way or another
3494POOL, you should check the status of the library
if AIX, "lsdev -Cctape", look for lmcp#, use in "mtlib -l/dev/lmcp#
-qL" look for anything like ~gripper not available~ or ~paused operational
state~, etc...
mtlib
have you zoned your switch yet ?
Dwight E. Cook
Systems Management Integration Professional, Advanced
Integrated Storage Management
TSM Administration
(918) 925-8045
Beatriz Villegas
nothing wrong with those last three...
as long as you have active management classes HOURLY, EACHWEEK, &
EACHMONTH in the domain to which the node is registered...
UNLESS you have an "exclude.fs" or an "exclude.dir" statement those are
processed initially and (for lack of a better term)
that is the beauty of export tapes
as long as the tape is OK and as long as you have a server version that can
read the export tape (probably created on an earlier version) and all the
other things that go along with media type and devices to read it
you can keep it forever!
I keep sa
occupancy #'s are real time
auditocc #'s only update at "audit license"
Dwight E. Cook
Systems Management Integration Professional, Advanced
Integrated Storage Management
TSM Administration
(918) 925-8045
hope this helps...
also use "mtlib -l /dev/lmcp0 -qV -V 002491" to see what the state of the
volume is...
Volume Data:
volume state.00
logical volume...No
volume class.3592 1/2 inch cartridge tape
volume type..3592 JA Cartridge
volser...B
also are you doing client compression ?
Dwight E. Cook
Systems Management Integration Professional, Advanced
Integrated Storage Management
TSM Administration
(877) 625-4186
Ben Bullock
Personally I'd look into processing that file system with an image backup.
A way to look at that would be "Which can you count to faster, 2,000,000
or 1 ?"
Depending on the data, 250 GB would compress down to 83 GB (if we assume
3/1 client compression) and if the client & server processors
Would greatly appreciate any input/feedback on what to look out for, tips,
hints, etc. in an AIX environment.
Remember, directories are just special files...
Add /* to the end of them and see how that works...
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
Pasta_Fazole
Sent: Tuesday, January 14, 2014 3:27 PM
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] Wh
Just brew a pot of coffee...
I tell folks, just because you "can" do something doesn't mean its a good
idea "to" do something and that much data, that many files in a single
mount point is a "poor choice" (but I deal with it almost daily so you're
not alone).
I haven't been dealing with the Win20
With the "no query restore" you are pushing the workload onto the TSM
server.
The client will still have to transmit the details on all the 7.5M files
up to the TSM server.
And I believe a simple "restore -replace=no -subdir=yes X:\*" will trigger
a ~no query restore~ by default.
OK, interesting, I
Did you run a checkin libvol using search=yes to see if TSM finds the extra
volume?
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
r4mzeso
Sent: Monday, February 24, 2014 7:32 AM
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] ALL Tapes known to TSM
Default status on a checkin is "private" so that might be why it came in
that way.
If you do a "q vol " and nothing is listed, that would indicate it
is not part of a storage pool (but could still be a dbb or dbs or export,
etc...) but those would show in your volume history file...
A "query conten
Have a need to try to thin down the "query nodedata" list of tapes to only
the tapes with active data on them for a node if at all possible.
Just wondering if anyone knows of one.
Dwight
OK, seeing a strange one.
I have a 10 TB DB
Running 6 sessions with multiplexing of 4
Tdp/erp will run at 300 GB/hr/session lan free until only one session
remains. then the through put drops to about 400 MB/hr
So I'm moving a total of 1700 GB/hr until the last 4 multiplexed files into
a single
trategic Outsourcing Delivery
(918) 493-4678
From: Dwight Cook
To: ADSM-L@VM.MARIST.EDU
Date: 03/17/2015 02:21 PM
Subject:TDP/ERP slows to a crawl when only one session remains.
Sent by:"ADSM: Dist Stor Manager"
OK, seeing a strange one.
I have a 10 TB DB
Runnin
Noo!
You will be missed!
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
Prather, Wanda
Sent: Friday, April 03, 2015 4:10 PM
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] So long, and thank you...
This is my last day at ICF, and the fi
You can always use a ~select distinct()~ to view what is within
your current environment.
tsm:>select distinct(status) from events
STATUS
--
Completed
Failed
Future
Missed
Severed
tsm:>
Dwight E. Cook
Technical Services Prof. Sr.
Strategic Outsourcing Delivery
(918) 493
OK, recently IBM created Operations Center but also included ~monitoring &
reporting~ through COGNOS.
Has anyone been taking advantage of that for anything?
I'm looking at setting it up and what I'm hoping it can do is:
Trend each individual client's activity over time (backup bytes, archi
Does anyone have a select statement for client occupancy by management
class (so I don't have to recreate the wheel)?
Dwight E. Cook
Technical Services Prof. Sr.
TSM Delivery Architect
IBM Cloud
(918) 493-4678
torage pools for each management class?
If no - it is very very difficult to calculate occupancy for each stored
object because you must select from backups table.
I don?t have select for do it.
Efim
> 11 апр. 2016 г., в 15:43, Dwight Cook написал(а):
>
> Does anyone have a select state
The library manager has to be the highest level of TSM for the
environment...
Dwight E. Cook
Technical Services Prof. Sr.
TSM Delivery Architect
IBM Cloud
(918) 493-4678
From: Zoltan Forray
To: ADSM-L@VM.MARIST.EDU
Date: 04/19/2016 12:11 PM
Subject:Re: upgrading 6.3 to 7.1.x
Well, all that code is out of date but other than that... make sure
SCHEDMODE PROMPT is in the client's dsm.opt file.
Also if you are running only a scheduler, look at switching over to the
CAD and letting it drive the scheduler.
If your backup network is set to use jumbo frames (MTU 9000) check
So who all out there has TSM servers with DB's in the 3+ TB range that
ingest 50+-ish TB's per night into a container storage pool?
Doesn't matter if it is 7.1 or 8.1 or AIX or Linux
I'm seeing behavior that I would classify as ~odd~ and wondering if others
might be too.
Dwight E. Cook
Seni
There are some patches that have to do with VSS issues which are only
available by request, they are:
Windows 2003 -IA64bit servers
WindowsServer2003-KB887827-v3-ia64-enu.exe
Windows 2003 - 32bit servers
WindowsServer2003-KB887827-v3-x86-enu.exe
Windows 2003 SP1 -IA64 bit servers
WindowsServer20
Look into using EXCLUDE.FS on the filesystem you don't want processed.
Dwight E. Cook
Systems Management Integration Professional, Advanced
Integrated Storage Management
TSM Administration
(405) 253-4397
(877) 625-4186
T/L 349-4361
Hi all,
I'm working with TSM version 5.3.4 on linux (fedora cor
Geoff,
(long time no see... been snowed under but coming up for air)
One thing I've notices is that if you have tapes being ejected AND tapes to
be inserted...
Once you remove all the tapes, you really need to swing the I/O station
door closed so the ATL can realize that all the tapes were in
I remember seeing something like this before... (~like~)
OK, so by your log, the client rolled straight to tape at some point
(probably because the disk pool filled or got full enough to where the
preallocation couldn't be made).
What is the maxnummp for your client?
In the past there was somethin
if your media is portable to your new platform, you should be able to just
export/import your data base...
then modify your library definition to point to the new device
definition...
put in new drive definitions...
and you are back in business...
NOW... for legally required long term data retenti
How many inactive versions exist??? (what are your retention characteristics
and how frequently do the files change)
Also, what is the total number of files stored for that file system???
If you have 7 million files in that file system, that can have ugly
results...
Also, do you have any looping li
OH...
Yea, toss in an
Exclude.compress /.../*.Z
(and remember to bounce the scheduler)
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Richard Rhodes
Sent: Friday, March 13, 2009 7:50 AM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] slow resto
If you have multiple TSM servers, just backup to a flat file and push the db
backup to your other TSM server. (same with hourly incr db backups)
(and please don't go down the ~just use server to server communications with
virtual volumes~ road... I don't like using virtual volumes on another
server
Typically in the AIX TSM server environments (can't speak for the others) it
will use empty scratch volumes in the order they ~became~ empty/scratch be
that from all the data expiring on it, reclamation, or checking into the
library so if you had a "brand new tape" you specifically wanted used, you
How long ago and how active is the old system???
As others have suggested, try bringing up a TSM image from a DB backup prior
to you deleting the filespace... but first, halt all reclamation on the 5.2
system... (to keep tapes as in tact as possible)
Once you get the image up on the other system...
Look at your other processes...
McAfee has an ~on access scan~ and I've seen windows servers have their
backup times more than double because the anti-virus software is scanning
every file being backed up when it is opened by TSM.
Also ck what is running on your TSM server... expiration can put loc
I've not used image backups on windows machines but by looking at the names,
I'm thinking those might be image backups... did you specify OBJTYPE=IMAGE ?
If they are regular files, you could use an admin account with full rights,
that would show them to you.
And you might try adjusting your query
Go look in the C:\adsm.sys directory...
You should find what you are looking for there...
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Shawn Drew
Sent: Friday, April 17, 2009 10:30 AM
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] Windows 2003 Reg
Why not just give them a warning/mandate that if they are working late, they
need to go out under services and disable the TSM client scheduler until
they are done working, then start the scheduler before leaving.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.
ftp://ftp.software.ibm.com/storage/tivoli-storage-management/maintenance/cli
ent/
goes all the way back to 3.7
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
RAYMOND J RAMIREZ RAMIREZ
Sent: Thursday, April 23, 2009 5:49 PM
To: ADSM-L@VM.MARIST.
Can I add a comment...
What if you have some form of file system corruption from anything and the
file appears to have changed, TSM backs up the corrupt version and purges
off the old copy???
If the files are as you say and never change, you won't have any inactive
versions as long as active versio
led, it will move to a monthly image backup. Am I
not making sense in my thinking?? Please straighten me out if not...
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Dwight Cook
Sent: Wednesday, April 29, 2009 3:40 PM
To: ADSM-L@VM.MARIST.EDU
Subject:
It isn't the backup that will kill you, it is the restore...
Trust me... if you have over 1.5 Million files in a mount point, expect
weeks or months to perform a full restore.
Remember, just because you can put 20 million files in a mount point doesn't
mean it is a good idea... discourage that at
Simply activate client encryption.
The data will be encrypted at the client before it is sent to the server;
from that point forward, it will be encrypted no matter where it resides in
TSM server storage / media. If you generate a backup set on portable media
readable to the client, the data will
You are aware that under AIX if any WWN in the path of a drive changes, then
that constitutes an entirely new device as seen by AIX. If you replaced
GBics, it possibly could have caused the tape drive to take on a new rmt as
seen under AIX unless you performed a "rmdev -dl rmt_" on the existing
de
1 - 100 of 155 matches
Mail list logo
