ackup set restoration if
encryption is enable for file system backup.?
Sachin C.
Dwight Cook
Sent by: "ADSM:To
Dist Stor ADSM-L@VM.MARIST.EDU
orted no problems with
these drives.
Thanks for the suggestions, we have looked at so many things it is hard
to recall them all, but we may have missed something simple.
Andy Huebner
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Dwight Cook
Sent:
I've noticed in 5.5.1 server (healthy AIX server running virtual tape
library so minimal mount delay, etc...) with 5.4.2 windows clients that if
you are performing a restore where over 1.5M files exist within TSM for a
single drive, it can take 7-10 days to perform the restore. If you have
2.5+Mil
Sorry... didn't see the initial email earlier...
As mentioned by Richard, the details on the TSM server and client platforms
& code levels would be helpful.
I seem to remember older clients only initiating a single thread per mount
point on retrieves (but I could be remembering incorrectly).
Als
Where does your directory management class go?
It could be the case that the storage location used as the destination of
your directory management class is full. That would explain being able to
successfully run commands that aren't tied to the DIRMC.
"The dirmc option specifies the management cl
You could use the move nodedata to move other nodes off that tape or move
the node you want to keep to another tape.
Say you have one node's data you want isolated... mark other filling vols in
that pool readonly and then just use a move nodedata.
If you want to purge just data on that tape for
Your guide to all software support :)
http://www-01.ibm.com/software/support/lifecycle/
Right now I don't see any end of support date for TSM 5.5
http://www-111.ibm.com/software/support/lifecycle/PLCDetail.wss?synkey=V6499
76C05962P29-J757048G77027Y87-T956174D94205J68
-Original Message---
You will probably see that only one node's data resides in the disk pool
still... I've noticed TSM only firing off a single migration process per
node for any given storage pool so even if you set migpr=10 and high=1 low=0
you'll still only see a single migration process if you have data from only
I've seen that but generally once the log has hit 80% (of a 13 GB log) and
TSM is imposing a 2 millisecond delay in transactions in order to try to
keep the log from filling and crashing TSM.
I only have seen it a few times a year across various very active
environments but one thing that seems to
You might be running into what I've seen if you are performing your
reclamation via RECLaim STGpool. If you have many very low utilized tapes
and you do not use an OFFSITERECLAIMLIMIT then it is possible that the
reclamation process will only yield another tape going off site and none
coming back.
Try these:
select distinct node_name as "Node Name ",filespace_name as "Filespace
Name",class_name as "Management Class Name" from backups
where TYPE='FILE' group by node_name,filespace_name,class_name
Node Name Filespace Name Managem
If you want to force it to use the cluster ip for the cluster node, set the
tcpclientaddress in the dsm.opt file used by the cluster node client
acceptor & scheduler... but if those follow the cluster OK, it really
shouldn't matter if they use the cluster address or the address of the
server curren
I always have that issue, I get around it by saving to my local PC and then
opening it from there...
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Timothy Hughes
Sent: Tuesday, May 25, 2010 2:50 PM
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] hav
Honestly Geoff, I think I'd have to tell them to stop smoke'n crack...
In reality, people do what is easiest for them and/or their application.
I'm sure it will be easier on them to make one big ~container~ for all their
~stuff~ rather than spread it out over lots of smaller areas. They might
say
Also have to wonder if maybe the VTL hasn't lost multiple drives in a raid
array and is (possibly) working off of parity or at minimum, rebuilding on a
spare drive though that would generally happen and be done with (not be
seen over and over again) but... if you have a raid array that has lost
e
You just have to remember that the directory under which the files will be
written needs to be mounted up to your local TSM server.
So filesystems /dev_class_file exists on your primary TSM server but is
either an NFS mount to /dev_class_file filesystem on your DR TSM server...
OR filesystem /dev_
Speaking of COM+ there is also the KB934016 which is required
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Lakshminarayanan, Rajesh
Sent: Wednesday, September 15, 2010 8:16 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] VSS error
HI Timo
In general computing you will want to have your production data center with
all your production servers and your remote site data center functioning as
your production fix / test / development / disaster recovery data center.
Wise processing practice is to perform monthly production fix refreshes f
>From a dsmc session, see what "q inclexcl" reveals.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Botelho, Tiago (External)
Sent: Tuesday, September 21, 2010 4:56 AM
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] Exluded file do not work - Can't u
Yes, you still need to define the client scheduler service but it is started
and stopped as required by the CAD.
So typically you'll have three items
dsmcutil inst /name:"TSM Client Scheduler" /node:NODE1 /password:NODE1PSWD
/autostart:no /startnow:no /schedlog:"C:\Program
Files\Tivoli\TSM\baclie
That sounds more like a case of you don't have something in the dsm.opt
correct, like the tcpserveraddress or maybe the routing at the system level.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of
Christian Svensson
Sent: Tuesday, December 14, 20
Did you check to make sure the drive is securely tightened down...
I had a 3494 that the little screws to tighten the drive's slide rails would
be loose and as the gripper would come forward to grab the tape, rather than
the tape going into the gripper, it would push the tape drive back just a
bit.
All I can say is what I've observed...
Question #4) TSM looks to preallocate space on and direct new inbound
traffic to a volume not currently in use within a storage pool. This is why
I try to have as many volumes in a storage pool as I expect concurrent
inbound sessions (within reason). 1 is a
I just got through with a DR test where the only entry in my volhist was the
single dbb entry I wanted it to use to recover with.
Worked like a champ after I figured out that async I/O wasn't activated at
the AIX level... that killed me for hours :-(
I also totally change around my devconfig becaus
I would say restore your secondary machine from a mksysb image of your
initial machine, then MAKE SURE AND ACTIVATE ASYNC I/O because that won't be
restored with the mksysb... then the drm file pretty much works like a
charm.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@
MARIST.EDU] On Behalf Of
Andrew Carlson
Sent: Monday, March 28, 2011 1:05 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Restore of TSM V6 DB on another machine
I guess I should have mentioned this, but unfortunately this is not an
inactive machine waiting for restore.
On Mon, Mar 28, 2011 at
>From the tsm server use
Q file
Then
Show version nametype=fsid
To see the actual file names that TSM is saving your tdp data under...
If varies across tdp products but typically the name used in tsm in a unique
name each backup.
Dwight
-Original Message-
From: ADSM: Dist Stor Manager
Probably the library has it assigned a category other than insert or
scratch.
Use your mtlib command to query the tape...
mtlib -l/dev/lmcp# -qV -VC01861(where # is the proper lmcp1 or lmcp2
etc... that your library is as found via aix command lsdev -Cctape)
Look for the category data... it
I've gone from 5.3.4 up to 5.4.1 and then up to 5.5.0 without issues.
I've also gone from 5.3.4 straight up to 5.5.0 without issues.
I also went to the latest Atape and atldd at the same time.
Atape.10.7.3.0.bin
atldd.6.5.5.0.bin
The installp commands associated with the Atape & atldd
If you only have to keep what is there for 10 years and can allow normal
processing of new data going forward... set up a new domain with grace
periods to fit your needs... for 10 years 3,666 days will work, my example
here is where I had to keep everything for an umlimited period of time.
(put in
That is probably where the event is so far back that it has fallen off the
event history.
Event Record Retention Period: 65 Day(s)
Use query stat to see your current setting.
Additionally, the last thing your client does is send the results to the
server, it is possible that the client
What I've seen in the past, in general, is that inbound work looking for a
place to store data will bind to a defined TSM volume that is not currently
in use.
Now, knowing that volume name is an internal key within TSM you can prevent
excessive TSM DB lock contention by breaking that storage pool u
Data on specific volumes...
How do you know what is on those volumes?
Need more information...
Is this copy pool data?
Is it data for a single node and you know what is on it because of
collocation?
Dwight
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behal
are affected.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Behalf Of
Dwight Cook
Sent: Thursday, July 31, 2008 2:05 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Change retention policy on specific volumes
Data on specific volumes...
How do you know what is on
Might I ask how you have determined it is going to itself rather than the
"next" storage pool? (by the volume assignment of the destination volume?)
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Hart, Charles A
Sent: Monday, August 25, 2008 3:59 P
So you folks have the following windows patches installed:
WindowsServer2003-KB940349-v3-ia64-ENU.exe
WindowsServer2003-KB940349-v3-x86-ENU.exe
WindowsServer2003.WindowsXP-KB940349-v3-x64-ENU.exe
(based on your platforms)
Dwight
-Original Message-
From: ADSM: Dist Stor Manager [mailto
Here is a thought...
Define a management class with unlimited characteristics...
define domain UNLIMITED backret= archret=3
define policy UNLIMITED standard
define mgmt UNLIMITED standard UNLIMITED migdest=spacedisk
define copy UNLIMITED standard UNLIMITED standard t=b dest=gbdp
verex=nol
You could have used the search=yes and the volrange to fire up multiple
label libvol commands... fire off 25 commands each doing a range of 100
tapes.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Andrew Collyer
Sent: Thursday, November 13, 2008 9:
Internally, within TSM there will be DB locks against keys such as ~volume~
and to help avoid lock contention what I like to do is to have as many
volumes in a storage pool as I expect maximum concurrent inbound client
sessions writing to that storage pool. So if you have a BACKUPPOOL that is
to a
Is it lots of little files (I know, silly question with it being a windows
file server).
Also, how long is "over night"?
Is that compressed client data or is it file space data?
Is that a backup or archive?
What is your network? 100 Mb/sec fast Ethernet? Gig Ethernet? Teamed
NIC's?
I'd run archi
arist.edu] On Behalf Of
Dwight Cook
Sent: Thursday, December 11, 2008 1:45 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data
Is it lots of little files (I know, silly question with it being a windows
file server).
Also, how long is "over night"?
Is
I've seen restores of 1+M files take days due to the delays associated with
general system over head (creating directory entries, etc...) and by days I
mean 5-7+.
And so again, I'll mention...
Just because you CAN put a million or more files on a single drive doesn't
mean it's a good idea!
Dwigh
You looking for something like this???
select session_id as "Sesija",cast(client_name as char(15)) as "Node", -
cont> cast(state as char(10)) as "Stanje", -
cont> current_timestamp-start_time as "Proteklo vrijeme", -
cont> cast((cast(bytes_sent as decimal(18,0)) /
cast((current_timestamp-start_t
Well, remember that under windows... upper case and lower case are the same
unless there are multiple files with various combinations of upper and lower
case characters so NTUSER.DAT* = ntuser.dat* if only one ntuser.dat
type file exists. So probably it is more so they are closing a hole (unless
wi
I actually saw a flood of these type messages across multiple systems...
enough to where I raised a red flag and declared a potential virus attack.
Within minutes of sounding the alarm it was discovered that it was actually
the anti-virus team performing a "vulnerability test" of the environments.
Changingretries 50 WOW!
OK, that is just the number of times to retry an open/active/changing
file... your issues are connection issues... that parm won't help you and
honestly, I'd back that way back down to something like 4 or less. Picture
sending a 4 GB file 50 times trying to capture a stati
Is it maybe getting killed because it is waiting on a resource and hitting
your resourcetimeout? Like a tdp backup going straight to tape and no tape
drives available?
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
Loon, EJ van - SPLXM
Sent:
Wanda,
Tell them to always run with the exact same multiplexing and restore
sessions as they did when backing up (match multiplexing and backup
sessions)
In the init.utl file there is also a MAXBACKSESSIONS and a
MAXRESTORESESSIONS etc... your maxsessions and sessions needs to be set at
the
Well, all you have to do is start a bare TSM server and define your library,
check in your DBB tape, halt TSM, and then do your restore.
I like doing it that way rather than using OS commands to mount a tape on a
drive and then classify that as a ~manual library~ in order to do my
restore.
All of m
Also all the TSM DB overhead
How big is your DB and its cache hit % (but I'm thinking you've
probably already looked into that end)
Are you doing journaling? (which I personally don't like)
Or INCRBYDATE (again, which I personally don't like)
What about memoryeff diskcache?
I should a
OK, so I have a file server with 6 volumes each of 2.5 TB's and each with
1-2.5M files on them.
Under Win2008 there is this ~funk~ called UAC such that an "administrative"
id has effective permissions to everything but not really any direct
permission.
That is, if I go into this server and under
identity of a more powerful user.
It is not uncommon for admins here to do what you describe without the
results you describe.
Andy Huebner
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
Dwight Cook
Sent: Thursday, October 10, 2013 2:20 PM
To: A
large backup occur again dump and compare.
Do you also have AMD Admins?
Andy Huebner
-Original Message-
From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of
Dwight Cook
Sent: Thursday, October 10, 2013 10:00 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Win2008
Select distinct node_name,type,filespace_name,hl_name,ll_name from backups
where node_name='' and type=DIR and
filespace_name='\\\c$' and hl_name='\' order by node_name,
type,filespace_name,hl_name,ll_name
It will run a while though...
-Original Message-
From: ADSM: Dist Stor Manager [mai
Getting it to not even look at the rest of the F drive probably isn't going
to happen...
And I'm old and more command line oriented but here it goes...
Depending on how many other directories there are at the root level of F you
could have in your dsm.opt file:
SUBDIR YES
DOMAIN F:
exclude.dir F:\
101 - 155 of 155 matches
Mail list logo
