Use soft mounts so the FS will return an error if it is not available?
On 27/10, Grigori Solonovitch wrote:
I have FILE copy storage pool for NFS-mounted file system from Disaster Site
via 1Gb Ethernet.
Everything is working fine, but I faced problem during Disaster Site testing,
when NFS
and make it very messy to have individual include/excludes on a
host by host basis, but that is in fact what you are asking for.
-km
On 12/08, Huebschman, George J. wrote:
You could use a client option set on the TSM Server with an include for
a specific management class with force=yes.
But, I
It is in the same directory as the Windows client on IBMs software FTP:
ftp://ftp.software.ibm.com/storage/tivoli-storage-management
Check the 5.5 manintenance directories since there probably isn't one for 6.x.
(or is there?)
-km
On 19/05, Michael Green wrote:
On Wed, May 19, 2010 at 2:20 AM, km k...@grogg.org wrote:
Please let me disagree with you. I think it's a wrong approach to
change locale for the entire OS for the sake of backups only.
Besides, I'm not fully aware of consequences of changing the locale
, if the locale does not exist, just install it. Since the en_US
locale is included in the glibc-common RPM try to reinstall or update that
RPM.
-km
On 18/05, Fred Johanson wrote:
For my users with Unicode files on UNIX clients, I recommend this be added to
the startup scripts for the scheduler
On 19/05, Michael Green wrote:
On Tue, May 18, 2010 at 10:49 PM, km k...@grogg.org wrote:
I would advice against overriding the default settings in a script and
instead to set the correct locale for the system. Most system settings in
RHEL based distros are made in the sysconfig directory
with
Blackboard but if you don't then the fastest way back would be a snapshot,
either in LVM or hardware, or a Windows image backup.
Depending on the application then it might even be that you need to do it this
way to get a consistent backup and keep metadata/filedata in sync.
-km
On 27/04, Lee, Gary D
low latency
noopno scheduling at all
You should probably give atleast the noop scheduler a try if you have
the time since it could be beneficial depending on your storage, otherwise
CFQ is an OK general purpose scheduler.
-km
On 26/04, Henrik Vahlstedt wrote:
God morning,
Have
Really nice!
Inotify has been a part of the kernel since 2.6.13 and is supported
on all RHEL 5 releases and atleast SLES 10+ so this should work on any
modern Linux.
-km
On 21/04, Allen S. Rout wrote:
I've got a few clients with rather large filesystems: ~6K dirs, ~2.2M
files, 10 filesystems
As long as you make sure all of your primary stgpools go to the COPY/DRM-pool
there is nothing else special about it.
On 18/04, Mehdi Salehi wrote:
Thanks David,
Does it matter from DRM point-of-view if there are many storage pools?
Yes. (At least for ESX3, but I would guess v4 is the same.)
Use standard Linux excludes for temporary or in-memory filesystems and VMFS,
nothing else special about it.
-km
On 14/04, Timothy Hughes wrote:
yoda, thanks for your reply. I understand what your saying about not
backing up the host
Yes.
On 06/04, Timothy Hughes wrote:
Hello all,
We did a Archive and it only Archived 18.15GB it should have Archived
76.6 GB, I am going to delete the Archive and try again.
Question - By deleting the filespace via command line or GUI that should
get rid of the data that was archived
It works.
Atleast with the 5.5 client, but I don't see any reason why 6.1+ wouldnt.
On 31/03, Lloyd Dieter wrote:
I'm not a Solaris type, so I need to ask...is Open Solaris the same as
Solaris x86?
Trying to find out if the Solaris x86 client will work on Open Solaris, or if
is something
No.
It states in the IBM OEM users guide that:
*Note: SQL Embedded Edition is not supported, SQL Express is supported
except for Platform incremental backups.
This is for DocAve version 5.1.2.0, but the above will probably be true
for all versions.
-km
On 23/02, Fred Johanson wrote:
All,
I
No.
TSM 5.x can only do implicit joins, which narrows it down to left inner
joins I believe. But hey, I'm not a DBA!
On 23/02, Erwin Zavala wrote:
does TSM 5.4 support full left or right outer joins??
Since you have caching turned off now and have emptied the diskpool (?)
fragmentation should not be an issue.
What is the performance when using only a single LTO4 drive, 1/4 of
30mb/s or 30mb/s?
Check the read performance of the disk with dd or something similar.
What speed do other tape
Without an HBA you wont be able to acces your tape drives from the storage
agent.
On 13/01, Botelho, Tiago (External) wrote:
Hello,
I'm planning to configure LANFree data movement.
I take a look on IBM for System Requirements for Storage Agent Server.
They didnt perhaps change the length of volser (i.e. if LTO generation
should be a part of the volume serial or not)? That would explain not
finding any already checked in volumes.
On 02/01, ashish sharma wrote:
Hello Howard,
Yes i received a lot of messages cant find tapes in slot. I just
No.
On 19/11, Bob Levad wrote:
TSM Version 5.5.2 on Windows
I have set up two active data pools (one physical-LTO4 and one virtual-LTO2)
for a group of nodes. When a restore is requested by the client, the active
data on the physical tape is usually mounted. Since I have set up
On 14/10, Erwin Zavala wrote:
Subject: how can I find out when a volume on a stgpool was last written to?
q vol volname f=D
Check out HELP QUERY VOLUME for syntax.
Last written date is only recorded for volumes in sequential storage pools.
-km
but for the server that will allow me to see what the server is spending
time on?
I've been looking in the problem determination guide but cant find any
flag similar to instrument_detail except for 'instr' which isnt documented
and only gives me DB LATCHes.
-km
TSM some more of those gigs and doing some instrumentation for
tonights backups.
-km
On 12/10, Lindsay Morris wrote:
I once saw this when we had set BUFPOOLSIZE way too high.Giving the TSM
server lots of memory starved the OS for memory; things slowed down horribly
until we set BUFPOOLSIZE back
that goes unasked.
The command line is your friend.
Good enough is the enemy of excellence.
ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU wrote on 10/12/2009
07:58:58 AM:
[image removed]
instrumentation for the server?
km
to:
ADSM-L
10/12/2009 08:05 AM
Sent by:
ADSM: Dist
.ibm.com/software/tivoli/products/storage-mgr-mail/
-km
On 12/10, ashish sharma wrote:
Hello all,
I have to configure TSM backup for Microsoft exchange server.The problem is
that TSM BAC client is unable to backup open files and the mail admin
requires these files to be backed up anyway.According
in tsm server)
# dsmc inc /home/mehdi -subdir=yes (does not set the time in tsm server)
That is because that field pertains to the last complete backup of the
filesystem.
-km
is a good thing since you
will have fewer common point of failures for both, ie less chance of losing
or having trouble with both at the same time.
-km
node1 \\node1\c$
ANR0852E DELETE FILESPACE: No matching file spaces found for node MEHRGAN.
ANS8001I Return code 11.
Delete by FSID:
DEL FILE node1 2 NAMET=FSID
-km
On 21/08, Loon, EJ van - SPLXM wrote:
Hi KM!
sho aggr 0 1086967169:
AggrId: 0.1086967169, AggrSize: 0.717, LogSize: 0.717, NumFiles: 1.
Aliases: None.
I guess I have to parc the undeletable nodes somewhere until the TSM
server is upgraded to a supported level...
Kind regards,
Eric van
On 20/08, Loon, EJ van - SPLXM wrote:
Hi KM!
I already tried the delete object command. In fact, the objectid is in
the error itself:
ANRD imutil.c(7001): ThreadId 51 unexpected rc=87 from bfDestroy
for objId=0.106696510
So when I issue a delete object 0 106696510 the same error
Manager [mailto:ads...@vm.marist.edu] On Behalf Of
km
Sent: dinsdag 18 augustus 2009 18:34
To: ADSM-L@VM.MARIST.EDU
Subject: Re: Dealloc prohibited - transaction failed
On 18/08, Loon, EJ van - SPLXM wrote:
Hi TSM-ers!
I have two nodes which I cannot delete. When I issue a DELETE
FILESPACE
return codes for TSM...
It's reproducible. Anybody seen this before?
Thank you very much for any reply in advance!
Kind regards,
Eric van Loon
Looks like an IBM support case, probably a missing object or something.
-km
...@vm.marist.edu] On Behalf Of km
Sent: 14 August 2009 00:56
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Change backup copy group settings
On 13/08, James Choate wrote:
I changed my backup copy group setting for a Policy domain that has one
test node in it from
Versions Data Exists: 35
be a backup summary with number
of expired files.
-km
storage needed is when moving from DISK stgpool to the
first FILE stgpool but could be mitigated by moving one volume at a time with
move data.
-km
process I believe the arguments for 6.1
are pretty strong.
-km
have it
doesnt matter.
-km
on the copy
stgpool so make sure there is some bandwidth available during the day.
-km
)}++;
}
for my $pool (sort keys %tapes)
{
for my $pct (sort keys %{$tapes{$pool}})
{
print $tapes{$pool}{$pct}\t . $pct*10 .-. ($pct*10+10) . %\t$pool\n;
}
}
-km
On 29/07, Huebschman, George J. wrote:
Greetings everyone,
I have a/an SQL Select question.
Most of my TSM Servers are at 5.5.1.0, one
volumes. The delete command, I believe, only
removes the backupset entries from the TSM DB and sets the tape(s) back
to scratch. Please do some research on this first so you dont lose any
data.
Maybe someones else on this list have done this recently and can verify
that this is the way to do it.
-km
Functions are not allowed in GROUP BY clause, only column names :(
On 30/07, Steven Harris wrote:
Hi George
I have a copy of the interesting bits of a TSM database in sqlite3 for a
side project I'm doing while resting. On that database I can use
select count(*),
Even easier (no OS involvment) would be to use:
virtualmountpoint c:\dir
domain c:\dir
in dsm.opt
-km
On 23/07, Adrian Compton wrote:
Hi Patryk
I found the best way to selectively backup directories was to create a
share on the directory, and you can then choose that share as a domain
Maybe the classic client encryption could solve this. Then only the customer
would be able to access the data and there would be no need for scrubbing.
On 20/06, Guenther Bergmann wrote:
Hi *TSMlers,
given the following scenario:
- one TSM Server
- a lot of filesystem clients (mostly AIX/RH
/
TSMtape is probably what you want.
-km
Im not sure what this has to do with TSM, you should maybe try a
backuppc maling list, but anyway...
You're gettings errors for door ipc files (type 9) and socket files (type 8).
Could they maybe be the cause for the ioctl error?
Exclude those files or maybe try TSM? :)
-km
On 18/05, ckandreou
What are the relevant trace classes in the TSM server to trace a stgpool
of type file?
Is there any kind of guide/index over what the different trace classes
displays?
-km
Is there any way to disable Direct I/O in TSM 5.5 and/or TSM 6.1 for AIX?
Apparently the option AIXDIRECTIO was removed in 5.3.2 and it is now
always on. This is killing my FILEDEVICE performance on a clustered
filesystem.
-km
with using a primary storage pool without media
redundancy.
-km
On 06/04, Jerome Alet wrote:
Hi,
I'd like to know if it's possible to use TSM in an environment with only
a single LTO2 tape drive and a few TB of hard disk space.
In particular I'd like to know how tape reclamation could be done
On 21/03, Hart, Charles A wrote:
It works well if you understand your data and how you can push to it
with in reason before you deploy another. The IBM product likes more
CPU cores ... Understand these are x86 boxes... We see up to 500MBS
Writes to one of our VTL's that ingests Exchange
by: 14% (SESSION:
18137)
AP
Because it includes alot more?
Network speed is: amount of data backed up / network time.
Aggregate speed is: amount of data backed up / entire backup session
time.
-km
a console session.
Change the password, halt the server and start it normally again.
-km
are
comfortable with.
And ofcourse, POWER when you need something that can push I/O.
-km
On 25/02, Zoltan Forray/AC/VCU wrote:
Been there...tried to do that..and would never put TSM on a
Windoze box.
Your environment is a big factor in platform decisions.
When we first discussed
for.
-km
as well.
-km
On 19/02, Larry Peifer wrote:
Just beginning to install TSM on some new RHEL servers but I have little OS
command experience for Linux. What is the equivalent syntax to get an AIX
'errpt' output on the RHEL system.
On 19/02, Allen S. Rout wrote:
On Thu, 19 Feb 2009 16:27:29 +0100, km k...@grogg.org said:
RAID5 will give you the maximum amount of IOPS for the DB with disk
redundancy, though 4 disks is on the low side.
TSM mirroring doesnt help performance when doing expiration or
reclamation
If all the ACLs are updated daily I guess thats not a problem, just
update them again after the restore :)
Or make a script that saves the ACLs before backup to a file and run as
a preschedulecmd.
Sounds like it could be worth it to avoid unnecessarily backing up 15 TB of
data every day.
-km
subfile but that is/was? a deficiency in NTFS.
But apart from that I guess only the ACL would be backed up...
-km
On 16/02, Len Boyle wrote:
What happens if you are using subfilebackup. Will it just backup the acl's as
the data blocks have not changed?
len
-Original Message
version)
works without a hitch against 5.5 server, even so for the 5.2/5.3 baclient
on AIX.
My rule of thumb is, try it, it will most probably work!
Ofcourse this is is not supported but neither is win2k :)
-km
The final answer to this question if anyone is interested is that the ISC for
TSM 5.5 (6.x) wont be released for Windows 2008.
The ISC for TSM 6.1 will support TSM 5.3, 5.4 and 5.5 on Windows 2008.
Really annoying gap in server/GUI OS support but I'm not really surprised.
-km
On 02/02
Hello,
Is there any word on when the ISC/Admin Center will be supported on
Windows 2008?
And has anyone been successful in installing the Admin Center on 2008
even though it is not supported?
-km
On 12/01, Kauffman, Tom wrote:
On Fri, Jan 9, 2009 at 3:45 PM, Kauffman, Tom kauffm...@nibco.com
wrote:
You can get that pretty easily from the SUMMARY table. You'll have
to play with it though, to pick out specifically which stgpool
process you want.
Start here with:
select *
On 12/01, Larry Clark wrote:
Wanda,
Do you have a query for displaying which volumes contain which filesystems
by node?
SELECT -
node_name AS Node, -
filespace_name AS FS, -
stgpool_name AS Pool, -
volume_name AS Volume -
FROM -
volumeusage -
GROUP BY -
node_name, -
MIGRATE acts on a per
node basis. That is, you wont get more than 1 process per node.
-km
)
settings of NICs and router ports.
-km
I would recommend waiting for TSM 6.1 and evalutating the new included
reporting tool.
On 05/12, hshahizul wrote:
Hi,
Anybody has ever test Servergraph. I wish to test this Servergraph but than.
so terrible in getting the trail version.
I wish to test this product before recommend to
On 02/12, Zoltan Forray/AC/VCU wrote:
I have a test TSM server (5.5.1) which is producing some strange DB
statistics.
**
*** --- Q DB F=D
**
Available Space (MB): 56,336
?
-km
66 matches
Mail list logo
