Works very well. There is a memory assertion error that may not be fixed in
this level, but that problem has always been there.
-Original Message-
From: Dale Gieseke [mailto:[EMAIL PROTECTED]
Sent: Thursday, April 03, 2003 1:27 PM
To: [EMAIL PROTECTED]
Subject: upgrade TDP for R3
Collocation can also cause fewer tape drives to be used.
-Original Message-
From: Hart, Charles [mailto:[EMAIL PROTECTED]
Sent: Monday, March 31, 2003 11:30 AM
To: [EMAIL PROTECTED]
Subject: Re: TSM Server not using maximum number of drives.
The Resource Utilization parameter in your
Ever tried maxpgahead? That is why you want JFS. Causes things to fly.
-Original Message-
From: Emil S. Hansen [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 27, 2003 4:54 AM
To: [EMAIL PROTECTED]
Subject: Re: Defining TSM Disk pools on ESS wisdom wanted
On Thu, Mar 27, 2003 at
This is a problem for all the newer tape technologies. Typically, this is
caused by a bad batch of tapes or a case of tapes that has been slammed on
the floor or the librarian stacked them and knocked them on the floor.
Early in the delivery of K tapes there was a packing problem.
Also, make
This problem was fixed in the F_0295 Level (I think that is right). It is
the latest level.
-Original Message-
From: Zoltan Forray/AC/VCU [mailto:[EMAIL PROTECTED]
Sent: Monday, March 03, 2003 3:45 PM
To: [EMAIL PROTECTED]
Subject: Re: Errors on 3590 K tapes
Sounds a lot like what I
I believe at the 4.2.1 level you have to get a special version of the client
to run on AIX V5.1.
ftp://service.boulder.ibm.com/storage/tivoli-storage-management/patches/clie
nt/v4r2/AIX/v423/AIX51/
-Original Message-
From: Jeff Rogers [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 04,
I have seen this product demonstrated, pretty impressive.
-Original Message-
From: Chris Young [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 04, 2003 7:20 AM
To: [EMAIL PROTECTED]
Subject: Re: Tivoli Storage Resource Reporting
While I couldn't comment in any great deal on this, I have
This is not a TSM issue it is a FC issue. Are you running your tape on the
same HBAs as the Disk containing the Oracle database. This is a no-no if
you are running random I/O to the Oracle database at the same time. You
will need dedicated HBAs for the tape from the Oracle database client if
We found the problem, but have not had time to reproduce it for Germany.
Back off to the 32 bit API and TDP. Everything will work just fine then.
-Original Message-
From: Bruce Lowrie [mailto:[EMAIL PROTECTED]
Sent: Friday, February 28, 2003 1:42 PM
To: [EMAIL PROTECTED]
Subject: TDP for
I do not think the first two require the library. Have you check the rights
on libobk.a? It may be that world rights are being used because libobk.a is
owned by someone other than who's requesting it. I bet world rights are not
executable.
-Original Message-
From: Guy Korn
I cannot remember now, but I believe this is one upgrade that absolutely
required both to go at the same time. I am thinking it is 4.2.1.8 that
required all to be updated. Prior to 4.2.2.0 they did not check the patch
level, but since it caused so much grief for support it is now enforced.
In encryption speak. The node name is usually called the public key. The
private key is what is used to encrypt the message. This is a nice
implementation because during password change (which is probably in the
message) the new encyption key (password) is not exposed.
Paul D. Seay, Jr.
If this is a dedicated AIX machine set vmtune maxperm to 20 and minperm to
10 otherwise paging will start happening.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180
-Original Message-
From: David le Blanc [mailto:[EMAIL PROTECTED]]
Sent:
Have you considered using the dsm.opt to contain exclude.fs commands for the
ones you do not want to backup and let it just backup everything else.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180
-Original Message-
From: Jitendra Thakur
Not Quite! This used to be true before 4.2.2.0, not any longer. They must
match right down to the patch level. I was bitten by 4.2.2.12 and 4.2.2.13.
However, TSM Development is promising to make this issue go away.
Hopefully, in V5.2. You will have to upgrade all at once then, probably,
but
If your 3590 drives have the green 2x sticker on the back of them (open the
cabinet and look), they can support K cartridges. 3590H drives can support
K-Carts. The K carts have a different knotch configuration than the J
cartridges. There is no issue with having a mixed bag except it throws off
I like to set it about 10% higher than the happy point and set it to yes on
AIX. Then check it every once and a while to see if it needs adjusting.
The problem is you can create a lot of GETMAINs on MVS if you do not set it
high enough to begin with. On MVS, you are probably best to set it to NO
This will give you the filespace usage, which is the active backupset size
except it includes excluded data. It should be good enough for an estimate.
select node_name, cast(sum(capacity*pct_util/100) as decimal(10,3)) as MB
of Used from filespaces group by node_name
However, I am not sure how
No, but there will probably be a V5.2 pre-announcement update at Share in
Dallas February 24-28.
BMR has been mentioned as part of their intended product plan, but no
commitments on when. The issue is Veritas BMR is fine for like
configurations, but not disaster recovery where the configurations
None of these. Install the latest one (D0IF_295) that fixes some nasty
problems especially if you are using ES-1000s. D0IF_26E regresses a fix
that the previous firware had for ES-1000s. There are also a lot fixes
related to bad tapes. There is a change to improve cleaning cycle needs and
to
This is why it took soo long for them to actually figure out what the
problem was. As it turns out, if the session did not exceed the coded
server timeout the problem did not show up on any release of the client or
the server. It is when it went over. In order to fix the problem, the
client
I think any aggregates that have deleted files in them are still counted as
whole. So, if reclamation ever runs against the primary and any copy pools
it is highly likely some aggregates in one pool would be reclaimed and in
others they may not. The other possibility is you have more than one
I suspect that you cannot even check both tapes into the library, though I
have never done it. This is why. When you check a volume in as scratch it
checks the volumes table to make sure the volume is not in the table. This
is to protect the volume contents from being destroyed (referential
I believe this is more than a server issue. I have been assured it is fixed
in 5.1.1.2 and higher, but I have not installed it yet. Apparently, the
client needs to be updated as well to get a complete fix. I do not know if
there is a client level for 4.2.3.2 and higher that completely
To: [EMAIL PROTECTED]
cc:
Subject:Re: summary-table
At 3:19 PM -0500 2/7/03, Seay, Paul wrote:
Apparently, the
client needs to be updated as well to get a complete fix. I do not
know
if
there is a client level for 4.2.3.2 and higher that completely
eliminates the problem
In V5.1 and up you can specify linked copy pools and actually create the
primary and the copy pools at the same time, but you have to have 3 tape
drives.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180
-Original Message-
From: Stapleton, Mark
It processes the files in order of the .anf file. As it turns out they are
recorded on the tape in that order. TDP for SAP will rewind and dismount
the tape if you do not have a mount point hold time on the device class.
So, if you have it set to zero, you will need to increase it a to maybe 2
Your ethernet adapter in the SUN machine probably took the default of AUTO
and is trying to continually negotiate with the switch. Change it to Full
or half as appropriate and the problem will go away. Seen this time and
time again.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman
I recommend TDP for Mail.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180
-Original Message-
From: Amini, Mehdi [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, February 04, 2003 9:17 AM
To: [EMAIL PROTECTED]
Subject: Domino Backup
We just got a
Yes, but it is ugly.
Move the filespace to a new primary pool temporarily, or permanently if you
like.
Create a new Copy storage pool and run a backup storage pool command of the
original storage pool command.
The delete the old copy storage pool (you have to delete all the volumes in
that copy
You have to do a CHECKIN LIBVOLUME command with the parameters particular to
your library. The tape will automatically come back as status=private even
if you specify status=scratch because they are still known to have good data
if that is really the case.
Now, you are requesting these for a
This is a similar approach to what we use to capture a set of Exchange data.
We determine the primary volumes that the data is on that we need. Move
those volumes to a new primary pool. Copy that primary pool to a copy pool,
database snapshot, eject the copy pool tapes, delete the copy pool
The typical cause is the tapes were not labeled. You checked them in, but
you should have done a LABEL LIBVOLUME .. CHECKIN=Scratch. Just check
them back out and do the LABEL command.
This is the normal action TSM takes to get a bad tape marked for no reuse
when it is not labeled.
Paul D.
There were some permission restore issues at this level, I do not remember
what they were. This level is not going to be supported after 4/15.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180
-Original Message-
From: Edgardo Moso
This is about what I would expect. Eventhough you are using the SAN agent,
you could be still using the IP stack internal to the machine to get the
data from the client to the SAN agent. And, if this is TDP for Exchange or
TDP for SQL, there is even more CPU overhead. Without more information,
In the management class you set the primary storage pool. To run the first
backup just direct it to the TAPE primary management class that is the
NEXTPOOL for the DISK management class. After the first backup, change the
storage pool to the DISK management class.
Backups are determined by the
This is an OS support and boot issue. HP has been slow to support open SAN.
This ability has only been available on AIX for a little while.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180
-Original Message-
From: [EMAIL PROTECTED]
If you have the 3494 mtlib code installed for Windows you can issue this
command and not even go to the library. The command is:
mtlib -l [library] -m -x [device serial] -V [volume number]
After TSM is done with the tape it will unload it but not put it away.
mtlib -l [library] -d -x [device
Check your default IP packet sizes and IP performance implementation. The
more packets you have the more overhead to process them. We do not have a
SUN TSM server, but we run SAN Storage agents and Clients on SUN.
Also, check your TDP for SAP implementation and make sure you are using the
most
With TSM V5.1 and below a drive has to be defined, visible, and online to
the TSM server and the SAN Storage Agent. Essentially, the SAN Storage
Agent is a cut down TSM server code set that does its database I/O remotely.
You have another option, install the TSM server code where you were going
Consistent return codes was a Share requirement for years so that production
processes could actually be coded and have a determinable consistent result.
You may not like the implementation, but it is what was asked for.
Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information
NO
NO
YES
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Nicolas Savva [mailto:[EMAIL PROTECTED]]
Sent: Thursday, January 23, 2003 6:54 AM
To: [EMAIL PROTECTED]
Subject: Some information...
Hi,
I need some help for the following:
1. Is TSM
This is exactly what we do.
Basically, you run two backup storage pool commands for each primary pool to
two different copy pools. Naming conventions become really important here
to help prevent mistakes and to use masks.
We call our pools like this:
DSK_Primary
, January 22, 2003 12:58 PM
To: [EMAIL PROTECTED]
Subject: Re: Tape Requests Never Mounting
I can't rule out a H/W problem, but...I did upgrade to 5.1.6 and the problem
went away. How solid is the tsmscsi stuff in 5.1.5 for Linux?
Mitch
Seay, Paul wrote:
You likely have a marginal SCSI
Yes, but you have to be at 4.1.4 and higher I believe. It may be higher
than that. It is called the system object for Windows 2000 clients and
contains a lot more than just the registry. It contains dlls and many other
files considered part of the system object. The biggest issue right now is
The bitfile is only needed to find out what tape an object is on.
This is a nasty issue. I am guessing what you really want is to know if a
specific machine registry gets broken and is not getting backed up so you
can address whatever happened. I think a select of the inactives not in the
This is why I run and incremental and a dbsnapshot every day. And two TSM
database backups for every offsite run. And a primary, onsite copy, and
offsite copy (DR) for filesystem backups. And, applicaton database and
application log file offsite copies to roll forward a bad application backup
Apparently, the BACKUP DB command does some pre processing now. Not sure
what. My system can take 3 to 5 minutes. I think what is happening is some
log commit logic is running for a while to try to unpin as much of the log
as possible, but that is just a guess. Then, my backup starts. Mine
Which 64 bit file system?
Whenever posting a question like this please provide:
Client or Server type and OS level.
Filesytem type (JFS, JFS2, Veritas, NTFS, XFS, UFS, etc).
Often folks just do not bother to respond to a question like this because of
the missing information.
You likely have a marginal SCSI cabling problem, missing terminator, tape
driver or something like this. This is probably not a TSM issue.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Mitch Sako [mailto:[EMAIL PROTECTED]]
Sent: Tuesday,
Roger, I agree with you on your point; however, there is one consideration
which we have found requires and unload/load. We do a lot of delete
filespaces because we have a lot of server migrations and are required to
keep the old image for 1 year, then delete it. This is also true for
filespaces
of this response
On Saturday, January 18, 2003, at 02:36 AM, Seay, Paul wrote:
As Zorg would say, I know the sound of this music.
The default maxperm is probably killing you. I am guessing you are
swapping
snip
I agree whole-heartedly! I'd go a step further, and knock maxperm down
As Zorg would say, I know the sound of this music.
The default maxperm is probably killing you. I am guessing you are swapping
more than you are running and your swap drives are I/O hot, a iostat will
tell you, or topas. This value dictates the amount of storage that can be
consumed by
There is only one size right now, 4096.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Farren Minns [mailto:[EMAIL PROTECTED]]
Sent: Thursday, January 16, 2003 5:37 AM
To: [EMAIL PROTECTED]
Subject: Database page size?
Hi TSMers
What is the
Tom, you are correct. This is the way we do the checkin commands when DR
tapes return because we do a lot of move data commands. What will happen is
tapes that have no data on them will end up being marked private in the
libvolumes table. However, it is easy to fix. All you have to do is an
Files that do not have an active version do not get rebound if you set them
to a new policy domain management class. In fact, we have had to create
dummy files and run a backup with a special management class to get this
kind of data to rebind.
I really want a command that I can say rebind to a
It really depends on the hardware you have.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Ruksana Siddiqui [mailto:[EMAIL PROTECTED]]
Sent: Thursday, January 16, 2003 6:19 PM
To: [EMAIL PROTECTED]
Subject: TSM database question
I have TSM
with 2 CPU's and 2 GB memory ( PHYSICAL
).
With regards,
-Original Message-
From: Seay, Paul [mailto:[EMAIL PROTECTED]]
Sent: Friday, 17 January 2003 10:31
To: [EMAIL PROTECTED]
Subject: Re: TSM database question
It really depends on the hardware you have.
Paul D. Seay, Jr.
Technical
4.2.3.0 is not good enough. You need to go to 4.2.3.2.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Charakondala, Chandrasekhar R. [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, January 08, 2003 4:19 AM
To: [EMAIL PROTECTED]
Subject: Any side
You have to be at 4.2.3.2 and there is a special patch you may have to put
on if cleanup backupgroups fails at the 4.2.3.2 level. Has a cleanup
backupgroups been run?
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Michael Moore [mailto:[EMAIL
How many tapes do you have in the tapepool?
Select count(*) from volumes where stgpool_name='TAPEPOOL'
How many drives do you have?
How many mounts are you getting during a stgpool backup?
Do you have maxpr specified on the stgpool backup?
What is the device class mount retention?
Are you
4.2.3.2 for the Server and latest client available.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Hokanson, Mark [mailto:[EMAIL PROTECTED]]
Sent: Monday, January 13, 2003 11:00 AM
To: [EMAIL PROTECTED]
Subject: AIX TSM Server, Client, and
Yeah, Roman numeral M is a 40+ year practice that sales people used at the
wholesale and manufacturing levels as kind of a shorthand for 1000. As they
have migrated to computers this has mostly gone away because the quantity
fields only supported ea items standing for each.
Paul D. Seay, Jr.
]
Subject: Re: Calculate 1 MB in TSM
Seay, Paul wrote:
Yeah, Roman numeral M is a 40+ year practice that sales people used
at the wholesale and manufacturing levels as kind of a shorthand for
1000. As they have migrated to computers this has mostly gone away
because the quantity fields only
I think you will have to run the dsmc as root.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Hokanson, Mark [mailto:[EMAIL PROTECTED]]
Sent: Friday, January 10, 2003 3:22 PM
To: [EMAIL PROTECTED]
Subject: Storage Agent Error: ANRD
Unfortunately, our bean counters started using 60K for $60,000. When the
mainframe 3380 came out they decided to start counting disk in 1000s and
100s. This has just perpetuated on for disk to make it easy for people
to calculate the space required because data records are measured in base
Keith, we worked with Tivoli to fix this problem. There was a memory leak
in the ACL processing routine. The latest TSM Client for SGI fixed this
problem and a previous patch level has the fix. I believe the current level
is 4.2.3.0.
What do you mean, they are in the inport. The 3494 should put the tapes
away with a category of FF00 and then you just check them in. You should be
able to do a mtlib command to see their status.
I believe you are getting confused with the J on the tape. That is not part
of the serial number.
:43 AM
To: [EMAIL PROTECTED]
Subject: Re: Gigabit Ethernet speed
Hi,
Can we use 2 or more gigabit interfaces to increase backup bandwidth? Any
experience on this approach?
Thanks in advance,
António Pires
Seay, Paul
seay_pd@NAPTHEON
with
CPU utilization.
-Original Message-
From: Seay, Paul [mailto:[EMAIL PROTECTED]]
Sent: December 27, 2002 2:53 PM
To: [EMAIL PROTECTED]
Subject: Re: Gigabit Ethernet speed
For some reason we forget that IP packet processing on the client and server
take a lot of CPU resources. Check your
For some reason we forget that IP packet processing on the client and server
take a lot of CPU resources. Check your CPU utilization on both to see what
is happening. My experience is a 450mhz x 4 P660 can process maximum of
about 7 kb/sec. When we had only one gigabit interface that is
Just being able to use the extended SAPDBA utilities that TDP provides the
hooks for is a plus. No SAPDBA should leave home without them.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Bill Zhang [mailto:[EMAIL PROTECTED]]
Sent: Friday,
This subject looked like burnt potatoes the last time it was discussed. It
is finished being discussed here, the archives are the answer!
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Mark D. Rodriguez [mailto:[EMAIL PROTECTED]]
Sent:
Try this on for size.
select node_name, cast(platform_name as char(16)) as OS/Name ,
cast(client_os_level as char(10)) as OS/Level, cast (client_version as
char(1)) || '.' || cast(client_release as char(1)) || '.' || cast
(client_level as char(1)) || '.' || trim(cast(client_sublevel as char(2)))
The short answer is probably yes. TSM does some checking of files that span
tapes in the reclamation criteria to prevent an endless chain of
reclamation.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: John [mailto:[EMAIL PROTECTED]]
Sent:
-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 12, 2002 3:26 PM
To: Seay, Paul
Subject: RE: Select Statement: The ANSWER
Thanks, I will try the command line.
Seay, Paul
seay_pd@naptTo:
'[EMAIL PROTECTED
I think it is more of a question when a VTS replacement would become
available and 3590 is significantly slower than other technologies. There
is still development going on for the 3494 library itself. There will be a
replacement 3590 drive soon. It is not clear if that drive will be in the
Try changing the buffers to 64K instead of the default. And, to a number of
4. Everyone that I have suggested this change to and our own site have
found that the performance is good and the fragmentation issues with getting
1MB buffers is eliminated.
Paul D. Seay, Jr.
Technical Specialist
I have seen this on the 4.2.x.x server code levels. What causes it to hang
is a storage pool related command by an ADMIN. If you do a q sesions, you
should find a admin that has been running for a long time. Cancel the admin
session and the problem will clear.
Paul D. Seay, Jr.
Technical
will be
proper :-)
Dwight
-Original Message-
From: Seay, Paul [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, December 11, 2002 11:47 PM
To: [EMAIL PROTECTED]
Subject: Re: K-cartridges besides J-cartridges
And, if you look at a J and a K tape on the notches you will notice that
they have
What have you set resourceutilization in the dsm.opt and mount point to for
the node?
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: J D Gable [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 12, 2002 1:55 PM
To: [EMAIL PROTECTED]
Subject:
PQ68076 is MVS ONLY related to GETMAIN activity, not AIX and there is no
equivalent issue on AIX. So, the problem is something else.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Dan Foster [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December
We have setup something almost identical to what Tom has done. This is an
excellent design based on our experiences as well. Our retentions are
different, but we use the same concepts.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Kauffman,
Do not read into anything about market share statistics. Suffice to say,
the only two products that get any significant press now are TSM and Veritas
NetBackup. They have market share in around the 20 percentile. A large
number of Backup Exec licenses exist and are continually sold, but that is
And, if you look at a J and a K tape on the notches you will notice that
they have a different configuration. So, if you put a K tape in a J only
drive, it will kick it right back out. I thought that all B to E upgrade
kits were K capable. I did not think that kit came out until the E upgrade
The reality is this. File system backups of small files are not a good
candidate for SAN Tape backups. Netware does not run any large file
databases, so there is no SAN Storage Agent requirement. The answer is
Gigabit between your TSM server and the client and let the TSM server stream
the data
You have to define them as a ADMIN with no grants. They will get the query
commands with that by default. That may be more than you want to give them.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Elio Vannelli [mailto:[EMAIL PROTECTED]]
The best way is to use two different nodes and two different management
classes/copygroups/pools. This way your offsite node is the only one used
to manage that data. We do some things very similar to this.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original
select summary.entity as NODE NAME, nodes.domain_name as DOMAIN,
nodes.platform_name as PLATFORM, cast((cast(sum(summary.bytes) as float) /
1024 / 1024) as decimal(10,2)) as MBYTES , count(*) as CONNECTIONS, cast
(client_version as char(1)) || '.' || cast (client_release as char(1)) ||
'.' || cast
Many do RL compression in the TDP. Some do client compression.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: brian welsh [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 10, 2002 5:19 PM
To: [EMAIL PROTECTED]
Subject: Help on TDP for SAP
As it turns out the clients also have to be updated to correct a problem
where the summary table statistics are missing. I talked to a TSM Level 2
person about this just last week. I do not know what levels you have to be
at to get the problem corrected, but I thought 5.1.5 was good enough.
My database saves at about 22,000,000 pages an hour. I have taken the
messages from the log and they are consistently the same. I do not know
what is causing this. There were some performance problems with mirrored
databases, but I thought they were fixed at your level.
Paul D. Seay, Jr.
Dwight, there is only one problem with this, SELECTIVE does not update these
fields in the database. There was a long discussion about this. The sum of
it all was these fields are to support incrbydate. However, I do the same
thing you do because we do not use selective backups.
Paul D. Seay,
We have a requirement to restore the transaction logs back to their disk
location either on the orginal or an alternate server or to be able to post
process the transaction logs as disk files some kind of way before they get
deleted after a TDP for SQL Server Backup.
My DBAs tell me SQL server
The easiest way is to use two different node stanzas and have 2 different
primary pools.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original Message-
From: Joshua Bassi [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 03, 2002 10:38 AM
To: [EMAIL PROTECTED]
OK, I see what you mean.
I guess I misunderstood what the product is probably doing. The company
that has this auditing product is LUMIGENT, www.lumigent.com. The product
is Entegra. I have not talked to their developers as yet, but they seem to
be interested in retrieving the objects from TSM
F_026E backleveled a fix from a previous release related to SAN LIP
protocols and login that was on, I believe, D_0388. To correct the problem
F_027E is the fix. This fix is not generally available yet.
If you have McData ES-1000 switches, you are especially exposed. Basically,
one drive can
Is anyone monitoring the drive internal logs or looking at them. It sounds
like the drive pressure is out of alignment in all the drives. This could
be because the CE you have has no idea what they are doing.
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-Original
The storage pool backup that you are doing could be directly affected by the
database performance. Is this 1.6TB a lot of small files?
On the IO configuration side, how are your tape drives connected, SCSI or
FC? I suspect SCSI. If you have more than 2 drives, on a SCSI adapter that
is the
Actually, there are 2 speeds to be concerned with.
Data transfer from the host to the 3590 controller.
SCSI, depends on length and type of connection (34 was the top on
SCSI E1A).
40 MB/sec on FC E1A advertised, but I have seen higher (FC)
70 MB/sec on FC H1A advertised
1 - 100 of 859 matches
Mail list logo
