Re: Question about Space Reclamation..
Thanks to all of you..
With a weekend of reading and some experimentation, I believe that the
problems have been overcome. I might still have lots of questions, but you
all have been very helpful.
Roark
"Maurice van 't
Loo"
<[EMAIL PROTECTED] To
TERS.NL> ADSM-L@VM.MARIST.EDU
Sent by: "ADSM:cc
Dist Stor
Manager" Subject
<[EMAIL PROTECTED] Re: [ADSM-L] Question about Space
.EDU> Reclamation..
08/30/2005 04:10
AM
Please respond to
"ADSM: Dist Stor
Manager"
<[EMAIL PROTECTED]
.EDU>
True ;-)
I use Operational Reporting and Access with a bunch of advanced selects (on
imported tables) to monitor and the command admin client to enter the
commands... that's cheaper... more Dutch ;-)
Regards,
Maurice
- Original Message -
From: "Bos, Karel" <[EMAIL PROTECTED]>
To:
Sent: Tuesday, August 30, 2005 12:20 PM
Subject: Re: [ADSM-L] Question about Space Reclamation..
Hi,
Just one thing to add. If the command line isn't your friend, tools like
tsmmanager can help you a lot. Just checkout the website (and look at
the pricing) and download the demo.
Regards,
Karel
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Maurice van 't Loo
Sent: dinsdag 30 augustus 2005 11:05
To: ADSM-L@VM.MARIST.EDU
Subject: Re: Question about Space Reclamation..
Hoi Roark,
If you cannot affort to upgrade to 5.3.x, you can also use more
tapepools.
So you can make a tapepool for very small nodes, with collocation off,
so it still uses less tapes and a tapepool for the big nodes with
collocation on.
If neccesary, you can make more tapepools, but remember, if you want to
use a diskpool (and most sites do) you need a diskpool for each
tapepool.
But best thing is to upgrade to 5.3.x and learn the command line
commands :-), so you can use collocation groups.
Collocation on copypools is most of the times nonsens... Even if you
want it in case of a big disaster, the costs are enormous and fast
restores per machine is not the biggest issue at that moment if you can
use a library with enoufh drives to restore multiple machines
simmultanious.
You cannot change a "filling" tape to "full", but you can use "move data
{volname} stg={diskpool}" to move the data of the tape quickly to disk.
If you set the access of the tapes you want to clean to readonly first,
the data will be migrated to the filling tapes you don't want to clean.
Regards,
Maurice van 't Loo
- Original Message -----
From: "Roark Ludwig" <[EMAIL PROTECTED]>
To:
Sent: Friday, August 26, 2005 1:56 AM
Subject: Re: [ADSM-L] Question about Space Reclamation..
> I have been handed a TSM system in the past week that has been running
for
> a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this
system
> is running TSM 5.2.2.
>
> We have run out of tapes as the setting for collocation (yes) is
> trying to use a tape for each node (as I understand it, ,,please
> correct me if I am wrong)..
>
> We have decided that since we have about 80 volumes with less then 1%
> utilized and 20 or so with large utilization %'s to set COLLOCATION=NO
> for the TAPEPOOL and COPYPOOLS.
>
> We are expecting to add another set of nodes and don't wish to consume
more
> of our volumes (with small percent utilization) as we add the nodes.
>
> Question ..
>
> Will this setting COLLOCATION=NO stop the addition of volumes? (I
> expect the answer is YES,, please correct me if I am wrong.)
>
> Now to the second question.
>
> I see no easy way to have Space Reclamation condense the volumes
> included in the two pools as it will only process "FULL" volumes.
> (Again Please
jump
> in here).
>
> QUESTION: is it acceptable to set the status from "FILLING" to "FULL"
> for the volumes with low percentage utilized to force Space
Reclamation ?
>
> OR
>
> is there an easier way to accomplish the GOAL of reducing the number
of
> volumes needed for the pools?(given that we have set
COLLOCATION=NO)
> or should we simply wait out the natural filling of the volumes and
> allow normal Space Reclamation as time proceeds?
>
>
> Has anyone done this in the past?
>
> How have others dealt with this??
>
>
> Thanks for any input.
>
Re: Question about Space Reclamation..
True ;-)
I use Operational Reporting and Access with a bunch of advanced selects (on
imported tables) to monitor and the command admin client to enter the
commands... that's cheaper... more Dutch ;-)
Regards,
Maurice
- Original Message -
From: "Bos, Karel" <[EMAIL PROTECTED]>
To:
Sent: Tuesday, August 30, 2005 12:20 PM
Subject: Re: [ADSM-L] Question about Space Reclamation..
Hi,
Just one thing to add. If the command line isn't your friend, tools like
tsmmanager can help you a lot. Just checkout the website (and look at
the pricing) and download the demo.
Regards,
Karel
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Maurice van 't Loo
Sent: dinsdag 30 augustus 2005 11:05
To: ADSM-L@VM.MARIST.EDU
Subject: Re: Question about Space Reclamation..
Hoi Roark,
If you cannot affort to upgrade to 5.3.x, you can also use more
tapepools.
So you can make a tapepool for very small nodes, with collocation off,
so it still uses less tapes and a tapepool for the big nodes with
collocation on.
If neccesary, you can make more tapepools, but remember, if you want to
use a diskpool (and most sites do) you need a diskpool for each
tapepool.
But best thing is to upgrade to 5.3.x and learn the command line
commands :-), so you can use collocation groups.
Collocation on copypools is most of the times nonsens... Even if you
want it in case of a big disaster, the costs are enormous and fast
restores per machine is not the biggest issue at that moment if you can
use a library with enoufh drives to restore multiple machines
simmultanious.
You cannot change a "filling" tape to "full", but you can use "move data
{volname} stg={diskpool}" to move the data of the tape quickly to disk.
If you set the access of the tapes you want to clean to readonly first,
the data will be migrated to the filling tapes you don't want to clean.
Regards,
Maurice van 't Loo
- Original Message -
From: "Roark Ludwig" <[EMAIL PROTECTED]>
To:
Sent: Friday, August 26, 2005 1:56 AM
Subject: Re: [ADSM-L] Question about Space Reclamation..
> I have been handed a TSM system in the past week that has been running
for
> a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this
system
> is running TSM 5.2.2.
>
> We have run out of tapes as the setting for collocation (yes) is
> trying to use a tape for each node (as I understand it, ,,please
> correct me if I am wrong)..
>
> We have decided that since we have about 80 volumes with less then 1%
> utilized and 20 or so with large utilization %'s to set COLLOCATION=NO
> for the TAPEPOOL and COPYPOOLS.
>
> We are expecting to add another set of nodes and don't wish to consume
more
> of our volumes (with small percent utilization) as we add the nodes.
>
> Question ..
>
> Will this setting COLLOCATION=NO stop the addition of volumes? (I
> expect the answer is YES,, please correct me if I am wrong.)
>
> Now to the second question.
>
> I see no easy way to have Space Reclamation condense the volumes
> included in the two pools as it will only process "FULL" volumes.
> (Again Please
jump
> in here).
>
> QUESTION: is it acceptable to set the status from "FILLING" to "FULL"
> for the volumes with low percentage utilized to force Space
Reclamation ?
>
> OR
>
> is there an easier way to accomplish the GOAL of reducing the number
of
> volumes needed for the pools?(given that we have set
COLLOCATION=NO)
> or should we simply wait out the natural filling of the volumes and
> allow normal Space Reclamation as time proceeds?
>
>
> Has anyone done this in the past?
>
> How have others dealt with this??
>
>
> Thanks for any input.
>
Re: Question about Space Reclamation..
Hi,
Just one thing to add. If the command line isn't your friend, tools like
tsmmanager can help you a lot. Just checkout the website (and look at
the pricing) and download the demo.
Regards,
Karel
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Maurice van 't Loo
Sent: dinsdag 30 augustus 2005 11:05
To: ADSM-L@VM.MARIST.EDU
Subject: Re: Question about Space Reclamation..
Hoi Roark,
If you cannot affort to upgrade to 5.3.x, you can also use more
tapepools.
So you can make a tapepool for very small nodes, with collocation off,
so it still uses less tapes and a tapepool for the big nodes with
collocation on.
If neccesary, you can make more tapepools, but remember, if you want to
use a diskpool (and most sites do) you need a diskpool for each
tapepool.
But best thing is to upgrade to 5.3.x and learn the command line
commands :-), so you can use collocation groups.
Collocation on copypools is most of the times nonsens... Even if you
want it in case of a big disaster, the costs are enormous and fast
restores per machine is not the biggest issue at that moment if you can
use a library with enoufh drives to restore multiple machines
simmultanious.
You cannot change a "filling" tape to "full", but you can use "move data
{volname} stg={diskpool}" to move the data of the tape quickly to disk.
If you set the access of the tapes you want to clean to readonly first,
the data will be migrated to the filling tapes you don't want to clean.
Regards,
Maurice van 't Loo
- Original Message -
From: "Roark Ludwig" <[EMAIL PROTECTED]>
To:
Sent: Friday, August 26, 2005 1:56 AM
Subject: Re: [ADSM-L] Question about Space Reclamation..
> I have been handed a TSM system in the past week that has been running
for
> a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this
system
> is running TSM 5.2.2.
>
> We have run out of tapes as the setting for collocation (yes) is
> trying to use a tape for each node (as I understand it, ,,please
> correct me if I am wrong)..
>
> We have decided that since we have about 80 volumes with less then 1%
> utilized and 20 or so with large utilization %'s to set COLLOCATION=NO
> for the TAPEPOOL and COPYPOOLS.
>
> We are expecting to add another set of nodes and don't wish to consume
more
> of our volumes (with small percent utilization) as we add the nodes.
>
> Question ..
>
> Will this setting COLLOCATION=NO stop the addition of volumes? (I
> expect the answer is YES,, please correct me if I am wrong.)
>
> Now to the second question.
>
> I see no easy way to have Space Reclamation condense the volumes
> included in the two pools as it will only process "FULL" volumes.
> (Again Please
jump
> in here).
>
> QUESTION: is it acceptable to set the status from "FILLING" to "FULL"
> for the volumes with low percentage utilized to force Space
Reclamation ?
>
> OR
>
> is there an easier way to accomplish the GOAL of reducing the number
of
> volumes needed for the pools?(given that we have set
COLLOCATION=NO)
> or should we simply wait out the natural filling of the volumes and
> allow normal Space Reclamation as time proceeds?
>
>
> Has anyone done this in the past?
>
> How have others dealt with this??
>
>
> Thanks for any input.
>
Re: Question about Space Reclamation..
Hoi Roark,
If you cannot affort to upgrade to 5.3.x, you can also use more tapepools.
So you can make a tapepool for very small nodes, with collocation off, so it
still uses less tapes and a tapepool for the big nodes with collocation on.
If neccesary, you can make more tapepools, but remember, if you want to use
a diskpool (and most sites do) you need a diskpool for each tapepool.
But best thing is to upgrade to 5.3.x and learn the command line commands
:-), so you can use collocation groups.
Collocation on copypools is most of the times nonsens... Even if you want it
in case of a big disaster, the costs are enormous and fast restores per
machine is not the biggest issue at that moment if you can use a library
with enoufh drives to restore multiple machines simmultanious.
You cannot change a "filling" tape to "full", but you can use "move data
{volname} stg={diskpool}" to move the data of the tape quickly to disk. If
you set the access of the tapes you want to clean to readonly first, the
data will be migrated to the filling tapes you don't want to clean.
Regards,
Maurice van 't Loo
- Original Message -
From: "Roark Ludwig" <[EMAIL PROTECTED]>
To:
Sent: Friday, August 26, 2005 1:56 AM
Subject: Re: [ADSM-L] Question about Space Reclamation..
> I have been handed a TSM system in the past week that has been running for
> a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this
system
> is running TSM 5.2.2.
>
> We have run out of tapes as the setting for collocation (yes) is trying to
> use a tape for each node (as I understand it, ,,please correct me if I am
> wrong)..
>
> We have decided that since we have about 80 volumes with less then 1%
> utilized and 20 or so with large utilization %'s to set COLLOCATION=NO for
> the TAPEPOOL and COPYPOOLS.
>
> We are expecting to add another set of nodes and don't wish to consume
more
> of our volumes (with small percent utilization) as we add the nodes.
>
> Question ..
>
> Will this setting COLLOCATION=NO stop the addition of volumes? (I expect
> the answer is YES,, please correct me if I am wrong.)
>
> Now to the second question.
>
> I see no easy way to have Space Reclamation condense the volumes included
> in the two pools as it will only process "FULL" volumes. (Again Please
jump
> in here).
>
> QUESTION: is it acceptable to set the status from "FILLING" to "FULL" for
> the volumes with low percentage utilized to force Space Reclamation ?
>
> OR
>
> is there an easier way to accomplish the GOAL of reducing the number of
> volumes needed for the pools?(given that we have set COLLOCATION=NO)
> or should we simply wait out the natural filling of the volumes and allow
> normal Space Reclamation as time proceeds?
>
>
> Has anyone done this in the past?
>
> How have others dealt with this??
>
>
> Thanks for any input.
>
Re: Question about Space Reclamation..
Hi, I can think of 2 ways around this - 1. install 5.3 (at level 5.3.1.13 to avoid a bad bug). I have done this And collocation groups work great, the feature is a great solution to this problem. 2. When migration occurs, TSM selects the volume currently used by the node. Do a process like this to put all the small nodes onto 1 volume. A. list all the low use volumes. B. make move data commands for all but one. C. update all the volumes to readonly except the one omitted in B. D. run the move data commands one at a time. E. make all the volumes readwrite again. Hope this helps, Bill Colwell > -Original Message- > From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] > On Behalf Of Prather, Wanda > Sent: Friday, August 26, 2005 1:38 PM > To: ADSM-L@VM.MARIST.EDU > Subject: Re: Question about Space Reclamation.. > > There is a middle ground here. > If you set COLLOCATION=NO, it will drastically affect your RESTORE > performance. > Over time, you could in theory have to mount EVERY TAPE to do a full > restore of a client. > > That is why collocation is a good thing. > But collocation will, by default, put each client on its own tape, > thereby using a LOT of tapes. > > The alternative is to control collocation by setting the MAXSCRATCH > value for your TAPEPOOL. > > Assume you are starting out with a new TSM server; if you have 80 > clients and set MAXSCRATCH to 40, > TSM will stack 2 clients per tape. > This is a good thing, too. > > You get 90% of the benefits of collocation, with only half the tapes. > This is a very common thing to do; there are plenty of sites that have > more clients than slots in their library! > > The drawback is that you have to check on the value of > MAXSCRATCH often. > As you get more tapes in FULL status (assuming you are like > the rest of > us and your site is CONSTANTLY growing its storage farm), you have to > keep increasing the value of MAXSCRATCH. That's easy: update stgpool > TAPEPOOL maxscratch=xxx. Or do it from the admin GUI. > > What I do is: > 1) Decide whether you want to stack 2, 3, etc. clients per tape. For > example, pick TWO. > 2) Then the number of TAPEPOOL tapes you wand to be in > FILLING status is > #clients/TWO. > 3) Count the number of tapes in TAPEPOOL that are already FULL. > 3) Set TAPEPOOL MAXSCRATCH to: DESIRED FILLING + ALREADY FULL. > > Check it one a week (or more often if you fill up a lot of > tapes daily) > and add more to MAXSCRATCH as tapes get full. > > Regarding COPYPOOL, I don't see any reason to collocate that at all. > > Wanda Prather > "I/O, I/O, It's all about I/O" -(me) > > > > > > > > -----Original Message- > From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] > On Behalf Of > Roark Ludwig > Sent: Thursday, August 25, 2005 7:56 PM > To: ADSM-L@VM.MARIST.EDU > Subject: Re: Question about Space Reclamation.. > > > I have been handed a TSM system in the past week that has been running > for > a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this > system > is running TSM 5.2.2. > > We have run out of tapes as the setting for collocation (yes) > is trying > to > use a tape for each node (as I understand it, ,,please correct me if I > am > wrong).. > > We have decided that since we have about 80 volumes with less then 1% > utilized and 20 or so with large utilization %'s to set COLLOCATION=NO > for > the TAPEPOOL and COPYPOOLS. > > We are expecting to add another set of nodes and don't wish to consume > more > of our volumes (with small percent utilization) as we add the nodes. > > Question .. > > Will this setting COLLOCATION=NO stop the addition of volumes? (I > expect > the answer is YES,, please correct me if I am wrong.) > > Now to the second question. > > I see no easy way to have Space Reclamation condense the volumes > included > in the two pools as it will only process "FULL" volumes. (Again Please > jump > in here). > > QUESTION: is it acceptable to set the status from "FILLING" to "FULL" > for > the volumes with low percentage utilized to force Space Reclamation ? > > OR > > is there an easier way to accomplish the GOAL of reducing the > number of > volumes needed for the pools?(given that we have set > COLLOCATION=NO) > or should we simply wait out the natural filling of the volumes and > allow > normal Space Reclamation as time proceeds? > > > Has anyone done this in the past? > > How have others dealt with this?? > > > Thanks for any input.
Re: Question about Space Reclamation..
==> On Fri, 26 Aug 2005 21:33:58 -0400, David E Ehresman <[EMAIL PROTECTED]> said: > Uhh, how about for restore performance when it matters most, during a > disaster! I come down on the other side of this, myself. Biggest reason is, you just can't afford to collocate the offsites. Imagine sending one tape per node offsite, every day. If you've got some sort of long-haul fiber solution, and can expose the physical volume granularity at the remote site to your local server, then it would be very useful to collocate. So in your site-wide disaster, you are going to have a mishmash anyway. On the other hand, the media-failure case is much more common. In those cases, I find the noncollocated copy pool to be a reasonable price to pay for the savings in slots and tapes. All these calculations might be different with collocation groups; hopefully in a week or two I'll be considering that as a concrete problem. (chortle) - Allen S. Rout
Re: Question about Space Reclamation..
>Regarding COPYPOOL, I don't see any reason to collocate that at all. Uhh, how about for restore performance when it matters most, during a disaster! David
Re: Question about Space Reclamation..
There is a middle ground here. If you set COLLOCATION=NO, it will drastically affect your RESTORE performance. Over time, you could in theory have to mount EVERY TAPE to do a full restore of a client. That is why collocation is a good thing. But collocation will, by default, put each client on its own tape, thereby using a LOT of tapes. The alternative is to control collocation by setting the MAXSCRATCH value for your TAPEPOOL. Assume you are starting out with a new TSM server; if you have 80 clients and set MAXSCRATCH to 40, TSM will stack 2 clients per tape. This is a good thing, too. You get 90% of the benefits of collocation, with only half the tapes. This is a very common thing to do; there are plenty of sites that have more clients than slots in their library! The drawback is that you have to check on the value of MAXSCRATCH often. As you get more tapes in FULL status (assuming you are like the rest of us and your site is CONSTANTLY growing its storage farm), you have to keep increasing the value of MAXSCRATCH. That's easy: update stgpool TAPEPOOL maxscratch=xxx. Or do it from the admin GUI. What I do is: 1) Decide whether you want to stack 2, 3, etc. clients per tape. For example, pick TWO. 2) Then the number of TAPEPOOL tapes you wand to be in FILLING status is #clients/TWO. 3) Count the number of tapes in TAPEPOOL that are already FULL. 3) Set TAPEPOOL MAXSCRATCH to: DESIRED FILLING + ALREADY FULL. Check it one a week (or more often if you fill up a lot of tapes daily) and add more to MAXSCRATCH as tapes get full. Regarding COPYPOOL, I don't see any reason to collocate that at all. Wanda Prather "I/O, I/O, It's all about I/O" -(me) -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Roark Ludwig Sent: Thursday, August 25, 2005 7:56 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: Question about Space Reclamation.. I have been handed a TSM system in the past week that has been running for a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this system is running TSM 5.2.2. We have run out of tapes as the setting for collocation (yes) is trying to use a tape for each node (as I understand it, ,,please correct me if I am wrong).. We have decided that since we have about 80 volumes with less then 1% utilized and 20 or so with large utilization %'s to set COLLOCATION=NO for the TAPEPOOL and COPYPOOLS. We are expecting to add another set of nodes and don't wish to consume more of our volumes (with small percent utilization) as we add the nodes. Question .. Will this setting COLLOCATION=NO stop the addition of volumes? (I expect the answer is YES,, please correct me if I am wrong.) Now to the second question. I see no easy way to have Space Reclamation condense the volumes included in the two pools as it will only process "FULL" volumes. (Again Please jump in here). QUESTION: is it acceptable to set the status from "FILLING" to "FULL" for the volumes with low percentage utilized to force Space Reclamation ? OR is there an easier way to accomplish the GOAL of reducing the number of volumes needed for the pools?(given that we have set COLLOCATION=NO) or should we simply wait out the natural filling of the volumes and allow normal Space Reclamation as time proceeds? Has anyone done this in the past? How have others dealt with this?? Thanks for any input.
Re: Question about Space Reclamation..
Why colloocate the offsite tapes? Original message >Date: Fri, 26 Aug 2005 08:26:44 +0200 >From: "Maibaum, Volker" <[EMAIL PROTECTED]> >Subject: Re: [ADSM-L] Question about Space Reclamation.. >To: ADSM-L@VM.MARIST.EDU > >Hello, > >one reason why I updated to version 5.3 was that there is the >possibility to create collocation groups. > >I have now packed all small servers (about 20 of them) in one group, >about 10 medium servers in another group and I made groups for large >servers with about 1-3 nodes each. So now no server is spread across >more than 1-4 tapes (except for the really large ones) and the tapes >have a good utilization. > >One problem still existing is that for each collocation group (we have 6 >of them) there is one tape send to the offsite location each day (we >have collocation=group turned on the copy pool). And as we do offsite >reclamation only on the weekend we have 5*6=30 tapes that are very >little utilized at the end of the week. But I think there is no way >around this. > >I'm not sure if this might work, but couldn't you just use the "move >data STGPOOL_NAME"-command to move the data from your 1% utilized >volumes? > >regards, > > >Volker Maibaum >Tel: +49 711 939 0972 >Fax: +49 711 939 160972 >[EMAIL PROTECTED] > > >Am Donnerstag, den 25.08.2005, 16:56 -0700 schrieb Roark Ludwig: >> I have been handed a TSM system in the past week that has been running for >> a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this system >> is running TSM 5.2.2. >> >> We have run out of tapes as the setting for collocation (yes) is trying to >> use a tape for each node (as I understand it, ,,please correct me if I am >> wrong).. >> >> We have decided that since we have about 80 volumes with less then 1% >> utilized and 20 or so with large utilization %'s to set COLLOCATION=NO for >> the TAPEPOOL and COPYPOOLS. >> >> We are expecting to add another set of nodes and don't wish to consume more >> of our volumes (with small percent utilization) as we add the nodes. >> >> Question .. >> >> Will this setting COLLOCATION=NO stop the addition of volumes? (I expect >> the answer is YES,, please correct me if I am wrong.) >> >> Now to the second question. >> >> I see no easy way to have Space Reclamation condense the volumes included >> in the two pools as it will only process "FULL" volumes. (Again Please jump >> in here). >> >> QUESTION: is it acceptable to set the status from "FILLING" to "FULL" for >> the volumes with low percentage utilized to force Space Reclamation ? >> >> OR >> >> is there an easier way to accomplish the GOAL of reducing the number of >> volumes needed for the pools?(given that we have set COLLOCATION=NO) >> or should we simply wait out the natural filling of the volumes and allow >> normal Space Reclamation as time proceeds? >> >> >> Has anyone done this in the past? >> >> How have others dealt with this?? >> >> >> Thanks for any input. >> Fred Johanson
Re: Question about Space Reclamation..
Hello, one reason why I updated to version 5.3 was that there is the possibility to create collocation groups. I have now packed all small servers (about 20 of them) in one group, about 10 medium servers in another group and I made groups for large servers with about 1-3 nodes each. So now no server is spread across more than 1-4 tapes (except for the really large ones) and the tapes have a good utilization. One problem still existing is that for each collocation group (we have 6 of them) there is one tape send to the offsite location each day (we have collocation=group turned on the copy pool). And as we do offsite reclamation only on the weekend we have 5*6=30 tapes that are very little utilized at the end of the week. But I think there is no way around this. I'm not sure if this might work, but couldn't you just use the "move data STGPOOL_NAME"-command to move the data from your 1% utilized volumes? regards, Volker Maibaum Tel: +49 711 939 0972 Fax: +49 711 939 160972 [EMAIL PROTECTED] Am Donnerstag, den 25.08.2005, 16:56 -0700 schrieb Roark Ludwig: > I have been handed a TSM system in the past week that has been running for > a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this system > is running TSM 5.2.2. > > We have run out of tapes as the setting for collocation (yes) is trying to > use a tape for each node (as I understand it, ,,please correct me if I am > wrong).. > > We have decided that since we have about 80 volumes with less then 1% > utilized and 20 or so with large utilization %'s to set COLLOCATION=NO for > the TAPEPOOL and COPYPOOLS. > > We are expecting to add another set of nodes and don't wish to consume more > of our volumes (with small percent utilization) as we add the nodes. > > Question .. > > Will this setting COLLOCATION=NO stop the addition of volumes? (I expect > the answer is YES,, please correct me if I am wrong.) > > Now to the second question. > > I see no easy way to have Space Reclamation condense the volumes included > in the two pools as it will only process "FULL" volumes. (Again Please jump > in here). > > QUESTION: is it acceptable to set the status from "FILLING" to "FULL" for > the volumes with low percentage utilized to force Space Reclamation ? > > OR > > is there an easier way to accomplish the GOAL of reducing the number of > volumes needed for the pools?(given that we have set COLLOCATION=NO) > or should we simply wait out the natural filling of the volumes and allow > normal Space Reclamation as time proceeds? > > > Has anyone done this in the past? > > How have others dealt with this?? > > > Thanks for any input. >
Re: Question about Space Reclamation..
I have been handed a TSM system in the past week that has been running for a year. IT has collocation = yes for TAPEPOOL and COPYPOOLS. this system is running TSM 5.2.2. We have run out of tapes as the setting for collocation (yes) is trying to use a tape for each node (as I understand it, ,,please correct me if I am wrong).. We have decided that since we have about 80 volumes with less then 1% utilized and 20 or so with large utilization %'s to set COLLOCATION=NO for the TAPEPOOL and COPYPOOLS. We are expecting to add another set of nodes and don't wish to consume more of our volumes (with small percent utilization) as we add the nodes. Question .. Will this setting COLLOCATION=NO stop the addition of volumes? (I expect the answer is YES,, please correct me if I am wrong.) Now to the second question. I see no easy way to have Space Reclamation condense the volumes included in the two pools as it will only process "FULL" volumes. (Again Please jump in here). QUESTION: is it acceptable to set the status from "FILLING" to "FULL" for the volumes with low percentage utilized to force Space Reclamation ? OR is there an easier way to accomplish the GOAL of reducing the number of volumes needed for the pools?(given that we have set COLLOCATION=NO) or should we simply wait out the natural filling of the volumes and allow normal Space Reclamation as time proceeds? Has anyone done this in the past? How have others dealt with this?? Thanks for any input.
Re: Newbie question about Space Reclamation
You did not provided version and OS of the server. Can you provide also
some information about your library - vendor, model, number of drives,
types of drives (it is possible the library is brand new with LTO2 or
upgraded with mixed LTO/DLT). Output of some commands may show the
situation in detail:
- q dri
- q path
- q v 476afxl1 f=d
- q v 505afxl1 f=d
- q stg backup_stg_pool f=d
- q stg f=d
- q dev ltoclass1 f=d
Plus any obvious error messages in the actlog. Is there any reason for
those 564 unreadable files?
Zlatko Krastev
IT Consultant
Stephen Comiskey <[EMAIL PROTECTED]>
Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
19.02.2003 18:18
Please respond to "ADSM: Dist Stor Manager"
To: [EMAIL PROTECTED]
cc:
Subject: Newbie question about Space Reclamation
Hi,
I'm only new to TSM and on a bit of a steep learning curve.
We have a scheduled job each Friday to reclaim our copy pool tapes (update
stg backup_copy_pool reclaim=75), the process starts and the output volume
is mounted OK however the server wait for the input (60K+ seconds). I've
pasted some output from the console below...
q proc
Process Description Status
-
Space ReclamationOffsite Volume(s) (storage pool
BACKUP_COPY_POOL), Moved Files: 0, Moved Bytes:
0, Unreadable Files: 564, Unreadable Bytes:
10,214,383. Current Physical File (bytes):
180,031,126
Waiting for mount of input volume 476AFXL1 (821
seconds).
Current output volume: 505AFXL1.
q mount
ANR8379I Mount point in device class LTOCLASS1 is waiting for the volume
mount to complete, status:
WAITING FOR VOLUME.
ANR8330I LTO volume 505AFXL1 is mounted R/W in drive LTO_DRIVE1 (\\.
\TAPE0), status: IN USE.
ANR8334I 2 matches found.
I know I'm missing something real simple here, I've search a number of
forums (including this lists archives) but I may be entering the wrong
search parameters
Any assistance you can provide is much appreciated.
cheers
Stephen
Re: Newbie question about Space Reclamation
From: Stephen Comiskey [mailto:[EMAIL PROTECTED]] > We have a scheduled job each Friday to reclaim our copy pool > tapes (update stg backup_copy_pool reclaim=75), the process > starts and the output volume is mounted OK however the server > wait for the input (60K+ seconds). I've pasted some output > from the console below... Please refer to the answer to question 04-17, posted in the TSM monthly FAQ, posted around the first of each month. Look for it in http://search.adsm.org. -- Mark Stapleton ([EMAIL PROTECTED])
Re: Newbie question about Space Reclamation
>I did a "q vol 476afxl1 f=d" and access=Read/Write, the tape is in the >library and there are no requests outstanding and Mount Limit = Drives :-( I'm chuckling a bit to myself, thinking of the old joke of how you discern a software person: That person will indefinitely look at command display indications in pondering the condition of a device. There are times when you need to just go look at the darned drive and see what its panel says about its condition. Never overlook the obvious: I can't tell you how many times in my career I've seen quite intelligent people obsess over erudite factors because they could be more intriguing than reality. Time is a limited resource. Start by looking in the areas which will most likely yield the most immediate results, and dig into laborious detail only as it becomes necessary. Richard Sims, BU
Re: Newbie question about Space Reclamation
In looking at the output of the "q mount" I see the ~waiting~ and remembered... often if a tape encounters errors upon being mounted, recovery work will be performed on the tape automatically (triggered by the drive) and it will show ~waiting on mount~ until the drive gets through with it (or at least with 3590's I've seen this) It might be that the tape is actually in a drive but hasn't ~officially~ been mounted yet. I don't know if LTO tapes have all the VCR information on them like a 3590 does but if so, you might be seeing this mount pause while the system attempts to fix problems that the drive has discovered... Dwight -Original Message- From: Stephen Comiskey [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 19, 2003 11:13 AM To: [EMAIL PROTECTED] Subject: Re: Newbie question about Space Reclamation Dwight, Thanks for the pointers... I did a "q vol 476afxl1 f=d" and access=Read/Write, the tape is in the library and there are no requests outstanding and Mount Limit = Drives :-( My head is starting to hurt !! Stephen |-+---> | | "Cook, Dwight E"| | | | | | | | | 19/02/03 16:47 | | | | |-+---> >--- | | | |To: Stephen Comiskey/Dublin/IE/RoyalSun@RoyalSunInt | | cc: | | Subject: RE: Newbie question about Space Reclamation | >--- | THIS MESSAGE ORIGINATED ON THE INTERNET - Please read the detailed disclaimer below. --- check : availability of the tape 476AFXL1 (q vol 476AFXL1 f=d) make sure it is available ie. acc=readwrite availability of the tape within the atl (q libvol * 476AFXL1) make sure it is checked in... try a "query request" because if a tape needs to be checked into a library a "request" will be issued availability of your drives (q drive) make sure they are all on-line and OK... mount limit associated with the device class used by these tapes (q dev f=d) look for "Mount Limit: DRIVES" OK, drives states limit is equal to available drives could have a hard number... try those things first... Dwight -Original Message- From: Stephen Comiskey [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 19, 2003 10:19 AM To: [EMAIL PROTECTED] Subject: Newbie question about Space Reclamation Hi, I'm only new to TSM and on a bit of a steep learning curve. We have a scheduled job each Friday to reclaim our copy pool tapes (update stg backup_copy_pool reclaim=75), the process starts and the output volume is mounted OK however the server wait for the input (60K+ seconds). I've pasted some output from the console below... q proc Process Description Status - Space ReclamationOffsite Volume(s) (storage pool BACKUP_COPY_POOL), Moved Files: 0, Moved Bytes: 0, Unreadable Files: 564, Unreadable Bytes: 10,214,383. Current Physical File (bytes): 180,031,126 Waiting for mount of input volume 476AFXL1 (821 seconds). Current output volume: 505AFXL1. q mount ANR8379I Mount point in device class LTOCLASS1 is waiting for the volume mount to complete, status: WAITING FOR VOLUME. ANR8330I LTO volume 505AFXL1 is mounted R/W in drive LTO_DRIVE1 (\\. \TAPE0), status: IN USE. ANR8334I 2 matches found. I know I'm missing something real simple here, I've search a number of forums (including this lists archives) but I may be entering the wrong search parameters Any assistance you can provide is much appreciated. cheers Stephen --- The following message has been automatically added by the mail gateway to comply with a Royal & SunAlliance IT Security requirement. As this email arrived via the Internet you should be cautious about its origin and content. Replies which contain sensitive information or legal/contractual obligations are particularly vulnerable. In these cases you should not reply unless you are authorised to do so, and adequate encryption is employed. If you have any questions, please speak to your local desktop support team or IT security contact. ---
Re: Newbie question about Space Reclamation
Dwight, Thanks for the pointers... I did a "q vol 476afxl1 f=d" and access=Read/Write, the tape is in the library and there are no requests outstanding and Mount Limit = Drives :-( My head is starting to hurt !! Stephen |-+---> | | "Cook, Dwight E"| | | | | | | | | 19/02/03 16:47 | | | | |-+---> >---| | | |To: Stephen Comiskey/Dublin/IE/RoyalSun@RoyalSunInt | |cc: | | Subject: RE: Newbie question about Space Reclamation | >---| THIS MESSAGE ORIGINATED ON THE INTERNET - Please read the detailed disclaimer below. --- check : availability of the tape 476AFXL1 (q vol 476AFXL1 f=d) make sure it is available ie. acc=readwrite availability of the tape within the atl (q libvol * 476AFXL1) make sure it is checked in... try a "query request" because if a tape needs to be checked into a library a "request" will be issued availability of your drives (q drive) make sure they are all on-line and OK... mount limit associated with the device class used by these tapes (q dev f=d) look for "Mount Limit: DRIVES" OK, drives states limit is equal to available drives could have a hard number... try those things first... Dwight -Original Message- From: Stephen Comiskey [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 19, 2003 10:19 AM To: [EMAIL PROTECTED] Subject: Newbie question about Space Reclamation Hi, I'm only new to TSM and on a bit of a steep learning curve. We have a scheduled job each Friday to reclaim our copy pool tapes (update stg backup_copy_pool reclaim=75), the process starts and the output volume is mounted OK however the server wait for the input (60K+ seconds). I've pasted some output from the console below... q proc Process Description Status - Space ReclamationOffsite Volume(s) (storage pool BACKUP_COPY_POOL), Moved Files: 0, Moved Bytes: 0, Unreadable Files: 564, Unreadable Bytes: 10,214,383. Current Physical File (bytes): 180,031,126 Waiting for mount of input volume 476AFXL1 (821 seconds). Current output volume: 505AFXL1. q mount ANR8379I Mount point in device class LTOCLASS1 is waiting for the volume mount to complete, status: WAITING FOR VOLUME. ANR8330I LTO volume 505AFXL1 is mounted R/W in drive LTO_DRIVE1 (\\. \TAPE0), status: IN USE. ANR8334I 2 matches found. I know I'm missing something real simple here, I've search a number of forums (including this lists archives) but I may be entering the wrong search parameters Any assistance you can provide is much appreciated. cheers Stephen --- The following message has been automatically added by the mail gateway to comply with a Royal & SunAlliance IT Security requirement. As this email arrived via the Internet you should be cautious about its origin and content. Replies which contain sensitive information or legal/contractual obligations are particularly vulnerable. In these cases you should not reply unless you are authorised to do so, and adequate encryption is employed. If you have any questions, please speak to your local desktop support team or IT security contact. ---
Re: Newbie question about Space Reclamation
With the infor you've included, I would sya you should do a "q request". It is probably requesting a volume that is out of the library. I would sya a restore could be using that volume, but that would show up in q mount. Other possibility is there is a problem with that atpe and it won't mount. In that case do a "q actlog" and see if any errors about this tape. David Longo >>> [EMAIL PROTECTED] 02/19/03 11:18AM >>> Hi, I'm only new to TSM and on a bit of a steep learning curve. We have a scheduled job each Friday to reclaim our copy pool tapes (update stg backup_copy_pool reclaim=75), the process starts and the output volume is mounted OK however the server wait for the input (60K+ seconds). I've pasted some output from the console below... q proc Process Description Status - Space ReclamationOffsite Volume(s) (storage pool BACKUP_COPY_POOL), Moved Files: 0, Moved Bytes: 0, Unreadable Files: 564, Unreadable Bytes: 10,214,383. Current Physical File (bytes): 180,031,126 Waiting for mount of input volume 476AFXL1 (821 seconds). Current output volume: 505AFXL1. q mount ANR8379I Mount point in device class LTOCLASS1 is waiting for the volume mount to complete, status: WAITING FOR VOLUME. ANR8330I LTO volume 505AFXL1 is mounted R/W in drive LTO_DRIVE1 (\\. \TAPE0), status: IN USE. ANR8334I 2 matches found. I know I'm missing something real simple here, I've search a number of forums (including this lists archives) but I may be entering the wrong search parameters Any assistance you can provide is much appreciated. cheers Stephen "MMS " made the following annotations on 02/19/2003 11:46:32 AM -- This message is for the named person's use only. It may contain confidential, proprietary, or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it, and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Health First reserves the right to monitor all e-mail communications through its networks. Any views or opinions expressed in this message are solely those of the individual sender, except (1) where the message states such views or opinions are on behalf of a particular entity; and (2) the sender is authorized by the entity to give such views or opinions. ==
Newbie question about Space Reclamation
Hi, I'm only new to TSM and on a bit of a steep learning curve. We have a scheduled job each Friday to reclaim our copy pool tapes (update stg backup_copy_pool reclaim=75), the process starts and the output volume is mounted OK however the server wait for the input (60K+ seconds). I've pasted some output from the console below... q proc Process Description Status - Space ReclamationOffsite Volume(s) (storage pool BACKUP_COPY_POOL), Moved Files: 0, Moved Bytes: 0, Unreadable Files: 564, Unreadable Bytes: 10,214,383. Current Physical File (bytes): 180,031,126 Waiting for mount of input volume 476AFXL1 (821 seconds). Current output volume: 505AFXL1. q mount ANR8379I Mount point in device class LTOCLASS1 is waiting for the volume mount to complete, status: WAITING FOR VOLUME. ANR8330I LTO volume 505AFXL1 is mounted R/W in drive LTO_DRIVE1 (\\. \TAPE0), status: IN USE. ANR8334I 2 matches found. I know I'm missing something real simple here, I've search a number of forums (including this lists archives) but I may be entering the wrong search parameters Any assistance you can provide is much appreciated. cheers Stephen
