osam
ADSM.ORG Member
Hi all
Has somebody any information when/if IBM plans to support Redhat EL5?
Rgds
Mario
Has somebody any information when/if IBM plans to support Redhat EL5?
Rgds
Mario
-
Please be aware that ADSM.ORG will be going through forum software conversion and migration over the next several weeks. The cut-over date and time will be posted in the ADSM.ORG News & Announcements section as we near the target date. Please visit the following link for more details ADSM News & Announcements
-
Community Tip: Forum Rules (PLEASE CLICK HERE TO READ BEFORE POSTING)
Click the link above to access ADSM.ORG Acceptable Use Policy and forum rules which should be observed when using this website. Violators may be banned from this website. This notice will disappear after you have made at least 3 posts.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
TSM CLient for Redhat 5
- Thread starter osam
- Start date
chenao
ADSM.ORG Member
Heys. it is a question about memory use with TSM.
I always get the recomendation about increase buffer pool size and although I did it, I still get the message. I incresed RAM from 2 to 6 GB but nothing changed. I resetd the buffer pool before leaving but same situation. Any suggestion about which process could be doing this.
thanks
I always get the recomendation about increase buffer pool size and although I did it, I still get the message. I incresed RAM from 2 to 6 GB but nothing changed. I resetd the buffer pool before leaving but same situation. Any suggestion about which process could be doing this.
thanks
PJ
ADSM.ORG Senior Member
???
ahhhm...
Maybe it would help if your post contained anything pointing in the general direction of what on earth you're talking about?
PJ
ahhhm...
Maybe it would help if your post contained anything pointing in the general direction of what on earth you're talking about?
PJ
chenao
ADSM.ORG Member
TSM server memory use
I have a Linux TSM server
what I am talking about is the message I get in my backup report:
Issues an recommendations: Database cache Hit Ratio is too low - condition 96<98 . Recommedation: Increase bufferpool size value.
I modified the dsmserv.opt file but still I get the same error, for sure is not for the memory size because the server has 6 GB and I use it only to run the backup. I have 7 clients (some of them as test, because planning to have 54).
If I run the free command it is the result
total used free shared buffers cached
Mem: 6231020 6212472 18548 0 19524 4933964
-/+ buffers/cache: 1258984 4972036
Swap: 2031608 6044 2025564
Any Idea is appreciated.
Thanks
I have a Linux TSM server
what I am talking about is the message I get in my backup report:
Issues an recommendations: Database cache Hit Ratio is too low - condition 96<98 . Recommedation: Increase bufferpool size value.
I modified the dsmserv.opt file but still I get the same error, for sure is not for the memory size because the server has 6 GB and I use it only to run the backup. I have 7 clients (some of them as test, because planning to have 54).
If I run the free command it is the result
total used free shared buffers cached
Mem: 6231020 6212472 18548 0 19524 4933964
-/+ buffers/cache: 1258984 4972036
Swap: 2031608 6044 2025564
Any Idea is appreciated.
Thanks
PJ
ADSM.ORG Senior Member
Its not really an error but rather a hint that the low cache hit rate may degrade performance. Blow up the buffpool to 2GB (you can do so online via the "setopt") and if it doesn't improve, consider splitting the DB. If you don't have a performance problem you might as well just forget about it and remove the annoying entry from reporting.
PJ
PJ
chenao
ADSM.ORG Member
Thanks I will work on it. Now I have one more issue .. it is about tapes use.. why even though some tapes are not full(48% or less) TSM marks them as full and takes a new tape to use 1% or so. I know I can move the data with the reclamation process and reuse the tape, but any reason it works like that, one more question any idea what is the life cycle for an IBM LTO ultrium 3 tape
PJ
ADSM.ORG Senior Member
TSM marks a sequential container like a tape as "full" as soon as it hits the end of it while writing. After that, data will expire and the percentage will decrease over time until it hits the reclaim threshold and will be recycled. You decide how much a Tape should hold. If you thought when creating the device class that a LTO3 Tape will hold 800GB and in reality it only holds 400 because the data is not comressable (maybe is already compressed or whatever) you will never see anything close to 100% until you reach "full". Reduce the estimated capacity in the deviceclass to 400GB to get more realistic percentages on filling tapes. Also keep in mind that data may already expire "off" a tape while it is still filling up. Say you backup a 10GB Database in 10 versions and it always goes to the sme tape, then that tape will never hold more than 100GB of valid data. So if its a 1TB tape it will be 10% utilized when it hits the EOT and becomes "full" (extreme example).
Eventually you'll find out that these values don't really matter that much in daily operations. Usually tape handling and reclamation just sort of... work.
I don't know about LTO specs but I think it is something between 5000 and 10000 cycles depending on the manufacturer. All it means is: if it breaks before that, you'll eventually get your 100 bucks back. Not much of a consolation when that tape contained precious data. Keeping copies is a good idea anyway.
PJ
Eventually you'll find out that these values don't really matter that much in daily operations. Usually tape handling and reclamation just sort of... work.
I don't know about LTO specs but I think it is something between 5000 and 10000 cycles depending on the manufacturer. All it means is: if it breaks before that, you'll eventually get your 100 bucks back. Not much of a consolation when that tape contained precious data. Keeping copies is a good idea anyway.
PJ
Last edited:
chenao
ADSM.ORG Member
Tape use
Thanks PJ for your reply.
Now... I change the device class advance option to 400 GB (before was blank) and I don't know why after removing some tapes, running the audit command, it is doing nothing, so I run a query and the library did not change the information, it seems the audit command is not running. I run the checkincommand and it doesn't do nothing. I noticed that after changing the device class option. Now I try to keep it blank but it does not let me do it. I am not sure if it is related with the change I did or is a different issue that showed up just after this.... Any idea what is going on.
Thanks.
I made the decision to reboot the server and then I run the commands and the worked. Any Idea what happened.. is this about any process (i did not find any) running that did not allow tu execute the commands? ....
Thanks PJ for your reply.
Now... I change the device class advance option to 400 GB (before was blank) and I don't know why after removing some tapes, running the audit command, it is doing nothing, so I run a query and the library did not change the information, it seems the audit command is not running. I run the checkincommand and it doesn't do nothing. I noticed that after changing the device class option. Now I try to keep it blank but it does not let me do it. I am not sure if it is related with the change I did or is a different issue that showed up just after this.... Any idea what is going on.
Thanks.
I made the decision to reboot the server and then I run the commands and the worked. Any Idea what happened.. is this about any process (i did not find any) running that did not allow tu execute the commands? ....
Last edited:
PJ
ADSM.ORG Senior Member
As I said - the estimated capacity is only valid for filling tapes. Once TSM reached an end of a container, it knows exactly how much data it could squeze on it and won't care about estimates. I don't know what caused the hangup with your library. I just changed the estcap on one of my devclasses and it didn't lock anything. If it locks for whatever reason, recycling the TSM server usually helps because that reinitializes the library. Frankly I never really cared about what TSM thought it might be able to write to a tape. Any prediction would be wrong anyway - unless you work with precomressed data.
PJ
PJ
