HPE GreenLake Administration
- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - Linux
- >
- Kernel Tuning w.r.t semaphores & shared memory
Operating System - Linux
1831806
Members
2437
Online
110029
Solutions
Forums
Categories
Company
Local Language
back
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Forums
Discussions
Discussions
Discussions
Forums
Discussions
back
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Blogs
Information
Community
Resources
Community Language
Language
Forums
Blogs
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-26-2003 09:29 PM
06-26-2003 09:29 PM
Kernel Tuning w.r.t semaphores & shared memory
Hi Guys,
System : - N-Class, hpux 11.00
We have approx. 15 db instances running on a N-Class server , all are with different versions eg. 8.1.7-32bit, 8.1.6-64bit, 9.2.0.2-32 , 9.2.0.2-64 , Yesterday we had have a serious problem where we have to brought down one database . After "shutdown abort" db was not going down and kill -9 was also not working for any of the oracle processes. I tried killing the semaphores & memory regions related to the particular instance with ipcrm but no luck. Atlast we have to reboot the server.
Now the question is why might this problem have aroused? Can we do any kernel tuning to take care it , to avoid this in future. Some of teh parameters which are of interest are
root@tbau01:>kmtune | grep -i shm
shmmax 2 GB
shmmni 512
shmseg 32
Can I tune this and to which extent , if yes. I am thinking to double the values of shmmax, shmmni, shmseg, Pl advise.
Thanx
System : - N-Class, hpux 11.00
We have approx. 15 db instances running on a N-Class server , all are with different versions eg. 8.1.7-32bit, 8.1.6-64bit, 9.2.0.2-32 , 9.2.0.2-64 , Yesterday we had have a serious problem where we have to brought down one database . After "shutdown abort" db was not going down and kill -9 was also not working for any of the oracle processes. I tried killing the semaphores & memory regions related to the particular instance with ipcrm but no luck. Atlast we have to reboot the server.
Now the question is why might this problem have aroused? Can we do any kernel tuning to take care it , to avoid this in future. Some of teh parameters which are of interest are
root@tbau01:>kmtune | grep -i shm
shmmax 2 GB
shmmni 512
shmseg 32
Can I tune this and to which extent , if yes. I am thinking to double the values of shmmax, shmmni, shmseg, Pl advise.
Thanx
Obstacles exist to challenge you to keep going. Not to quit.
2 REPLIES 2
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-30-2003 05:02 AM
06-30-2003 05:02 AM
Re: Kernel Tuning w.r.t semaphores & shared memory
To be honest, it looks like your configuration is like a lot of those with Oracle - quite complex!!
With such specifics, I would ask Oracle for their views on what your parameters should be.
I have some servers here that have been tuned similarly but it too is a very specific environment which may not truely represent yours.
Good luck, (sorry, perhaps not the informative response you might have hoped for).
With such specifics, I would ask Oracle for their views on what your parameters should be.
I have some servers here that have been tuned similarly but it too is a very specific environment which may not truely represent yours.
Good luck, (sorry, perhaps not the informative response you might have hoped for).
There's no substitute for experience
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-30-2003 06:43 AM
06-30-2003 06:43 AM
Re: Kernel Tuning w.r.t semaphores & shared memory
Few things to consider
shmmax will ignore settings greater than 25% of totoal memory.
Total memory is defined as physical memory plus swap.
Swap in general for systems with less than 16 GB of ram should be between 1.5 and 2 times physical memory.
shmseg might be a little low, but you need to collect some performance data to see if that is the issue.
To that end, I'm attaching a script. I also have some information from one of HP's prime oracle tuners and may want to collect additional data.
Script is attached....
With so many versions of Oracle running on the server, you may have a combination that Oracle never tested.
As a long term goal, I would suggest trying to migrate to all instances running in one central version in the 8 series, say 8.1.7.4.0 and one common version in the 9 series.
This could be an oracle problem.
itrc is giving me trouble accessing mhy Oracle Tuning documentation. I'll get back to you with it when itrc lets me get back to you with it.
SEP
shmmax will ignore settings greater than 25% of totoal memory.
Total memory is defined as physical memory plus swap.
Swap in general for systems with less than 16 GB of ram should be between 1.5 and 2 times physical memory.
shmseg might be a little low, but you need to collect some performance data to see if that is the issue.
To that end, I'm attaching a script. I also have some information from one of HP's prime oracle tuners and may want to collect additional data.
Script is attached....
With so many versions of Oracle running on the server, you may have a combination that Oracle never tested.
As a long term goal, I would suggest trying to migrate to all instances running in one central version in the 8 series, say 8.1.7.4.0 and one common version in the 9 series.
This could be an oracle problem.
itrc is giving me trouble accessing mhy Oracle Tuning documentation. I'll get back to you with it when itrc lets me get back to you with it.
SEP
Steven E Protter
Owner of ISN Corporation
http://isnamerica.com
http://hpuxconsulting.com
Sponsor: http://hpux.ws
Twitter: http://twitter.com/hpuxlinux
Founder http://newdatacloud.com
Owner of ISN Corporation
http://isnamerica.com
http://hpuxconsulting.com
Sponsor: http://hpux.ws
Twitter: http://twitter.com/hpuxlinux
Founder http://newdatacloud.com
The opinions expressed above are the personal opinions of the authors, not of Hewlett Packard Enterprise. By using this site, you accept the Terms of Use and Rules of Participation.
Company
Events and news
Customer resources
© Copyright 2025 Hewlett Packard Enterprise Development LP