- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- Re: Internet production server
Categories
Company
Local Language
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
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
Community
Resources
Forums
Blogs
- 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
05-23-2001 07:20 AM
05-23-2001 07:20 AM
The databases hosted on this L1000 are less than 1Gb in size. Httpd requests are handled by a K420 across the firewall which then retrieves data from the L1000.
Performance has been reported as being slow on the L1000 and I have been monitoring and patching whatever daemons appear to create memory overflows, ie mib2agt.
The dba runs xoffice(GroupWise), Wordperfect, Netscape and occasionally xomni on the L1000. I want to remove these applications to ensure all cpu cycles are dedicated to the production server.
The dba argues with me that the above mentioned applications do not have an impact on the performance of the server and that the production server is his home base.
Am I being overly pro-active? I welcome any feedback.
TIA
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-23-2001 07:35 AM
05-23-2001 07:35 AM
Re: Internet production server
I would add another 512 meg memory and another processor depending on how many users are on at any one time.
To see what is taking the cpu, try using top for at least a few minutes and you might want to watch it for an hour during the busy part of the day. If you have glance, that is a much better product to use.
Give us more stats to go on, and the answers will be better also.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-23-2001 08:26 AM
05-23-2001 08:26 AM
Re: Internet production server
In general, these applications take approximately 5% of the cpu, but I did notice Netscape going rogue occasionally and I had to kill the session.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-24-2001 09:36 AM
05-24-2001 09:36 AM
Re: Internet production server
L1000, 512Mb RAM, 1 360Mhz cpu, 64 bit o.s.
Use: Oracle database production server
Users: 15
Apps: xoffice(GW5), WP7, Netscape, OmnibackII, ArcInfo, SDE
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-24-2001 10:49 AM
05-24-2001 10:49 AM
SolutionRegardless of the availability of CPU cycles, the default HP-UX scheduler will tend to penalize applications that run for longer periods of time and that take a relatively higher percent of CPU cycles. Therfore I don't want anything to compete with the database, especially things that seem to be lightweight, since by default, the scheduler will give these transient, lightweight apps a more generous look at the CPU.
Given that CPU cycles are available, and that Oracle (from a process standpoint) doesn't seem to be getting killed by the scheduler, I'd
1) nail tables in memory
2) run a database diagnostic tool (like Quest's SQLab) to see what's going on with the database. We find that perfomance problems generally require database tunes much more often than hardware upgrades
3) check the firewall for a potential bottleneck. Since the DB is behind a firewall, make sure to compare SQL performance locally (on the database server) to performance through the firewall.