Showing posts with label OS Changes. Show all posts
Showing posts with label OS Changes. Show all posts
Tuesday, April 21, 2015
Linux OS: Opening raw device can slow down on certain kernels
Last week at COLLABORATE I talked about measuring your baselines so you can determine when a negative (or even positive) change has happened, and the MOS document "Open Raw Device is Slower with Kernel 2.6.32-358.el6.x86_64" from article ID 1999421.1 proves that we have a lot of measuring still to automate because this is a basic OS operation which needs investigation in order to find out when it gets off kilter.
Wednesday, February 25, 2015
OS: Missing backout data and unable to remove patch
Do you want to add more tools to your toolbox? How much do you know about OS patching? Would you like to know how to remove an OS patch? Read up on MOS document "Unable to remove patch due to missing backout data" from article ID 1984126.1 to learn the commands for backing out a patch, and how you might be able to salvage a bad situation!
Monday, October 27, 2014
Signal 17 and RHEL upgrades
This weekend we upgraded servers to Red Hat Enterprise Linux (RHEL) 5.11 but found quite a surprise when the DBA ran a simple ps command and received the following error message:
Signal 17 (CHLD) caught by ps (procps version 3.2.8).
What did we find out by searching the Interwebs in the middle of the night? A KB article from Red Hat (subscription required) that told us that the error is confirmed on version 6.6 and has a few simple solutions. Luckily our system administrators were able to implement one of them in order to return the ps command back to service!
Signal 17 (CHLD) caught by ps (procps version 3.2.8).
What did we find out by searching the Interwebs in the middle of the night? A KB article from Red Hat (subscription required) that told us that the error is confirmed on version 6.6 and has a few simple solutions. Luckily our system administrators were able to implement one of them in order to return the ps command back to service!
Saturday, October 11, 2014
Weekend Learning: R12: Solaris on SPARC OS updates required
Do you run EBS R12 with the Oracle Solaris OS on a SPARC platform? If so, do you know about changes coming in December requiring you to be above a specific release of Solaris in order to continue receiving EBS patches? True story!! Check out the My Oracle Support note "New Oracle E-Business Suite R12 Operating System and Tools Requirements for the Oracle Solaris on SPARC platform" via document 1912669.1 for all of the details!
Sunday, April 20, 2014
FYI: Oracle Hyperion Essbase 64 Bit Information
Are you thinking about installing Oracle Hyperion Essbase? Maybe thinking about upgrading your platform to a 64-bit OS client? Then you should visit the article "Oracle Hyperion Essbase and 64 Bit Master Note" in My Oracle Support note 1090635.1 to get information on APIs, ODBC drivers, and SmartView issues with an eye towards a 64-bit OS client.
Friday, August 9, 2013
Finding out more about your OS
Recently I had to go back though our servers to verify OS information, and these sites proved invaluable to that effort so I'll be adding them to the Frequently Used Links:
http://www.unixtutorial.org/2007/08/how-to-find-out-the-release-version-of-your-unix/ http://www.unixtutorial.org/commands/uname/
http://itsiti.com/windows-server-2003-determine-32-bit-or-64-bit-version
http://www.unixtutorial.org/2007/08/how-to-find-out-the-release-version-of-your-unix/ http://www.unixtutorial.org/commands/uname/
http://itsiti.com/windows-server-2003-determine-32-bit-or-64-bit-version
Thursday, August 2, 2012
OS Changes - Commands
So we've upgraded from SUN Solaris to RedHat Linux, and I've found some key differences so far. The command prstat no longer really works well to find out how healthy your server/CPUs are, as it needs to be combined with the new top command to get a fuller understanding of how healthy your system is and this site will help you understand why there is a difference between not only the commands but how the OS is operating as well.
To investigate printing issues I grew accustomed to the lpstat command combined with the -lp switches so that a command lpstat -lp PRINTERNAME would give me information as to how the printer was connected to our system and if the printer queue was up or not. I've found out that issuing this command in a RedHat Linux environment doesn't provide that old type of feedback, so we've had to alter our approach and instead use the -a switch so that the entire command looks like lpstat -a PRINTERNAME now.
To investigate printing issues I grew accustomed to the lpstat command combined with the -lp switches so that a command lpstat -lp PRINTERNAME would give me information as to how the printer was connected to our system and if the printer queue was up or not. I've found out that issuing this command in a RedHat Linux environment doesn't provide that old type of feedback, so we've had to alter our approach and instead use the -a switch so that the entire command looks like lpstat -a PRINTERNAME now.
Subscribe to:
Posts (Atom)