Comments on: Debugger dangers http://blog.tanelpoder.com/2008/06/14/debugger-dangers/ Oracle troubleshooting, internals and performance tuning Mon, 15 Mar 2010 01:50:11 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: tanelp http://blog.tanelpoder.com/2008/06/14/debugger-dangers/comment-page-1/#comment-419 tanelp Wed, 18 Jun 2008 14:00:29 +0000 http://tanelpoder.wordpress.com/2008/06/14/debugger-dangers/#comment-419 Hi Ganesh, I don't think this a coincidence. I've used this kit for numerous long-running stress tests and have never hit this issue before. If it's a coincidence, then a pretty big one! Decent (and well configured) OSes do not just reject IOs unless there's some fundamental problem. Anyway I'll try reproduce this some time in future, will update if get any results. Hi Ganesh,

I don’t think this a coincidence. I’ve used this kit for numerous long-running stress tests and have never hit this issue before.

If it’s a coincidence, then a pretty big one! Decent (and well configured) OSes do not just reject IOs unless there’s some fundamental problem.

Anyway I’ll try reproduce this some time in future, will update if get any results.

]]> By: Ganesh http://blog.tanelpoder.com/2008/06/14/debugger-dangers/comment-page-1/#comment-418 Ganesh Tue, 17 Jun 2008 20:44:32 +0000 http://tanelpoder.wordpress.com/2008/06/14/debugger-dangers/#comment-418 Hello: I suspect that the File I/O error during your pstack trace could be a mere coincidence. I have seen this in the past at client locations. Though there are published bugs for this if running 10.2.0 3 or less w/ AIX , I have seen it in versions 10.2.0.3 with the O/s fully patched. The only explanation I can come up with is that the disk was too busy w/ many PX operations going on. (What? Why? How? .. I don't have an answer). As this was very difficult to reproduce I left it at that. HTH, Ganesh Hello:

I suspect that the File I/O error during your pstack trace could be a mere coincidence.

I have seen this in the past at client locations. Though there are published bugs for this if running 10.2.0 3 or less w/ AIX , I have seen it in versions 10.2.0.3 with the O/s fully patched.

The only explanation I can come up with is that the disk was too busy w/ many PX operations going on. (What? Why? How? .. I don’t have an answer).

As this was very difficult to reproduce I left it at that.

HTH,
Ganesh

]]> By: Advanced Oracle Troubleshooting Guide, Part 6: Understanding Oracle execution plans with os_explain « Tanel Poder’s blog: Core IT for geeks and pros http://blog.tanelpoder.com/2008/06/14/debugger-dangers/comment-page-1/#comment-417 Advanced Oracle Troubleshooting Guide, Part 6: Understanding Oracle execution plans with os_explain « Tanel Poder’s blog: Core IT for geeks and pros Sun, 15 Jun 2008 14:40:49 +0000 http://tanelpoder.wordpress.com/2008/06/14/debugger-dangers/#comment-417 [...] proceeding though, please read this post about safety of different stack sampling [...] [...] proceeding though, please read this post about safety of different stack sampling [...]

]]>