Comments for OraInternals http://www.orainternals.com A company specializing in RAC, Performance, and E-business suite Tue, 24 Jan 2012 23:14:29 +0000 hourly 1 http://wordpress.org/?v=3.3 Comment on SCN – What, why, and how? by Kevin http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-390 Kevin Tue, 24 Jan 2012 23:14:29 +0000 http://www.orainternals.com/?p=1268#comment-390 Regarding the code, yes. But could only commit so much to memory. Having access (periodic) to Oracle code over 20+ years does allow a lot to sink in though. I historically only cared about things that happen with nose-bleed frequency. As a platform guy that and bandwidth are all that matters. Bandwidth naturally improves due to simple HW evolution. Critical code, on the other hand, is a software problem that victimizes hardware. In short, spin-locks suck. Now, all this aside, I'd like to reiterate that my words regarding SCN internals is knowledge obtained by anyone with a debugger and 5 minutes of time. In other words, Oracle Legal, if you are reading this you have too much time on your hands so don't bother sending me any "fan mail." Enough is enough. Regarding the code, yes. But could only commit so much to memory. Having access (periodic) to Oracle code over 20+ years does allow a lot to sink in though. I historically only cared about things that happen with nose-bleed frequency. As a platform guy that and bandwidth are all that matters. Bandwidth naturally improves due to simple HW evolution. Critical code, on the other hand, is a software problem that victimizes hardware. In short, spin-locks suck.

Now, all this aside, I’d like to reiterate that my words regarding SCN internals is knowledge obtained by anyone with a debugger and 5 minutes of time.

In other words, Oracle Legal, if you are reading this you have too much time on your hands so don’t bother sending me any “fan mail.” Enough is enough.

]]> Comment on SCN – What, why, and how? by admin http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-389 admin Tue, 24 Jan 2012 20:15:39 +0000 http://www.orainternals.com/?p=1268#comment-389 Hi Kevin Thank you for reading my blog and Thanks for your valuable comment. Of course, you would know the struct details of this code :) Cheers Riyaj Hi Kevin
Thank you for reading my blog and Thanks for your valuable comment. Of course, you would know the struct details of this code :)
Cheers
Riyaj

]]> Comment on SCN – What, why, and how? by Kevin http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-388 Kevin Tue, 24 Jan 2012 19:05:45 +0000 http://www.orainternals.com/?p=1268#comment-388 "SCN is a huge number with two components to it: Base and wrap" ...Hi Riyaj, Actually, the SCN is represented as two separate words in a structure. They are organized to "feel" like a huge number but they are two words, one is 16b as you say and the other is 32b. This information is a) obtainable via any debugger and b) doesn't change anything about what you wrote :-) Many years ago while working in the Oracle port team at Sequent we implemented a latch-free SCN using atomic 8byte pentium instructions. If the organization of the SCN wasn't as per what I wrote such an approach would not be possible because there would be no way to ensure the two words are adjacent in memory. Trivial pursuit I suppose. “SCN is a huge number with two components to it: Base and wrap”

…Hi Riyaj,

Actually, the SCN is represented as two separate words in a structure. They are organized to “feel” like a huge number but they are two words, one is 16b as you say and the other is 32b. This information is a) obtainable via any debugger and b) doesn’t change anything about what you wrote :-)

Many years ago while working in the Oracle port team at Sequent we implemented a latch-free SCN using atomic 8byte pentium instructions. If the organization of the SCN wasn’t as per what I wrote such an approach would not be possible because there would be no way to ensure the two words are adjacent in memory.

Trivial pursuit I suppose.

]]> Comment on SCN – What, why, and how? by The Oracle flaw: Clarifications and more information - TECHNOLOGY GADGETS – TECHNOLOGY GADGETS http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-386 The Oracle flaw: Clarifications and more information - TECHNOLOGY GADGETS – TECHNOLOGY GADGETS Tue, 24 Jan 2012 07:12:10 +0000 http://www.orainternals.com/?p=1268#comment-386 [...] of a database services association OraInterals, has zeroed in on another aspect of a smirch in a Jan. 20 post entitled “SCN — What, why, and how?” Shamsudeen records that he had hold off posting a blog entrance “for many [...] [...] of a database services association OraInterals, has zeroed in on another aspect of a smirch in a Jan. 20 post entitled “SCN — What, why, and how?” Shamsudeen records that he had hold off posting a blog entrance “for many [...]

]]> Comment on SCN – What, why, and how? by The Oracle flaw: Clarifications and more information - Security news - Tech Around World http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-385 The Oracle flaw: Clarifications and more information - Security news - Tech Around World Tue, 24 Jan 2012 03:14:22 +0000 http://www.orainternals.com/?p=1268#comment-385 [...] of the database services company OraInterals, has zeroed in on another aspect of the flaw in a Jan. 20 post entitled “SCN — What, why, and how?” Shamsudeen notes that he had held off posting the blog entry “for many months.” [...] [...] of the database services company OraInterals, has zeroed in on another aspect of the flaw in a Jan. 20 post entitled “SCN — What, why, and how?” Shamsudeen notes that he had held off posting the blog entry “for many months.” [...]

]]> Comment on SCN – What, why, and how? by jee http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-381 jee Sat, 21 Jan 2012 02:12:33 +0000 http://www.orainternals.com/?p=1268#comment-381 Awesome explanation. Thanks for sharing this valauable information. Great day ahead. Awesome explanation. Thanks for sharing this valauable information. Great day ahead.

]]> Comment on SCN – What, why, and how? by Andrew Rosati http://www.orainternals.com/2012/01/20/scn-what-why-and-how/#comment-377 Andrew Rosati Fri, 20 Jan 2012 15:20:40 +0000 http://www.orainternals.com/?p=1268#comment-377 Its not difficult to figure out how to raise the scn on an pc based oracle instance to just below the soft limit. The infoworld article even mentions that its done with an undocumented parameter. Also, its common practice to issue very restrictive read only production database accounts to non dbas. I have my doubts that their - 24 hours worth of scn jump limit fix - will slow down a persistent opponent. Oracle needs to take this more seriously. Its not difficult to figure out how to raise the scn on an pc based oracle instance to just below the soft limit. The infoworld article even mentions that its done with an undocumented parameter. Also, its common practice to issue very restrictive read only production database accounts to non dbas. I have my doubts that their – 24 hours worth of scn jump limit fix – will slow down a persistent opponent. Oracle needs to take this more seriously.

]]> Comment on Tuning ‘log file sync’ wait events by Log file synch tuning #2 - OraInternals http://www.orainternals.com/2008/07/07/tuning-log-file-sync-wait-events/#comment-371 Log file synch tuning #2 - OraInternals Wed, 04 Jan 2012 19:46:25 +0000 http://orainternals.wordpress.com/?p=40#comment-371 [...] reading my earlier blog about log file sync tuning a reader sent an interesting issue, worth blogging about [...] [...] reading my earlier blog about log file sync tuning a reader sent an interesting issue, worth blogging about [...]

]]> Comment on Tuning latch contention: Cache buffers chain latches by Cardinality feedback to resolve a Cache buffers chains latch contention issue - OraInternals http://www.orainternals.com/2008/07/30/tuning-latch-contention-cache-buffers-chain-latches/#comment-370 Cardinality feedback to resolve a Cache buffers chains latch contention issue - OraInternals Wed, 04 Jan 2012 19:42:10 +0000 http://orainternals.wordpress.com/?p=56#comment-370 [...] I blogged about resolving cache buffers chains latch contention in my earlier entry , in which, root cause was excessive index access due to Nested Loops join. Recently, we resolved [...] [...] I blogged about resolving cache buffers chains latch contention in my earlier entry , in which, root cause was excessive index access due to Nested Loops join. Recently, we resolved [...]

]]> Comment on Scripts by Shared pool freelists (and durations) - OraInternals http://www.orainternals.com/scripts/#comment-369 Shared pool freelists (and durations) - OraInternals Wed, 04 Jan 2012 17:49:19 +0000 http://www.orainternals.com/?page_id=390#comment-369 [...] Scripts [...] [...] Scripts [...]

]]>