Comments on: Bugs

By: Block size – again « Oracle Scratchpad

Block size – again « Oracle Scratchpad — Fri, 20 May 2011 21:22:40 +0000

[...] problem is similar to the problem of row migration and ASSM – there are some hard-coded limits in what you can put into a single Oracle block, and those [...]

By: Book Review: Oracle Tuning: The Definitive Reference Second Edition « Charles Hooper's Oracle Notes

Sun, 28 Nov 2010 17:18:27 +0000

[...] technical editor of this book) reported fixing the problem, and not the book author (reference reference2 reference3 [...]

By: Jonathan Lewis

Jonathan Lewis — Fri, 12 Feb 2010 08:57:04 +0000

Gary,

Sorry about the late response – this one dropped out of sight before I’d got around to it. I’d be interested to hear what comments you get from Oracle.

When you say MBRC = 512, do you mean the parameter db_file_multiblock_read_count, or do you mean the system statistics named ‘MBRC’ ? And what block size are you using – the 512 suggests it might be 2KB.

By: Gary S

Gary S — Sat, 06 Feb 2010 03:29:05 +0000

I think I have one (a weird bug, that is).

We have a staging table with two partitions: a little shrimpy one that isn’t used and a large one that can be dropped and recreated as new data is staged into the table. If we try to do a full scan against the table, we see really slow performance; Oracle reports the scan, but no scattered reads. All reads against the table are sequential. Extent size is fixed 1 MB (too small I think), ASSM, MBRC is 512. Reads against this table are abysmal.

It does load sort records, then do an update that lengthens the rows. pctfree is 10, but there are no “continue row events” during the scan … so row chaining is playing no visible role.

We’ll send it to Oracle Support on Monday … thought this was odd, though.

By: Database using ASSM Tablespace Exhibits Slow Insert Performance After an Uncommitted Delete « Charles Hooper's Oracle Notes

Thu, 28 Jan 2010 06:02:35 +0000

[...] gets, 18.03 seconds spent performing single block reads. This seems to be behaving similar to the bug that Jonathan Lewis found with ASSM 16KB block size tablespaces in 2008 when column values in existing rows were [...]

By: Bugs – 2 « Oracle Scratchpad

Bugs – 2 « Oracle Scratchpad — Tue, 15 Dec 2009 21:31:17 +0000

[...] Troubleshooting — Jonathan Lewis @ 12:00 pm UTC Oct 10,2009 After asking people about their strangest bugs, it’s a little ironic that I should immediately be subject to one. Here’s a transcript [...]

By: Glossary « Oracle Scratchpad

Glossary « Oracle Scratchpad — Sun, 11 Oct 2009 10:56:38 +0000

[...] three passes against a table, for example to merge data from different source into a single table. There is a bug (fixed the latest releases only) that can turn this error into a performance disaster if you happen [...]

By: Hemant K Chitale

Hemant K Chitale — Thu, 08 Oct 2009 15:03:11 +0000

Here’s one interesting (and very disconcertingly hidden) bug : 6008290 as a duplicate of 4743582.
There’s just not enough information in the published text.
But a forums posting has possibly some evidence :
http://forums.oracle.com/forums/thread.jspa?messageID=3816337

Hemant K Chitale

By: Blogroll Report 18/09/2009 – 25/09/2009 « Coskan’s Approach to Oracle

Blogroll Report 18/09/2009 – 25/09/2009 « Coskan’s Approach to Oracle — Tue, 06 Oct 2009 19:36:05 +0000

[...] Jonathan Lewis – Bugs [...]

By: Analytic Agony « Oracle Scratchpad

Analytic Agony « Oracle Scratchpad — Sat, 03 Oct 2009 15:15:22 +0000

[...] 4: 3rd Oct 2009: In a comment on one of my postings about bugs, Timur Akhmadeev has referenced an OTN thread that references Metalink document id 330818.1 and [...]