I had a batch program chew up 16 gb of PGA and run for extended time doing updates to ( my only ) table that has a text index on it ( customer last name ) ... used for "soundex" lookups.
It choked eventually and terminated with ORA 4030 ... have not (more...)
My OLTP system just cutover from 220.127.116.11.x to 18.104.22.168.5 and a bunch of my custom monitoring stuff now looks a whole lot different.
I had some custom slicing and dicing of the contents of sgastat ( I query it once an hour (more...)
Sorry for anyone that may have seen this already via Oracle L ... some volleying back and forth on this item ( and thanks to Jonathan Lewis ).
This is the same approximate workload ( one week in between ) on an 11.1 system and now running on 11.2. (more...)
We are working on a set of upgrades to our environment and replacing a set of very stable but now old IBM x3650 servers ( currently running 5.7 ) with a set of new Dell R 710 servers.
New Dell servers on Oracle Linux 6.2 ... using red hat compatible kernel aka: 2.6.32-220.el6.x86_64
The new dell boxes have an internal raid controller ( Perc H700 ? ) and are connected to EMC direct attached storage using emulex HBA's. All operating system and linux software installed on internal disks ( mirrored ) ... all database stuff going (more...)
My usergroup ( NEOOUG = Northeast Ohio Oracle Users Group ) has been running a big two day conference at Cleveland State in May for a number of years. Last year our keynotes were from Cary Millsap and Rich Niemic.
This year we are hitting it out of the ball park with Chris Date ( CJ Date ) the godfather of relational database design theory and principles along with Tom Kyte.
Conference renamed this year as GLOC Great Lakes Oracle Conference on May 14th/15th with additional 1/2 day workshop ( extra cost ) on monday May 13th by Tom (more...)
Geez some days it is hard for a grumpy old dba to be so grumpy ... no worries I will find something to make me complain eventually.
This geek is pretty pumped up to have my presentation accepted for hotsos 2013
I am presenting with the exciting title of "OS Truth, Little White Lies, and the Oracle Wait Interface" ... kind of a play on "Truth, Lies, and Videotape" ( which I think I saw once a long long time back ).
This is my presentation abstract here ... John Hurley
Cary Millsap I believe is the ultimate decision maker on who makes (more...)
In my prior post I showed a bad inline view that was causing a huge performance hit. Running in an OLTP environment no less but well thats kind of outside of the solution.
It was bugging me all weekend and with the help of Mladen I came up with a solution.
The original SQL ( again part of an inline view in a much bigger piece of SQL ) looked like this:
FROM SCHEMA_NAME.PO_COMMENT POC
Part of a very long query put together by a developer and of course implemented in production. Lots of inline views ( not a bad idea many times ) ... big tables it was going against ... many of them. Still we have a pretty fast system ...
Things were still ok until you get to this part of the SQL:
(SELECT POC.PO_ID, POC.LINE_ITEM_PO, POC.COMMENT_PO, POC.DATE_ADD_COMMENT FROM PO_COMMENT POC WHERE (POC.PO_ID,POC.LINE_ITEM_PO,POC.ROW_TIMESTAMP) IN (SELECT A.PO_ID,A.LINE_ITEM_PO,MAX(A.ROW_TIMESTAMP) FROM PO_COMMENT A WHERE A.TYPE_PO_COMMENT='LE' GROUP BY A.PO_ID, A.LINE_ITEM_PO) ) T193
So were are operating against (more...)