RE: LONG TRANSACTION!!!!!!!!!!1
Posted in 2000
What messages are in the log?
-----Original Message-----
From: Simbarashe.Pangeti@icl.com [mailto:Simbarashe.Pangeti@icl.com]
Sent: Thursday, November 16, 2000 7:06 AM
To: kagel@bloomberg.net; informix-list@iiug.org
Subject: LONG TRANSACTION!!!!!!!!!!1
people one of my operators shut down the serve while i t was processing
along transaction and its refusing to come up i am using solaris and 7.31
online what can i odo i need you help pronto
-----Original Message-----
From: Art S. Kagel [mailto:kagel@bloomberg.net]
Sent: Wednesday, November 15, 2000 12:06 AM
To: informix-list@iiug.org
Subject: Re: table extents in what chunks?
c-eriks@algonet.se wrote:
>
> Hi!
>
> In my efforts to "eliminate extent interleaving" I came about to study
> the distribution of table extents among dbspace chunks. For a specific
> table I want to know in what chunks it's extents are located. (I can do
> a "oncheck -pe | more" and look through the output but I want to query
> the SMI-tables instead). Any ideas about such a sql-query?
SELECT *
FROM sysextents
WHERE dbsname = "mydatabase" and tabname = "mytable"
ORDER BY start;
> What I then want to do is a "ALTER INDEX TO CLUSTER" on a table
> (described in INFORMIX-Online Dynamic Server, Administrator's Guide,
> Volume 1 Version 7.1", page 15-23) to rebuild the table with _one_
> extent only. My focus on chunks is because of the following lines in
> the Guide: "The chunk must contain adequate contiguos space in which to
> rebuild the table". But what if the table has extents in many chunks?
> Does anybody have experience in doing such a rebuild of a table?
Use ALTER FRAGMENT ON TABLE mytable INIT IN somedbspace; where somedbspace
can even be the same dbspace in which the table already resides. This is
the fastest reorg and if you repeatedly reorg your tables over time there
will not be many small free extents so the tables will have very few
extents after a very short time.
Art S. Kagel