Re: master db diagram

Yeah, what he said. Yes mostly trial and error. I am in the process of building a monitoring tool using SMI but it is still just an outline of an application. BTW anyone have any suggested reports? Current plan includes: o Basic statistics analysis including BR, RA Utilization, Buffer cache turnover rate (anyone know how to determine when stats were last zero'd?), etc. o Dbsavail report of dbspace freespace. o Table freespace report ala PrintFreeB.ec Any suggestions appreciated. Tim Schaefer wrote: > > Paul, Sunil, Art, etc. :-) > > You can get a few tips on sysmaster right out of the sysmaster.sql > script in the $INFORMIXDIR/etc directory. There are also some > great write-ups from Lester Knutsen regarding sysmaster, as Art > pointed out. > > See also: > http://www.advancedatatools.com/articles.htm > lester@advancedatatools.com > > You could also use a tool I wrote a while back to do your own > analysis, available at my website. It produces static HTML and > a cross-ref based on column-name similarities, which are of course > are not guaranteed join-columns, but probably are. It may allow > you to do some of your own analysis of the sysmaster data base, > as well as any other data base you add to the list. I have a > dynamic version of this tool, but it is unreleased, so you'll > have to do with this version for now. The dynamic version is > coming out sometime this quarter, but no specific release date. > > You could also use an ER program to add to your knowledge. It's > also important to note that the 7.x engine is not quite the same > as the 8.x engine, and probably the same for the latest release of > the 9.x engine, all three engines are unique when it comes to the > sysmaster tables in the engine. 8 and 7 are merging from what I > hear, but this will be a major upgrade, and eventually I guess > there will only be one engine as 9.x and the next 8.x-7.x engines > merge. > > Hope this helps, > > Tim > > "Howard, Paul" wrote: > > > > Art, > > > > Are you saying in a round about way that learning about the sysmaster > > database is still a trial and error process? > > > > Yes, for the most part. :-) > > -- > . > .- > .-- > .--- > .---- Tim Schaefer > .----- tschaefe@bellsouth.net > .---- http://www.inxutil.com > .--- > .-- > .- > .

"Art S. Kagel" wrote: > > Yeah, what he said. Yes mostly trial and error. I am in the process of > building a monitoring tool using SMI but it is still just an outline of an > application. BTW anyone have any suggested reports? Current plan includes: > > o Basic statistics analysis including BR, RA Utilization, Buffer cache > turnover rate (anyone know how to determine when stats were last zero'd?), > etc. Got it . . . check out sysmaster:sysshmvals.sh_curtime?? -- John Carlson Informix DBA WHSmith USA #include std_disclaimer.h /* These are my opinions, not my company's opinion */

"Carlson@WHSmith" wrote: > > "Art S. Kagel" wrote: > > > > Yeah, what he said. Yes mostly trial and error. I am in the process of > > building a monitoring tool using SMI but it is still just an outline of an > > application. BTW anyone have any suggested reports? Current plan includes: > > > > o Basic statistics analysis including BR, RA Utilization, Buffer cache > > turnover rate (anyone know how to determine when stats were last zero'd?), > > etc. > > Got it . . . check out sysmaster:sysshmvals.sh_curtime?? > > -- > John Carlson > Informix DBA > WHSmith USA > > #include std_disclaimer.h /* These are my opinions, not my company's > opinion */ I really missed this one . . . . should have been sysmaster:sysshmvals.sh_pfclrtime . . . as the sysmaster.sql line says: sh_pfclrtime int, { time profilers were last clr } Gotta watch what I'm typing . . . . make sure the line that I'm looking at . . . . etc. <G> -- John Carlson Informix DBA WHSmith USA #include std_disclaimer.h /* These are my opinions, not my company's opinion */