Re: IPLOAD KABOUM

Jake Salomon wrote: > > gosport wrote: > > > > Hi, here is a nice bug I have discovered this afternoon : > > We use Online DSA 7.20.UC2 on an 3CPU HP9000 K200... > > > > TEST CASE: > > create table junk( field integer ) > > fragment by expression > > (mod(field,2)=0) in dbs1, > > (mod(field,2)=0) in dbs2;> > Valery, > > what you have above is known as "overlapping fragments", except yours is > more blatant than most cases of this. > > > Obviously, the creation script is incomplete/stupid/insane, BUT...Online does > > not refuse to create this table with such a fragmentation scheme ;-) > > The SQL parser has enough to do just parsing your commands. Noticing > poor logic is not the parser's line of work. I have thought it might be > nice to put some AI into the parser, but then my burned-in skepcism > about AI steps in and tells me to sit down and stop rocking the boat. > ;-) > Gee, thank you for your Re: BUT...I have something new about IPLOAD and fuzzy fragmentation: Well, yesterday and today, we have reorganized our whole instance (we bought 4 brand new 2Go disks and 1 HSC F/W controller card. Because we have big (millions of records) tables, we planned to use ipload. Good choice ! We've unloaded our databases in 4 hours ! Brilliant ! During the reload, ipload crashed online once (it also crashed our HP9000!). Why: Before the reorg, we had a table which was not fragmented. We have decided to fragment it over 4 fragments, with a mod expression on one column (a correct one this time ;-) BUT we did not know that there were NULL values for 4-5 rows (they shouldn't have existed, but they did actually). Of course, those columns didn't match our so-called correct fragmentation scheme. And what happened ? I let you guess.... We figured two solutions : using the REMAINDER keyword in place of the last line of the frag scheme; or...deleting the nauty rows... Doesn't seem it's the infamous "overlapping fragments"; don't you think so ? And again, when we tried to load the table with dbload, the nasty rows were only rejected... :-D Is it a known bug ? Is it fixed in the 7.22 ? Don't answer, I finally called the Informix Tech Support in France... > > > > _ __/| > > \\`O.o' "For every problem there is a solution which > > =(_ _)= is simple, clean and wrong" > > U > > What? You get the Dilbert comic strip in France? ;-) Na, I found it somewhere else. What d'you think of the one forward ??? -- Valery Fouques valou@hotmail.com SYSADM GOSPORT-FRANCE /\\oO/\\ _ __/| \\`O.o' "355/113 -- Not the famous irrational =(_ _)= number PI, but an incredible simulation!" U