Re: Chunks Down&/Out/Not?

In article <1md098INN8tb@emory.mathcs.emory.edu> rridley@sin-pss.DHL.COM (Richard Ridley) writes: >Running Informix On-line 4.00.UH1 on an HP 827 (HP-UX 8.02), when we >run tbmonitor it tells us that ALL chunks (including those of the >ROOTDBS) are Down. Also, running tbcheck -cD tells us that all >chunks are off-line. > >Now, the Online instance is up and running and we can still access >the data in our tables (Yes - the chunks are marked Y for mirroring >BUT have no mirror devices assigned to them) This is an old bug/feature in Informix Online which I believe is still supposed to be around through 5.0. Whenever Informix receives some kind of "error message" or bad signal from a hard drive, it sets an internal flag which says that all chunks on that drive are forevermore bad and "down". There is no way to reset this flag, whether or not the drive is actually bad or not. However, if the mirroring indicator is set to Y in tbconfig then the "bad chunk" flag is not read and Informix lets you use the chunk whether or not it is set. Therefore, you can "fix" this problem -- once -- by changing your mirroring flag to "Y" even if you have no mirroring. I'm not sure what happens if Informix thinks that it has received a bad signal again after that. We had this happen to us as well. Luckily someone from Ix tech support knew of this "bug" or our production system would have been down until we could completely recreate the database from tape (about 2 days work). As far as we could tell, once we used the mirroring trick, there was absolutely nothing wrong with our data. - Rich Puchalsky