Re: The greatest undocumented environmental variable ever

Thomas J. Girsch wrote: > OK, maybe it's not that great, but it's proved damn useful to me. > > IFX_DIRTY_WAIT > > Basically, you set it to a timeout value, and with it set, you can > issue, say, an ALTER TABLE statement without immediately getting an ISAM > -106 (non-exclusive access). Instead, it keeps trying, waiting for the > timeout, before giving the error. So if you need to alter a table or > build an index where you have dirty readers out there, this is quite handy. > > Detailed info here: > > http://www-1.ibm.com/support/docview.wss?uid=swg21174239 > > If it's documented anywhere else, I haven't found it. I only learned of > it from the IBM Premium Support newsletter for November of 2006. > Apparently, the feature was added in 7.31.xD5, 9.30.xC3, 9.40.xC1, and > 10.00.xC1, back before February of 2005, but I just now learned of it. > I thought I might not be the only one in the dark about this, and that > others might find it as useful as I have. Indeed useful feature. There is some additional discussion well written by Fernando Nunes at http://informix-technology.blogspot.com, section "When 'Exclusive' is not really exclusive...". Thanks, Fernando. Darko Krstic

darko wrote: > Thomas J. Girsch wrote: >> OK, maybe it's not that great, but it's proved damn useful to me. >> >> IFX_DIRTY_WAIT >> >> Basically, you set it to a timeout value, and with it set, you can >> issue, say, an ALTER TABLE statement without immediately getting an ISAM >> -106 (non-exclusive access). Instead, it keeps trying, waiting for the >> timeout, before giving the error. So if you need to alter a table or >> build an index where you have dirty readers out there, this is quite handy. >> >> Detailed info here: >> >> http://www-1.ibm.com/support/docview.wss?uid=swg21174239 >> >> If it's documented anywhere else, I haven't found it. I only learned of >> it from the IBM Premium Support newsletter for November of 2006. >> Apparently, the feature was added in 7.31.xD5, 9.30.xC3, 9.40.xC1, and >> 10.00.xC1, back before February of 2005, but I just now learned of it. >> I thought I might not be the only one in the dark about this, and that >> others might find it as useful as I have. > > Indeed useful feature. There is some additional discussion well written > by Fernando Nunes at http://informix-technology.blogspot.com, section > "When 'Exclusive' is not really exclusive...". Thanks, Fernando. > > Darko Krstic > I'm glad you liked it! I was also very surprised when I discovered this variable... It can be very helpful sometimes. I will add some more info about this... I have a script that can help to identify sessions preventing the ALTER TABLE. But as always, time is short... -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently...