Re: Corrupt Logical Log

In article <D6oEpz.BtM@cix.compulink.co.uk> akent@cix.compulink.co.uk ("Andy Kent") writes: > > A couple of days ago I experienced the unfortunate problem of the a > > logical log that contained the current checkpoint had become corrupt. > > The engine had aborted while holding a latch and shut itself down. > When > > restarting the initialization process failed as it couldn't find the > > last checkpoint record to roll forward from and thus shutdown again. > We > > had to restore from backup *but* is there away to amend the current > > logical log number so that it would have looked at a previous one. I > > didn't get much response from our support VAR. Any body got any bright > > ideas... > > > > David > > I seem to recall seeing a utility, called something like tbstop or tbshut > or something, which modifies a specified log record so as to stop OnLine > looking beyond it when it does fast recovery. The copy I saw was supplied > as C source, presumably from Informix Tech Support. The readme with it > implied that it wasn't all that top secret. > > akent@cix.compulink.co.uk (Andy Kent) > ------------------------------------------------ > Freelance Informix Database Specialist, > Redland, Bristol, England Sometime ago I had this problem, and I ended with some corrupt data. At the time the utility you talk about would have been welcome. So, in case it happens again (hope not), I would like to know more about it, if you can of course. Thank you. -- .signature under construction