R
Rob Fitzpatrick
Guest
> Wow, based on your screenshots it looks like the problem developed over the course of 3.5 hours. The anomalous activity started around noon and the DB stalled at 13:48. I haven't pinpointed yet exactly when the growth started, which wasn't necessarily when the first long-running transaction started. > I sure wish OE had some sort of reasonable transaction timeout which could be enforced by the DBMS itself. The problem with trying to design a generic platform solution is that everyone has their own idea of what is "reasonable". In this case, the client was not monitoring BI growth or transaction age. I am having that conversation with them now. Both are possible programmatically, even if OEE isn't working for you as you expect. You can look at _trans._trans-duration and _dbstatus._dbstatus-bisize.
Continue reading...
Continue reading...