I think we can all agree (well most of us) that ratios are not a very useful measurement in most situations. Because most tuning problems exist at the SQL level rather than the database.I wouldn't call that a ringing endorsement of the BCHR as even an in-extremis
init.ora parameter event = "10269 trace name context forever"Steve Adams makes that suggestion with several restrictions and under a section entitled 'Oracle 7.3'. Can you confirm whether it is appropriate or even supported in 8i (for which you appear to be suggesting it) and under what circumstances it is effective?
I would prefer to say "only people who don't know any better". This may be because they are idiots or it may be because nobody has told them a different way of doing things. Some people can learn and other people are idiots.only complete idiots use ratios as a diagnostic tool.
Actually Laurent Schneider told him. Which I originally posted with a smiley but there is a serious point here. The guy had looked at the init.ora parameters because he posted them to the forum and he still couldn't see what the problem was. But what can we do? People are allowed to have kids without getting a breeding licence and people are allowed to administer databases without understanding how the darn things work.the BCHR didn't tell him what to do to fix his problem: a quick squiz at the
init.ora did that.
Whatever.So, my point: no, that's a lousy example of when the BCHR was put to good use.
Ha, good comment!People are allowed to have kids without getting a breeding licence and people are allowed to administer databases without understanding how the darn things work.
From your Blog (my emphasis):Fact is, it wasn't dropped functionally, and I am happy to acknowledge as much.