On Wed, 9 Oct 2019 at 20:06, Pól Ua Laoínecháin <linehanp@xxxxxx> wrote:
One of my courses is "Advanced Databases" - yummy I thought - it's not
even compulsory for me but I just *_had_* to take this module. The
lecturer is a bit of an Oracle fan-boy (ACE director no less...
hmmm...) and I want(ed) - becoming less enthusiasic by the minute - to
do my dissertation with him.
It's a good thing that you have the opportunity to do that course and to have an lecturer with strong real-world experience.
So, we're having a chat and I make plain
my love of good 'ol PostgreSQL as my RDBMS of choice and he tells me
that there are problems with random block corruption with PostgreSQL.
I said "really" and before that conversation could go any further,
another student came over and asked a question.
I think its true that there have been reported problems with block corruption with both Oracle and PostgreSQL. The main difference is that the PostgreSQL project is open enough for people to see much of that on public record.
Given the efforts made on resilience and recovery, such as PITR, block checksums, those problems are pretty much solved, based upon a statistically sufficient sample of the real world: 2ndQuadrant customers. Some block-level problems do still recur - the recent fsync() problems were reported by us and have been handled (not resolved), but they were and are rare. There are still many issues of other kinds.
Many Oracle customers I have worked with years ago experienced block corruptions and it was very common to hear the reply "talk to your disk vendor". Those happened probably 20-30 years earlier, so in many cases have now been forgotten. There isn't an easy way to go back and check for trends on that.
Having said all of that, its easy to point at some of these things and use them as FUD - fear, uncertainty and doubt. No doubt unintentionally.
I'd go and learn more from your lecturer. Your disagreement has made you think, so he helped you. Learning from your own mistakes takes longer.
