Re: Allow ERROR from heap_prepare_freeze_tuple to be downgraded to WARNING

Hi,

On 2020-09-14 13:26:27 -0400, Robert Haas wrote:
> On Sat, Aug 29, 2020 at 4:36 AM Dilip Kumar  wrote:
> > One example is,  suppose during vacuum, there are 2 tuples in the hot
> > chain,  and the xmin of the first tuple is corrupted (i.e. smaller
> > than relfrozenxid).  And the xmax of this tuple (which is same as the
> > xmin of the second tuple) is smaller than the cutoff_xid while trying
> > to freeze the tuple.  As a result, it will freeze the second tuple but
> > the first tuple will be left untouched.
> >
> > Now,  if we come for the heap_hot_search_buffer, then the xmax of the
> > first tuple will not match the xmin of the second tuple as we have
> > frozen the second tuple.  But, I feel this is easily fixable right? I
> > mean instead of not doing anything to the corrupted tuple we can
> > partially freeze it?  I mean we can just leave the corrupted xid alone
> > but mark the other xid as frozen if that is smaller then cutoff_xid.
> 
> That seems reasonable to me. Andres, what do you think?

It seems pretty dangerous to me. What exactly are you going to put into
xmin/xmax here? And how would anything you put into the first tuple not
break index lookups? There's no such thing as a frozen xmax (so far), so
what are you going to put in there? A random different xid?
FrozenTransactionId? HEAP_XMAX_INVALID?

This whole approach just seems likely to exascerbate corruption while
also making it impossible to debug. That's ok enough if it's an explicit
user action, but doing it based on a config variable setting seems
absurdly dangerous to me.

Greetings,

Andres Freund