Lately my database crashed and I’ve had some strangeness following.
I found that some tables would have two distinct rows with identical primary key.
Now, VACUUM complains thusly
WARNING: index "pg_statistic_relid_att_index" contains 2984 row versions, but table contains 2983 row versions
HINT: Rebuild the index with REINDEX.
After trying to reindex, I get
ERROR: could not create unique index
DETAIL: Table contains duplicated values.
I think I’m seeing the same issue – a duplicated row.
I know this should not happen – but does it in fact come to happen sometimes?
Is there a smarter way of dealing with it than deleting the duplicate?
How can these duplicates get produced?
And, for pg_statistic in particular, is it safe to muck with it? What should I do?
Thanks
Jaime