I wrote;
> Hannes Dorbath <light@theendofthetunnel.de> writes:
>> I have the bad feeling that I need to correct this into "It happens when
>> autovacuum is active on the table".
> Ah-hah, I realize how to explain that too, now.
Hmm, no, scratch that: neither VACUUM nor ANALYZE use a standard
heapscan, so they won't move the syncscan pointer. Were you performing
some other query that did a partial seqscan of the table?
regards, tom lane