On Fri, Aug 12, 2016 at 3:09 PM, Ioana Danes <ioanadanes@xxxxxxxxx> wrote: .... > drawid | 318220 ... > drawid | 318216 > Here are the facts I know: > > August 10 @ 11:10 > - The record was created on db1 and replicated to db2 and db3 > August 11 @ 2:30 > - db1, db2 and db3 are in sync (I have a script that compares the data > for all 3 dbs every night @ 2:30 am) > August 12 @ 2:30 > - db3 is out of sync because of this field (drawid) > - drdb (which is PITRed from db3) is in sync with db1 and db2????? given 318220 and 318216 are just a bit away ( 4db08/4db0c ), and it repeats sporadically, have you ruled out ( by having page checksums or other mechanism ) a potential disk read/write error ? > Also the index is correct on db3 as the record in case (with drawid = > 318216) is retrieved if I filter by drawid = 318220 Specially if this happens, you may have some slightly bad disks/ram/ leading to this kind of problems. Francisco Olarte. -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
