On Wed, Aug 9, 2017 at 6:27 AM, ADSJ (Adam Sjøgren) <adsj@xxxxxxxxxxxxx> wrote: > On 2017-06-21 Adam Sjøgren <adsj@xxxxxxxxxxxxx> wrote: > >> Adam Sjøgren <adsj@xxxxxxxxxxxxx> wrote: > >>> Meanwhile, I can report that I have upgraded from 9.3.14 to 9.3.17 and >>> the errors keep appearing the log. > > Just to close this, for the record: We haven't seen the errors since > 2017-06-30. We upgraded to 9.3.17 (latest 9.3 point-release at the time > of writing) on 2017-06-10. > > Whether this means that the affected rows gradually got overwritten > after switching to .17 and thus got fixed, or if something subtle in our > workflow changed, so we aren't hitting this anymore, or something else > entirely is the answer, we're not sure. > > We didn't get to trying Alvaro Herrera's suggestion of removing > 6c243f90ab6904f27fa990f1f3261e1d09a11853 before the errors stopped > appearing "by themselves". This sounds a lot like bad sectors getting remapped. -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
