Re: After replication failover: could not read block X in file Y read only 0 of 8192 bytes

Oops, i missed including pgsql-general in my earlier replies..

> > > I have data_checksums switched on so am suspecting a streaming
> > > > > replication bug. Anyone know of a recent bug which could have
> caused
> > > > > this?
> > > > >
> > > >
> > > > I cannot conclude at this point. I encountered these kind of errors
> with
> > > > Indexes and re-indexing fixed them.
> > >
> > > This is actually the second time I am seeing these kinds of errors, in
> > > the past, after verifying that no data was lost I used VACUUM FULL to
> > > recover the ability to INSERT. There was no pitchfork uprising...
> > >
> >
> > Did you check your disk file systems for any issues ?
>
> I'm on EC2 and praying that data_checksums takes care of that for me...
>

Please check if any issues with the storage.

> > These errors only
> > happen after you do pg_restore ?
>
> they happened after the failover, I'm not convinced they were caused by
> the pg_restore.
>

mmh ok. Any issues on slave before failover ?

>
> I still have the backups and WAL logs and could probably try use them to
> build a bad slave. But that's a few hundred GB of WAL and is not worth
> it unless I know what to look for.
>

Well, that is the work around if you are encountering data loss due to this
bug.
Developers must be able to tell you if this is a bug in streaming
replication. I haven't encountered such bug till now.

Regards,
Venkata B N

Fujitsu Australia