Re: Bugs in TOAST handling, OID assignment and redo recovery

Am Dienstag, den 10.04.2018, 15:59 +0530 schrieb Pavan Deolasee:
> One of our 2ndQuadrant support customers recently reported a sudden
> rush of
> TOAST errors post a crash recovery, nearly causing an outage. Most
> errors
> read like this:
>
> ERROR: unexpected chunk number 0 (expected 1) for toast value nnnn
>
> While we could bring back the cluster to normal quickly using some
> workarounds, I investigated this in more detail and identified two
> long
> standing bugs in TOAST as well as redo recovery.

Wow! I'm currently also investigating issues on a customer system, too,
where suddenly TOAST errors arised after shutdown immediate.

I haven't dug into your findings yet, but it seems to perfectly fit
into the reports i got.

Thanks,
Bernd