Re: [GENERAL] pg_dump 8.3.3 ERROR: invalid page header in block2264419 of relation "pg_largeobject"

On Wed, May 24, 2017 at 04:45:51PM -0700, David Wall wrote:

> They do have a slave DB running via WAL shipping.  Would that likely help us
> in any way?

If you can find out which blobs are afflicted you may be able
to extract those from the slave and re-insert them into the
new DB.

> Because the DBs are big (they have two at 191GB and 127GB), it
> takes a fair bit of time to do backups, transfers and restores.  I'm trying
> to find options as it likely means downtime for them that they are not
> expecting (yet).

Oh, they likely have downtime already, because what you first
reported smells of bad hardware ?

Karsten
--
GPG key ID E4071346 @ eu.pool.sks-keyservers.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346