On 12 February 2016 at 17:56, Oliver Stöneberg <oliverst@online.de> wrote:
A few weeks ago we already had a data corruption when the disk was full. There are other services running on the same machine that could cause the disk to fill up (e.g. local chaching when the network is acting up). It happened a few times so far but the database was never compromised. In that case thought it was but fortunately we only lost a huge table/toast (300+ GB) that has very verbose data stored which is not essential. That happened with an earlier 9.4 version.
What remedial action was taken to restore the database to normal functionality at this time?
Is the current database a direct descendant of the one that was corrupted here? i.e. has it had a complete dump, initdb and restore since then, or not?
Unfortunately we don't have a recent backup of the database (a tool to back up all the relevant data was just finished recently and was not set up for this system yet).