Re: BUG: *FF WALs under 9.2 (WAS: .ready files appearing on slaves)

Am 31.12.2014 um 11:40 schrieb Michael Paquier <michael(dot)paquier(at)gmail(dot)com>:
>> As long as master is fixed, I don't actually care. But I agree with Dennis
>> that it's hard to see what's been commited with all the different issues
>> found, and if any commits were done, in which branch. I'd like to be able to
>> tell my customers: update to this minor release to see if it's fixed, but I
>> can't even do that.
> This bug does not endanger at all data consistency as only old WAL
> files remain on the standby, so I'm fine as well with just a clean fix
> on master, and nothing done on back-branches.

I’d like to point out that this issue is causing severe disk space bloat over time, as described in my previous posting. I’d assume that, as time goes by, this issue will pop up more and more, as long as back-branches remain unfixed.

And as the general recommendation is to never ever fiddle with pg_xlog/ contents, I’m not even sure what to do about it.

- D.