Jan Wieck wrote:
> On 7/11/2006 12:43 PM, Christopher Browne wrote:
> 
>> I have set up tarballs for a first release candidate; Jan should be
>> uploading those soon, and I'll see about updating the web site when they
>> are uploaded...
> 
> I have run them through a compile and a few tests and uploaded them to 
> the usual location at
> 
> http://developer.postgresql.org/~wieck/slony1/download/
> 
> One thing I wanted to discuss is related to a recent discussion on the 
> hackers mailing list with respect to possible btree corruption due to 
> xxid not being suitable. With a single origin setup, there is not 
> problem during normal operation. But even then, any failover of move set 
> could immediately cause the btree on the xid only (not the one on 
> node-id, xid) to corrupt. I remember that this index was added because 
> of the backend failure to use the one with the node-id. But that IMO 
> doesn't warrant the risk of corrupting the index at the very time there 
> is already trouble in the house (move set or failover isn't done because 
> it is such a boring day).

there is another issue with the current code - while it compiles and
works fine against postgresql -HEAD with standards_conforming_strings =
off(the default) it will silently corrupt data with
standards_conforming_strings = on.
The reason for this is the rather ad-hoc quoting in slon_quote_literal()
- it will for example double-escape '\\' in strings leading to
regression failures with for example testddl.
Maybe we should take a look at the new escaping functions in the recent
postgresql releases to make that more safe and future proof ?


Stefan