Wed Nov 12 13:27:19 PST 2014
- Previous message: [Slony1-bugs] [Bug 355] New: execute script () modifies search_path
- Next message: [Slony1-bugs] [Bug 355] execute script () modifies search_path
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
http://www.slony.info/bugzilla/show_bug.cgi?id=355 --- Comment #1 from Christopher Browne <cbbrowne at ca.afilias.info> 2014-11-12 13:27:19 PST --- The stored function should be protecting itself; if it depends on having a particular search path, it should do so as part of the stored function definition. So it seems to me that there's a bug in the stored function. (And I suspect I myself might very well get assigned to fix that bug...) Slony itself operates relatively safely as it specifies fully qualified object names in most (hard to prove "all") cases. It seems like it would be a good idea for the code that runs at the end of DDL processing (see src/backend/slony1_funcs.c) to add in "RESET ALL;" to reset the environment at the end of DDL processing in case search paths or other GUCs got messed with by DDL. There's a further bit of risk; if a DDL script is messing around with GUCs, that risks messing with another DDL script that might be run in the same SYNC. It seems a good idea to add a warning to users that messing around with GUCs in DDL could injure other users... -- Configure bugmail: http://www.slony.info/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. You are the assignee for the bug.
- Previous message: [Slony1-bugs] [Bug 355] New: execute script () modifies search_path
- Next message: [Slony1-bugs] [Bug 355] execute script () modifies search_path
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-bugs mailing list