2011/11/7 Ivan Sergio Borgonovo <mail@webthatworks.it>:
> On Mon, 7 Nov 2011 17:55:11 +0100
> Pavel Stehule <pavel.stehule@gmail.com> wrote:
>
>> Hello
>>
>> this is know bug/feature based on caching plans
>
> What puzzled me is I'm operating in a similar way in a different
> system and I'm not experiencing the same problem.
>
> Do different users have different caches?
depend on usage - cache is per session
> What about different sessions?
if you don't change a search_path inside session, then all will works well
>
>> There is workaround - you can put a copy of test_insert function to
>> every schema - no to "public" schema.
>
> That's pretty ugly. I'll take the chance to refactor everything and
> learn.
yes, this is workaround - it's not nice
>
> Where can i learn about creation and invalidation of plans in
> postgres documentation?
>
> BTW it looks to me you just answered to my address and not to the
> list. If it was by mistake feel free to repost everything to the
> list for other people's reference.
>
> thanks
>
> --
> Ivan Sergio Borgonovo
> http://www.webthatworks.it
>
>