On Wed, Nov 9, 2011 at 3:40 PM, Dimitri Fontaine wrote:
> Back to the problem you raised, it's DDL and we're sitting in between
> SnapshotNow and catalog cache entries. Not so comfy. I would guess
> that the problem (I confess didn't read carefully enough) happens after
> having done a cache loo
Robert Haas writes:
> e.g. Suppose that you have a table with a unique index on column a.
> Transaction A deletes the tuple where a = 1. Transaction B attempts to
That's DML, I agree with you there, no need. In DML we have MVCC.
Back to the problem you raised, it's DDL and we're sitting in betw
On Wed, Nov 9, 2011 at 10:50 AM, Cédric Villemain
wrote:
> 2011/11/9 Robert Haas :
>> On Wed, Nov 9, 2011 at 8:37 AM, Cédric Villemain
>> wrote:
>>> Maybe I miss something but:
>
> I read that the error was produced by first session and didn't check
> carefuly (it fails silently in 9.0! and 'work
On Wed, Nov 9, 2011 at 10:33 AM, Tom Lane wrote:
> Robert Haas writes:
>> On Wed, Nov 9, 2011 at 9:54 AM, Tom Lane wrote:
>>> I would think you need to drop the now-useless lock, and I sure hope
>>> that RangeVarGetRelid does likewise.
>
>> It doesn't currently. The now-useless lock doesn't rea
2011/11/9 Robert Haas :
> On Wed, Nov 9, 2011 at 8:37 AM, Cédric Villemain
> wrote:
>> Maybe I miss something but:
I read that the error was produced by first session and didn't check
carefuly (it fails silently in 9.0! and 'works' as expected in 9.1)
No objection, but I would like to still be a
Robert Haas writes:
> On Wed, Nov 9, 2011 at 9:54 AM, Tom Lane wrote:
>> I would think you need to drop the now-useless lock, and I sure hope
>> that RangeVarGetRelid does likewise.
> It doesn't currently. The now-useless lock doesn't really hurt
> anything, aside from taking up space in the lo
On Wed, Nov 9, 2011 at 9:54 AM, Tom Lane wrote:
> Robert Haas writes:
>> I'd like to propose the attached patch, which changes
>> get_object_address() in a manner similar to what we did in
>> RangeVarGetRelid() in commit 4240e429d0c2d889d0cda23c618f94e12c13ade7.
>
> I would think you need to drop
Robert Haas writes:
> I'd like to propose the attached patch, which changes
> get_object_address() in a manner similar to what we did in
> RangeVarGetRelid() in commit 4240e429d0c2d889d0cda23c618f94e12c13ade7.
I would think you need to drop the now-useless lock, and I sure hope
that RangeVarGetRe
On Wed, Nov 9, 2011 at 8:37 AM, Cédric Villemain
wrote:
> Maybe I miss something but:
> The ERROR message is misleading: the schema 'x' does exist.
No, it doesn't. The concurrent transaction has dropped it.
> And also
> why a drop schema would fail and a drop+create would success ?!
Because y
2011/11/9 Robert Haas :
> I'd like to propose the attached patch, which changes
> get_object_address() in a manner similar to what we did in
> RangeVarGetRelid() in commit 4240e429d0c2d889d0cda23c618f94e12c13ade7.
> The basic idea is that, if we look up an object name, acquire the
> corresponding
I'd like to propose the attached patch, which changes
get_object_address() in a manner similar to what we did in
RangeVarGetRelid() in commit 4240e429d0c2d889d0cda23c618f94e12c13ade7.
The basic idea is that, if we look up an object name, acquire the
corresponding lock, and then find that the objec
11 matches
Mail list logo
