Simon Riggs <[EMAIL PROTECTED]> writes:
> In general, the hint bits are good. In *some* cases, not. I still seek
> control over that as a designer.
> Specifically, the scenario I want to optimize is this:
> - we load a table with lots of real time measurement data, as one child
> out of a large nu
Simon Riggs <[EMAIL PROTECTED]> writes:
> On Wed, 2005-07-20 at 09:24 -0400, Tom Lane wrote:
>> We don't rely on any one write of them to work, but that doesn't mean
>> that we can indefinitely postpone writing them.
> OK, I think I understand where you're coming from now.
Apparently not :-(
> W
What really interests me here: Where is you code located in order to
make sure that this things can work reliably?
We are integrated into the PostgreSQL backend.
So they have to get a new PostgreSQL release from you for every minor
upgrade, I assume at no cost?
Yep :)
Sincerely,
Joshua D. Drake
Joshua D. Drake wrote:
Joshua,
This sounds interesting. If you don't use the WAL but a transaction
log shipping - what does it mean in terms of PostgreSQL? Do you
create your own transaction log?
Yes.
What really interests me here: Where is you code located in order to
make sure that this thing
Joshua D. Drake wrote:
> >
> >
> > Joshua,
> >
> > This sounds interesting. If you don't use the WAL but a transaction log
> > shipping - what does it mean in terms of PostgreSQL? Do you create your
> > own transaction log?
>
> Yes.
>
> > What really interests me here: Where is you code loca
Joshua,
This sounds interesting. If you don't use the WAL but a transaction log
shipping - what does it mean in terms of PostgreSQL? Do you create your
own transaction log?
Yes.
What really interests me here: Where is you code located in order to
make sure that this things can work reliably?
We
Neil Conway wrote:
Hans-Jürgen Schönig wrote:
The idea: We are looking for a way to implement a synchronous
single-master / multiple slaves systems.
Meanwhile we are able to serialize / deserialize WAL records and send
them to a group communication system which transports those records to
the sl
Joshua D. Drake wrote:
Alvaro Herrera wrote:
On Sun, Apr 24, 2005 at 08:10:34AM +0200, Hans-Jürgen Schönig wrote:
The idea: We are looking for a way to implement a synchronous
single-master / multiple slaves systems.
Meanwhile we are able to serialize / deserialize WAL records and send
them to a
Joshua D. Drake wrote:
> >>
> >>Very close. We don't use the WAL (yet, slated for probably 8.1) but we
> >>do use a transaction log shipping method. So the implementation is
> >>almost the same.
> >
> >
> > Can you run queries on the slave? If so, how do you handle xid collisions?
>
> You can
Very close. We don't use the WAL (yet, slated for probably 8.1) but we
do use a transaction log shipping method. So the implementation is
almost the same.
Can you run queries on the slave? If so, how do you handle xid collisions?
You can run any query that does not modify data on a replicated ta
Joshua D. Drake wrote:
> Alvaro Herrera wrote:
> > On Sun, Apr 24, 2005 at 08:10:34AM +0200, Hans-J?rgen Sch?nig wrote:
> >
> >
> >>The idea: We are looking for a way to implement a synchronous
> >>single-master / multiple slaves systems.
> >>Meanwhile we are able to serialize / deserialize WAL
Alvaro Herrera wrote:
On Sun, Apr 24, 2005 at 08:10:34AM +0200, Hans-Jürgen Schönig wrote:
The idea: We are looking for a way to implement a synchronous
single-master / multiple slaves systems.
Meanwhile we are able to serialize / deserialize WAL records and send
them to a group communication sy
On Sun, Apr 24, 2005 at 08:10:34AM +0200, Hans-Jürgen Schönig wrote:
> The idea: We are looking for a way to implement a synchronous
> single-master / multiple slaves systems.
> Meanwhile we are able to serialize / deserialize WAL records and send
> them to a group communication system which tra
On Sun, Apr 24, 2005 at 11:41:17AM -0400, Tom Lane wrote:
> Klaus Naumann <[EMAIL PROTECTED]> writes:
> >> what. Allowing that to be turned off would be interesting for a number
> >> of purposes, such as burning a database onto CD.
>
> > FWIW, Oracle suggests a "transportable tablespace" for this
Klaus Naumann <[EMAIL PROTECTED]> writes:
>> what. Allowing that to be turned off would be interesting for a number
>> of purposes, such as burning a database onto CD.
> FWIW, Oracle suggests a "transportable tablespace" for this feature.
> Which is a tablespace that is not written too and which
what. Allowing that to be turned off would be interesting for a number
of purposes, such as burning a database onto CD.
FWIW, Oracle suggests a "transportable tablespace" for this feature.
Which is a tablespace that is not written too and which can be read by
any database.
Would that solve the pur
Hans-Jürgen Schönig wrote:
The idea: We are looking for a way to implement a synchronous
single-master / multiple slaves systems.
Meanwhile we are able to serialize / deserialize WAL records and send
them to a group communication system which transports those records to
the slave database.
BTW,
=?ISO-8859-1?Q?Hans-J=FCrgen_Sch=F6nig?= <[EMAIL PROTECTED]> writes:
> We have toyed around with PostgreSQL's WAL structure and we are
> wondering whether it is (theoretically) possible to replay WAL records
> coming from a remote host while a system is working in read only mode.
There's been so
We have toyed around with PostgreSQL's WAL structure and we are
wondering whether it is (theoretically) possible to replay WAL records
coming from a remote host while a system is working in read only mode.
The idea: We are looking for a way to implement a synchronous
single-master / multiple sl
19 matches
Mail list logo
