I am building a new server with a separate partition for data and
xlogs. What is the correct way to do this?
Can I create a symlink from /var/lib/pgsql/9.1/data -> /pg_data (and
then a symlink in /pgdata for xlogs-> /pg_xlog)
Or do I have to modify $PGDATA in the init script?
This is all after in
> do many of the statistical queries use the whole month's data?
> have you considered partitioning the log table by day? if you can
> reduce the number of rows involved in the table-scans by partitioning
> it'll be help performance.
>
I am summarising by hour, day, week and month. So I guess par
>
>
> Since you are on RHEL 6 I would use ext4 throughout.
>
Great, I'll look into it. However my research suggested a journalled
filesystem is unnecessary for xlogs and I assume ext4 is journalled?
> You say you have I/O problems when "stats jobs" run. Can you describe
> those jobs
> and what t
I have a postgres server I need to move onto a new OS (RHEL6) on a new VM
and am looking for advice on how to partition the disks to gain some
performance improvement.
In the current environment I am given a single VHDD which I have not
partitioned at all. The SAN performance is pretty good, but w
Hmm, no-one seemed to mention the obvious - a pl/pgsql function, either
triggered or run manually depending if you want to update on insert/update
or on demand.
On 7 August 2011 16:05, Sim Zacks wrote:
> **
> On 08/05/2011 07:32 PM, jeffrey wrote:
>
> I have a table that looks like this:
>
> ho
Actually there is a section in the manual on this problem:
http://www.postgresql.org/docs/9.0/static/plpgsql-implementation.html#PLPGSQL-VAR-SUBST
On 29 June 2011 11:41, Ben Carbery wrote:
>
> Hm, "a" isn't a variable,it's the value of stuff[1].
>
>
> It&#
> Hm, "a" isn't a variable,it's the value of stuff[1].
It's both..
DECLARE
a integer;
b integer;
The point is I want a dynamically named variable. Here I've named them the
same as stuff[i] but they can be anything provided it is a different
variable name on each loop iteration. The manu
Hi,
I am trying to EXECUTE .. INTO a variable that I want to be dynamically
named.
stuff := '{a,b,c,d}';
FOR i IN 1..4 LOOP
thing := stuff[i];
-- stuff_a, stuff_b etc are functions - substitution works here
exec_string := 'SELECT stuff_' || thing || '(''' || arg1
Thanks for the responses all, I have this working now. I had to create a
base backup before copying to the standby for replication to start, but the
main sticking point was actually understanding the terms and concepts
involved..
I think the Binary Replication Tutorial page on the wiki basically e
ec 21, 2010 at 2:40 PM, Ben Carbery wrote:
> Hi,
>
> I am having some trouble trying to figure out how to configure this
> particular scenario..
>
> I have a pair of pg servers that I want to put in a Master/Standby
> configuration. Currently a script dumps the master db ever
Hi,
I am having some trouble trying to figure out how to configure this
particular scenario..
I have a pair of pg servers that I want to put in a Master/Standby
configuration. Currently a script dumps the master db every hour, copies it
to the standby, restores, and restarts the server. The aim i
You can always recreate the dependent views in the same process, since you
won't lose any data. If there are also linked tables you could use COPY
before deleting and recreating, it's fairly fast.
But..does the structure of your data really need to change? Just guessing,
but it just sounds a littl
Well, the objects indices 1,2,3 point to changed when you changed column a4,
but I don't know if that's the reason. I would guess that the indices are
structured as pointers of some kind though.
On Wed, Oct 13, 2010 at 9:03 AM, sunpeng wrote:
>
> the question is why all four indices updated in t
For the record, I did find a way to do with another application..
mathematica makes it fairly trivial:
Say a query such as:
data = SQLSelect[someconnection,"sometable"]
Printing it:
TableForm[data]
Then simply:
Transpose[TableForm[data]]
The private keys needs to be readable by the same user the server runs
under. This is distribution-dependent and may not be 'root'.
In my case I run Red Hat which uses the 'postgres' user, so:
chown postgres.postgres /var/lib/pgsql/data/server.*
On Sun, Oct 10, 2010 at 2:52 PM, Mike Christensen
> Or that OP has a networking issue. Some firewalls are known for
> dropping what they think are idle connections when they aren't.
>
>
I don't think so.. EOF is an explicit termination, not a timeout as would
caused by a firewall dropping traffic. It's more like what happens when the
remote proce
Depending on the significance of the primary key, another option may have
been simply a unique constraint (b,c) on the table before the data was added
On Fri, Oct 8, 2010 at 9:55 PM, A B wrote:
> Thank you all for your replies.
>
>
> 2010/10/8 Alban Hertroys :
> > On 8 Oct 2010, at 8:59, A B wro
On Thu, Sep 30, 2010 at 9:36 PM, Vincenzo Romano <
vincenzo.rom...@notorand.it> wrote:
> 2010/9/30 Raymond O'Donnell :
> > On 30/09/2010 12:17, Ben Carbery wrote:
> > Googling on "sql swap rows columns" found this:
> >
> >
> http://stackoverf
On Thu, Sep 30, 2010 at 9:19 PM, Tommy Gildseth
wrote:
> Ben Carbery wrote:
>
>> I have a query that returns many columns but few rows. I would like to
>> display output horizontally instead of vertically, i.e. rotating by 90
>> degress, so column headings appear in
I have a query that returns many columns but few rows. I would like to
display output horizontally instead of vertically, i.e. rotating by 90
degress, so column headings appear in the left margin, and the output is not
'wrapped'.
Is this possible? I have had no luck searching for this as rotate usu
20 matches
Mail list logo
