Ron Johnson wrote:
> Does PG set up buffers at the postmaster level or the database level?
>
> If at the database level, then you'll be allocating memory to
> databases that might not be in use at any one time, thus wasting it.
> One database buffer pool would make more efficient use of RAM.
On Wed, 30 May 2007 05:24:57 +0200, Erwin Brandstetter
<[EMAIL PROTECTED]> wrote:
On May 30, 2:11 am, Rodrigo De León <[EMAIL PROTECTED]> wrote:
(... useful code example snipped)
Now see:
http://www.postgresql.org/docs/8.2/static/plpgsql-control-structures
Thanks for your hints, Rodri
On May 30, 6:48 am, Rodrigo De León <[EMAIL PROTECTED]> wrote:
> On May 29, 11:35 pm, Erwin Brandstetter <[EMAIL PROTECTED]> wrote:
>
> > EXECUTE
> > 'SELECT '
> > || (SELECT array_to_string(ARRAY(
> > SELECT a.attname
> > FROM pg_class c, pg_namespace nc, pg_attribute a
> > WHERE c.relname = 'v_
Hmm, well if both of you say that ORDER BY is not somehow ignored by
refcursor functions then I'm sure you're right. I'm just very clusmy in my
testing of the output...somehow the output order by is being lost. I've been
testing via the results in a web app and have had issues with executing the
p
On May 29, 11:35 pm, Erwin Brandstetter <[EMAIL PROTECTED]> wrote:
> EXECUTE
> 'SELECT '
> || (SELECT array_to_string(ARRAY(
> SELECT a.attname
> FROM pg_class c, pg_namespace nc, pg_attribute a
> WHERE c.relname = 'v_event'
>AND c.relnamespace = nc.oid
>AND nc.nspname = 'stdat'
>AND
To conclude (to the best of my current knowledge), here is a plpgsql
code sample based on what was said here:
EXECUTE
'SELECT '
|| (SELECT array_to_string( ARRAY(
SELECT column_name::text
FROM information_schema.columns
WHERE table_schema = 'my_schema'
AND table_name = 'my_relation'
On May 30, 2:11 am, Rodrigo De León <[EMAIL PROTECTED]> wrote:
(... useful code example snipped)
> Now see:
>
> http://www.postgresql.org/docs/8.2/static/plpgsql-control-structures
Thanks for your hints, Rodrigo!
I am aware I can consult pg_catalog / information_schema to retrieve
the informa
No, you can't. The reason there is no "nonvolatile" list of timezones
is that timezones are subject to the whims of politicians, who can and
do change them at the drop of a hat. Read the historical comments in
the zic source files sometime...
It used to be that "fully one third of timezones in
"Stuart Cooper" <[EMAIL PROTECTED]> writes:
> My system is stabilised and I'm looking at this now. I'm not getting a
> lot from pg_prepared_statements yet: perhaps this view only reports on
> statements you've prepared using PostgreSQL's PREPARE through their
> SQL interface, and not DBI's DBD::Pg
Mageshwaran <[EMAIL PROTECTED]> writes:
> We have our application which uses postgres 7.1.3 version in
> that we use a time functions like now(),timestamp. etc mostly
> timestamp without timezone.We don't need the timezone in that variable
> so we ignore it using timestamp without
Tom Lane wrote:
> Naz Gassiep <[EMAIL PROTECTED]> writes:
>
>> Ok, that's kinda cool. But can I trust those names to not change from
>> version to version?
>>
>
> No, you can't. The reason there is no "nonvolatile" list of timezones
> is that timezones are subject to the whims of politicia
novnov <[EMAIL PROTECTED]> writes:
> It seems that at least in the way I've written the function below, ORDER BY
> is ignored.
Please provide a test case backing up that statement?
regards, tom lane
---(end of broadcast)---
Hi,
We have our application which uses postgres 7.1.3 version in
that we use a time functions like now(),timestamp. etc mostly
timestamp without timezone.We don't need the timezone in that variable
so we ignore it using timestamp without timezone. The same is not
working with high
Naz Gassiep <[EMAIL PROTECTED]> writes:
> Ok, that's kinda cool. But can I trust those names to not change from
> version to version?
No, you can't. The reason there is no "nonvolatile" list of timezones
is that timezones are subject to the whims of politicians, who can and
do change them at the
Lew writes:
Strange? Why? Did you expect a particular statistical distribution? Perhaps
The impression was that one query was returning everything.. and the other
only the records that did not exist in the one table.
you were surprised by the extent of the situation, not thinking there
On Tue, May 29, 2007 at 05:18:42PM -0700, novnov wrote:
> It seems that at least in the way I've written the function below, ORDER BY
> is ignored.
Works here (see below). Can you post a complete example that shows
ORDER BY being ignored? Do you get different results from the
cursor than if you
Alvaro Herrera wrote:
> alvherre=# select * from pg_timezone_names ;
>name | abbrev | utc_offset | is_dst
> --+++
> Africa/Algiers | CET| 01:00:00 | f
> Africa/Luanda
Rodrigo De León escribió:
> Now see:
>
> http://www.postgresql.org/docs/8.2/static/plpgsql-control-structures.html#PLPGSQL-RECORDS-ITERATING
Just a quick reminder that it's usually painful to build "generic"
functions in plpgsql because it's not prepared to deal with column or
table names built
I know that a few versions back PostgreSQL's DBD driver didn't support prepared
statements (the operation was still there, it just wasn't doing the
whole magic),
I imagine that it does today but will investigate further.
There's a very extensive writeup in the documentation of DBD::Pg,
perldoc D
The pg_prepared_statments view will give you information on prepared statments
currently resident in the database backend. I suggest you run your progam
(with a couple of "Enter to continue" breakpoints in the code) and, in another
session, select from pg_prepared_statements and see if the prepare
Is there an advantage to using something like $dbh->prepare($sql) if
the SQL is going to be run once within the scope of the code? The
code block may be run many times in a minute as in a function call ---
while (<>) {
insert_something($_);
}
Will the prepare statement be cached @ the d
It seems that at least in the way I've written the function below, ORDER BY
is ignored. I've seen hints that one can declare the refcursor as a specific
query and that apparently the order by clause there is respected. But I
don't find much by way of examples in the docs or on this list. I will
ev
On May 29, 5:42 pm, [EMAIL PROTECTED] wrote:
> Hi!
>
> Title says it pretty much. I am wondering if there is a short way to
> form a query that retrieves all fields of a table (of which I do not
> know names and number beforehand) except for one (or more, of which I
> know the name(s)). I have stum
Naz Gassiep wrote:
> I've been trying to sort out the answer to this question for a while
> now, I've received different answers from different places.
>
> I'm looking for a definitive non-volatile list of timezones for use in a
> web application. I can't use the OS's time zone list, as changing O
I'm not one of the developers either but this was posted to the
hackers list just today:
We currently have processed 15 of the 40 patches in the patch queue
during our two months in feature freeze. Based on that progress, I
estimate we will enter beta in September.
Kevin.
-Original Messag
I've been trying to sort out the answer to this question for a while
now, I've received different answers from different places.
I'm looking for a definitive non-volatile list of timezones for use in a
web application. I can't use the OS's time zone list, as changing OSes
may cause some listed tim
Is there an advantage to using something like $dbh->prepare($sql) if
the SQL is going to be run once within the scope of the code? The
code block may be run many times in a minute as in a function call ---
while (<>) {
insert_something($_);
}
Will the prepare statement be cached @ the da
Make sure there are some queries been executed on the server from
pg_stat_activity. Make sure your logging level for log files is such that it
can log the queries in db server log file.
In addition to log_statement you can also set log_min_duration_statement to
0 and that will also log queries in
On 30/05/2007 00:11, Guy Rouillier wrote:
I'm not one of the developers, but I found the following information
here http://www.postgresql.org/developer/roadmap:
That's great - thanks.
Ray.
---
Raymond O'Donnell, Director of Music, G
Raymond O'Donnell wrote:
I realise that (i) this is something of a FAQ, and (ii) the definitive
answer is "When it's ready", but when (roughly) is 8.3 planned to be
released?
I'm not one of the developers, but I found the following information
here http://www.postgresql.org/developer/roadmap:
I realise that (i) this is something of a FAQ, and (ii) the definitive
answer is "When it's ready", but when (roughly) is 8.3 planned to be
released?
I'm planning a server upgrade in the next 4-5 months, and a ballpark
guess would be helpful.
Thanks,
Ray.
--
On May 29, 2007, at 15:28 , John D. Burger wrote:
Even ISO country codes are not guaranteed to be stable
I'm not sure where the idea that primary keys must be stable comes
from. There's nothing necessarily wrong with updating a primary key.
All a primary key does is uniquely identify a ro
Hi!
Title says it pretty much. I am wondering if there is a short way to
form a query that retrieves all fields of a table (of which I do not
know names and number beforehand) except for one (or more, of which I
know the name(s)). I have stumbled across the need for this a couple
of time during th
On May 29, 2007, at 14:50 , Oliver Elphick wrote:
On Tue, 2007-05-29 at 13:49 -0500, Michael Glaesemann wrote:
If you're handling more than one country, you'll most likely want to
associate the states with their respective countries.
-- Listing 4
CREATE TABLE countries
(
country_id INTE
show log_statement;
log_statement
---
all
(1 fila)
select pg_stat_get_backend_pid(s.backendid),pg_stat_get_backend_activity(
s.backendid) from (select pg_stat_get_backend_idset() as backendid)s;
pg_stat_get_backend_pid | pg_stat_get_backend_activity
-+--
Richard P. Welty wrote:
but what are the consequences of backing up a WAL file
if the archive process (probably scp in this case) is running
when the backup copy is made? the whole thing won't make it onto
tape, are there any downsides to running a recover with
an incomplete WAL file?
The WAL f
Did you do a reload or restart of the server after doing this change?
what do you see when you do this query? --> show log_statement
--
Shoaib Mir
EnterpriseDB (www.enterprisedb.com)
On 5/29/07, Diego Fernando Arce <[EMAIL PROTECTED]> wrote:
this is a part of postgrsql.conf
log_statement =
this is a part of postgrsql.conf
log_statement = 'all' # none, mod, ddl, all
#log_hostname = off
#---
# RUNTIME STATISTICS
#---
# -
hello, I have a question, does not work monitor stats in postgres 8.1.8
I cannot see querys in execution
help me please..
DiegoF
Cyril VELTER wrote:
> OK, I've the results of my tests :
>
> With the previous code, then message "SOCKERROR" is printed 5 times
> during the
> whole process (100 Gb dump import with psql). There one group of three and
> one
> group of two, but I don't have timestamps and am not su
Oliver Elphick wrote:
You have assumed that state codes are unique integers, but for a
worldwide database that is probably a bad design. The USA knows its
states by two-letter codes, as does India and one should surely not
invent a new set of codes for them. I would make this field a
VARCHAR(3
again thanks - bit of a noob question I know, but it's good to
learn :-)
Well not really since the answer is quite subtle...
You kave two columns A and B.
Say you have index on A, and index on B.
These queries will make direct use of the index :
A=... o
I just did some checks on two seperate indexes c.f. one combined one.
I saw almost no difference between making select statements.
Haven't tried what happens with many updates - makes sense that more
indexes will slow that down though.
again thanks - bit of a noob question I know, but it's good t
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:pgsql-general-
> [EMAIL PROTECTED] On Behalf Of Lew
> Sent: Tuesday, May 29, 2007 6:38 AM
> To: pgsql-general@postgresql.org
> Subject: Re: [GENERAL] optimisation for a table with frequently used
query
>
> danmcb wrote:
> >> SELECT * f
On Tue, 2007-05-29 at 13:49 -0500, Michael Glaesemann wrote:
>
> If you're handling more than one country, you'll most likely want to
> associate the states with their respective countries.
>
> -- Listing 4
> CREATE TABLE countries
> (
> country_id INTEGER PRIMARY KEY
> );
>
> CREATE TABL
danmcb wrote:
SELECT * from my_table where id_1 = x and id_2 = y;
Neither id_1 or id_2 or the combination of them is unique. I expect
this table to become large over time.
PFC wrote:
Create an index on (id_1, id_2), or (id_2,id_1).
What are the strengths and weaknesses compared to creating t
On May 24, 2007, at 8:57 , btober wrote:
I'm not sure it is a bad design. Country has a country_id.
That's the primary key. State has a state_id, and exactly
one country, so really state has a compound primary key,
namely (country_id, state_id).
While each state may have a single state_id and
Another alternative is the SQL editor in pgEdit. It is an interface
for psql and supports tab completion (and several other completion
methods).
John
On May 29, 2007, at 3:50 AM, Hannes Dorbath wrote:
I think this is because readline is not available/broken on win32.
Use the cygwin versio
On May 29, 2007, at 11:42 , ABHANG RANE wrote:
I have a column real[] in my table. I need to create a gist index
on it. But the error is obvious as follows.
ERROR: data type real[] has no default operator class for access
method "gist"
HINT: You must specify an operator class for the ind
ABHANG RANE <[EMAIL PROTECTED]> writes:
> I have a column real[] in my table. I need to create a gist index on
> it.
An index to do what --- ie, what queries do you hope the index will help
with?
regards, tom lane
---(end of broadcast)
On May 29, 2007, at 12:41 , Jeff Davis wrote:
"CREATE CONSTRAINT TRIGGER is used within CREATE TABLE/ALTER TABLE and
by pg_dump to create the special triggers for referential
integrity. It
is not intended for general use."
Is there harm in using constraint triggers outside of pg_dump? It
Jeff Davis <[EMAIL PROTECTED]> writes:
> Is there harm in using constraint triggers outside of pg_dump? It seems
> like deferrable constraints would be useful for many applications, but
> the docs imply that it's a bad idea.
It's un-deprecated as of 8.3.
regards, tom lane
Albe Laurenz wrote:
Advantages of many databases:
- Each database is smaller.
- No danger of one user accessing another user's data (because of
misconfigured permissions and similar).
- Guaranteed independence of each user's data.
- More scalable: If you decide that one machine or one cluster
Hi All,
I have set up a jdbc connection using jdeveloper(on my windows
machine). I am connecting to a postgres database on a linux machine. I
used the Database connections are shown in the Navigator, in the
Connections | Database folder. I tested the connection and it said
"Success!" in the d
On Tue, 2007-05-29 at 10:59 -0500, Michael Glaesemann wrote:
> On May 25, 2007, at 3:22 , Andrus wrote:
>
> > CREATE TRIGGER puhkus_sequenced_trigger BEFORE INSERT OR UPDATE ON
> > puhkus
> >FOR EACH ROW EXECUTE PROCEDURE puhkus_sequenced_pkey();
>
> You can also use CREATE CONSTRAINT TRIGG
Joshua wrote:
I checked the table and found that none of my fields in the SELECT
statement contain NULLs.
Any other suggestions?
Why are you using such constructions in your query: ',' || ',' || ',' ?
May be this set of commas makes you think that some of your fields are
empty? Do you have e
On Wednesday 23 May 2007 1:04 pm, George Pavlov wrote:
> Hoping to resurrect this thread. I am seeing more and more of
> this as the database gets more usage and it really messes up
> query log analysis.
>
>
> A quick summary: When I posted this was getting corrupted
> query log entries. I still am
Joshua wrote:
I checked the table and found that none of my fields in the SELECT
statement contain NULLs.
Any other suggestions?
Please post the queries you used.
--
Richard Huxton
Archonet Ltd
---(end of broadcast)---
TIP 9: In versions b
I checked the table and found that none of my fields in the SELECT
statement contain NULLs.
Any other suggestions?
PFC wrote:
SELECT 'PV.LINEITEM:' || partnum || ',' || round(onhand) || ',' ||
round(qm5) || ',' || round(lsm4) || ',' || ',' || ',' ||
round(onorder) || ',' || ',' || ',' || bino
Hi,
I have a column real[] in my table. I need to create a gist index on
it. But the error is obvious as follows.
ERROR: data type real[] has no default operator class for access
method "gist"
HINT: You must specify an operator class for the index or define a
default operator class for the
SELECT 'PV.LINEITEM:' || partnum || ',' || round(onhand) || ',' ||
round(qm5) || ',' || round(lsm4) || ',' || ',' || ',' || round(onorder)
|| ',' || ',' || ',' || binone || ',' || ',' || round(backorderqty) ||
',' || ',' || round(onhold) || ',' || ',' || ',' || ',' || ',' || ',' ||
',' || r
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
> Anyone have an implementation of the Jaro Winkler fuzzy string
> matching algo for Postgres?
I've not seen one, but it should be a Simple Matter of Programming
to whip something up in Perl or C.
- --
Greg Sabino Mullane [EMAIL PROTECTED]
P
2. Is there any performance or other advantage to using PL/pgsql over
Pl/Perl or Python?
Yes, if you want to loop over large amounts of data (FOR row IN SELECT)
plpgsql will be faster since it does not have to convert the data from
postgres to python/perl format.
---
Hello,
I am new to this list and have been working with PostgreSQL since January.
Here is my problem, I hope someone here has some experience or can point
me in the right direction. I am writing the following query for a C#
program I am writing:
SELECT 'PV.LINEITEM:' || partnum || ',' || rou
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
> Which list is the most appropriate for proposing features
> and ideas for postgres ?
Generally, -general. If the proposal is very technically specific,
you might prefer -hackers instead.
- --
Greg Sabino Mullane [EMAIL PROTECTED]
PGP Key:
On May 25, 2007, at 3:22 , Andrus wrote:
CREATE TRIGGER puhkus_sequenced_trigger BEFORE INSERT OR UPDATE ON
puhkus
FOR EACH ROW EXECUTE PROCEDURE puhkus_sequenced_pkey();
You can also use CREATE CONSTRAINT TRIGGER, which allows you to have
deferrable constraints. This is useful if stat
On May 29, 2007, at 10:30 , Michael Glaesemann wrote:
You can take a look in the pg_proc table, which is part of the
system catalog, to see which functions are installed.
file:///usr/local/pgsql/pgsql-8.2.0/doc/html/catalog-pg-proc.html
Ha! That link won't be very helpful, will it :)
This
On May 29, 2007, at 9:48 , Robert James wrote:
I'd like to be able to detect if a record has associations. I don't
want to actually delete it, just know if it could be deleted. (This
is to build an intelligent menu on a GUI)
On 5/29/07, Albe Laurenz <[EMAIL PROTECTED]> wrote: > Is there
On Tue, 29 May 2007, Ron Johnson wrote:
1) The data contains the price of the underlying stock, the strike price
of the option, and the option premium. From this I can calculate the
"cost basis" and the "maximum potential profit", which are elements I
would like to be able to SELECT and ORDER.
Robert James wrote:
1. How can I get a list of available functions (ie, user defined or
contrib)
using SQL?
To see how PG does it:
psql -E
\df
2. Is there any performance or other advantage to using PL/pgsql over
Pl/Perl or Python?
1. It's more likely to be available (not relevant if y
On May 29, 2007, at 9:49 , Robert James wrote:
1. How can I get a list of available functions (ie, user defined or
contrib) using SQL?
You can take a look in the pg_proc table, which is part of the system
catalog, to see which functions are installed.
file:///usr/local/pgsql/pgsql-8.2.0/
am Tue, dem 29.05.2007, um 10:48:21 -0400 mailte Robert James folgendes:
> I'd like to be able to detect if a record has associations. I don't want to
> actually delete it, just know if it could be deleted. (This is to build an
> intelligent menu on a GUI)
Set a savepoint, try to delete the recor
On 05/24/07 15:20, cjl wrote:
PG:
Sorry it's taken so long for anyone to answer you, but it appears
that some emails were hung up for a while.
I am playing around with some historical stock option data, and have
decided to use a database to make my life easier. The data is "end-of-
day" fo
On 05/29/07 04:02, Albe Laurenz wrote:
I have an application with some hundreds users, each one
having the same
data definitions, and each one storing up to 2 GB of data.
A user have just access to his own data. His data will have its own
tablespace.
Therefore, it seems to me I have a choice
On 05/29/07 09:48, Robert James wrote:
I'd like to be able to detect if a record has associations. I don't want
to actually delete it, just know if it could be deleted. (This is to
build an intelligent menu on a GUI)
Are you wanting to know this in a generic way or for a specific
database?
1. How can I get a list of available functions (ie, user defined or contrib)
using SQL?
2. Is there any performance or other advantage to using PL/pgsql over
Pl/Perl or Python?
I'd like to be able to detect if a record has associations. I don't want to
actually delete it, just know if it could be deleted. (This is to build an
intelligent menu on a GUI)
On 5/29/07, Albe Laurenz <[EMAIL PROTECTED]> wrote:
> Is there anyway to know if a DELETE will violate an FK
> withou
On 5/23/07, lest mitsui <[EMAIL PROTECTED]> wrote:
Does anyone have an idea on where to turn for information/research papers on
online application upgrades? I am well-versed in offline upgrades, but I am
now faced with customer requirements for transparent software upgrades of
our application se
On 5/29/07, Albe Laurenz <[EMAIL PROTECTED]> wrote:
> I have an application with some hundreds users, each one
> having the same
> data definitions, and each one storing up to 2 GB of data.
> A user have just access to his own data. His data will have its own
> tablespace.
>
> Therefore, it seems
On May 29, 2007, at 6:09 , Sahoo, Ranjan Rashmi wrote:
I am new to this postgres and now struggling with creation of a
user specific to a database, so the user can login to that specific
database and can do the activities as per the given credentials and
can not touch or see any informa
On Thu, May 24, 2007 at 02:12:01PM -0300, Simon De Uvarow wrote:
> may someone tell me where to put the source of the tool for free access?
You could set up a project on pgfoundry:
http://pgfoundry.org/
Better still, contact the maintainers of some of the already-started
monitoring projects, and
I
would
like
to
leave
the
list
...
--
Acelerador POP
Acelere a sua conexão discada em até 19 x. Use o Acelerador POP. É grátis, pegue já o seu.
http://www.pop.com.br/acelerador
Nikolay Samokhvalov wrote:
On 22 May 2007 17:47:46 -0700, EMS Database Management Solutions
(SQLManager.net) <[EMAIL PROTECTED]> wrote:
We, here at EMS Database Management Solutions, are pleased to announce
SQL Manager 2007 for PostgreSQL - the new major version of the
powerful PostgreSQL admini
"David Fetter" <[EMAIL PROTECTED]> writes:
> On Sun, May 27, 2007 at 01:23:48PM -0400, ABHANG RANE wrote:
>> Hi,
>> I need to use the cube operator in postgresql 8.2. Whats the syntax
>> for the cube or a link to the documentation which clearly explains
>> how to use cube operator syntactically.
Thanks!
---(end of broadcast)---
TIP 1: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to [EMAIL PROTECTED] so that your
message can get through to the mailing list cleanly
> Is there anyway to know if a DELETE will violate an FK
> without actually trying it?
I don't know what you mean by 'without trying it', but does the
following answer your question?
CREATE TABLE a (id integer PRIMARY KEY);
CREATE TABLE b (id integer PRIMARY KEY,
a_id integer NOT NULL CON
> I have an application with some hundreds users, each one
> having the same
> data definitions, and each one storing up to 2 GB of data.
> A user have just access to his own data. His data will have its own
> tablespace.
>
> Therefore, it seems to me I have a choice between "one database per
On May 21, 2007, at 8:25 AM, David Fetter wrote:
On Mon, May 21, 2007 at 02:21:52PM +0400, Akmal Akmalhojaev wrote:
For example I have a role ID1 with members ID2 and ID3.
Role ID2 has also members ID4 and ID5. It means that roles ID4 and
ID5 are
members of ID1.
The question: Is there any fun
On May 21, 2007, at 1:44 PM, PFC wrote:
Which list is the most appropriate for proposing features and
ideas for postgres ?
Well, I think hackers tend to pay more attention to -hackers. OTOH,
in order to get something accepted you've got to show why it should
be accepted, so if it's a bran
On May 17, 2007, at 4:14 AM, Sorin N. Ciolofan wrote:
It seems that the problem was the max_fsm_relations parameter which
was increased
I still do not understand why in the past I received always the
message:
ERROR: out of shared memory
Is this an appropriate message for the need for increa
On 29.05.2007 09:26, Robert James wrote:
Any way to get psql Tab Completion in Windows?
I think this is because readline is not available/broken on win32. Use
the cygwin version of psql.
--
Regards,
Hannes Dorbath
---(end of broadcast)---
TIP
On 23.05.2007 18:55, lest mitsui wrote:
Does anyone have an idea on where to turn for
information/research papers on online application
upgrades? I am well-versed in offline upgrades, but I am now faced with
customer requirements for transparent software upgrades of our application
server/servi
Any way to get psql Tab Completion in Windows?
93 matches
Mail list logo
