[ADMIN] Re: Dual-CPU RAID database server: What Hardware ?

2001-03-26 Thread J.H.M. Dassen (Ray)

Paul BOYER <[EMAIL PROTECTED]> wrote:
>As we are trying to determine the better hardware choice for a database
>server aiming to serve from 1GB up to 15GB with less than 10 connections
>from application servers, with rather intensive requests, operating
>__24x7__ under Linux.

For 24x7 availability, don't forget to look into UPSes, journaling
filesystems and good backups procedures and technology.

>-> Q3: Any experience for a database under Linux SMP with those cards ?

One of our database servers is a dual PIII-850 with a Mylex RAID controller
("DAC1164P PCI RAID Controller" according to dmesg) running PostgreSQL 7.0.3
under Debian GNU/Linux 2.2 with kernel 2.2.19pre17 without any problems
whatsoever (300+ days uptime under the previous kernel before it was
moved to a different location).

HTH,
Ray
-- 
Pinky, Are You Pondering What I'm Pondering?
I think so Brain, but if we give peas a chance, won't the lima beans feel
left out? 
Pinky and the Brain in "All You Need Is Narf"


---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]



[ADMIN] Error migrating from Postgres 6.3.2

2001-03-26 Thread Luis Martínez

I am migrating from postgres 6.3.2 to 7.1 Beta6. After performing a 
pg_dumpall > backupfile on my old pgsql version I try to restore the data 
with psql -d template1 -f backupfile, but I get "Unable to locate type name 
'char16' in catalog" on tables such as onek, tenk1 ,tenk2

What is the problem?
Shoul I replace this type with some other?

Thanks for your help in advance,

Luis,
-- 
--
Luis   Javier Martínez Martínez   Instituto Tecnológico de Informática
Telf:  34 (9) 63 87 72 33 Universidad  Politécnica de Valencia
Email: [EMAIL PROTECTED]Camino de Vera s/n46071 Valencia
--

---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster



Re: [ADMIN] Error migrating from Postgres 6.3.2

2001-03-26 Thread Tom Lane

Luis =?iso-8859-1?q?Mart=EDnez?= <[EMAIL PROTECTED]> writes:
> I am migrating from postgres 6.3.2 to 7.1 Beta6. After performing a 
> pg_dumpall > backupfile on my old pgsql version I try to restore the data 
> with psql -d template1 -f backupfile, but I get "Unable to locate type name 
> 'char16' in catalog" on tables such as onek, tenk1 ,tenk2

Those look suspiciously like tables in the regression tests; I bet you
ran the regression tests in 6.3.2 and never bothered to clean up afterwards.
Consider just cutting the whole regression-database section out of the
backupfile.

regards, tom lane

---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]



RE: [ADMIN] Sorry, too many clients already

2001-03-26 Thread Peter Eisentraut

Mahesh Guleria writes:

> Use these commands :
>
> ps -ef|grep post
> kill -9 [process id]

Don't do that, it might destroy your database.  SIGTERM should be used to
kill individual backends.

-- 
Peter Eisentraut  [EMAIL PROTECTED]   http://yi.org/peter-e/


---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])



Re: [ADMIN] Sorry, too many clients already

2001-03-26 Thread Ross J. Reedstrom

On Mon, Mar 26, 2001 at 12:00:03PM +0530, Mahesh Guleria wrote:
> Use these commands :
>  
> ps -ef|grep post
> kill -9 [process id]

No, no, a thousand times, no! Haven't you seen this appended to
messages on the lists?

-
TIP 4: Don't 'kill -9' the postmaster
-

If you kill -9 the postmaster, various shared memory structures never
get cleaned up correctly, so you risk damaging the database. 

What Subra is experiencing is the maximum simultaneous client limit,
which defaults to 32 7.X series (it was 64 in 6.X). You can affect this
with the -N parameter to postmaster, when it starts up. If you increase
it, you may need to increase -B (shared buffers), as well. Check the
pg_ctl man page. Each installation uses a slightly different way to
startup the postmaster, so you'll have to track that down yourself.

Ross

> 
> Mahesh Guleria
> 
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Subra
> Radhakrishnan
> Sent: Friday, March 23, 2001 4:31 PM
> To: [EMAIL PROTECTED]
> Subject: [ADMIN] Sorry, too many clients already
> 
> 
> Hi All,
> 
> We are getting an error every now and then in
> PostgreSQL saying 'Sorry, too many
> clients already'. We have recently installed the
> database version 7.0.3. Calls to
> database are made from our application written in
> Java, JSP. Can anyone help
> me with this?
> 
> Thanks,
> 
> Subra
> _
> 
> 
> __
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail. 
> http://personal.mail.yahoo.com/
> 
> ---(end of broadcast)---
> TIP 4: Don't 'kill -9' the postmaster
> 
> ---(end of broadcast)---
> TIP 4: Don't 'kill -9' the postmaster

---(end of broadcast)---
TIP 3: 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



RE: [ADMIN] Sorry, too many clients already

2001-03-26 Thread David Lizano

At 19.42 26/3/01 +0200, you wrote:
>Mahesh Guleria writes:
>
> > Use these commands :
> >
> > ps -ef|grep post
> > kill -9 [process id]
>
>Don't do that, it might destroy your database.  SIGTERM should be used to
>kill individual backends.
>
>--
>Peter Eisentraut  [EMAIL PROTECTED]   http://yi.org/peter-e/
>
You can do it like this:

 killall -TERM postmaster
or
 kill -TERM `pidof postmaster`

The SIGTERM signal can be trapped by a process and when is invoked  the 
process can close all the opened resources like files and pipes.




~~
David Lizano - Director área técnica
correo-e: [EMAIL PROTECTED]

I Z A N E T - Servicios integrales de internet.
web: http://www.izanet.com/
Dirección: C/ Checa, 57-59, 3º D - 50.007 Zaragoza (España)
Teléfono: +34 976 25 80 23Fax: +34 976 25 80 24
~~


---(end of broadcast)---
TIP 6: Have you searched our list archives?

http://www.postgresql.org/search.mpl



Re: [ADMIN] Sorry, too many clients already

2001-03-26 Thread Tom Lane

"Ross J. Reedstrom" <[EMAIL PROTECTED]> writes:
> On Mon, Mar 26, 2001 at 12:00:03PM +0530, Mahesh Guleria wrote:
>> Use these commands :
>> 
>> ps -ef|grep post
>> kill -9 [process id]

> No, no, a thousand times, no! Haven't you seen this appended to
> messages on the lists?

> TIP 4: Don't 'kill -9' the postmaster

Actually, as of 7.1RC1 that issue is gone away; it's still not
considered good practice to kill -9 any postgres process, but
at least you can't shoot yourself in the foot that way anymore.

regards, tom lane

---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster



Re: [ADMIN] psql doesn't recognize some keys

2001-03-26 Thread Björn Lundin

Simone Tellini wrote:

> Hi,
> 
> I've got a problem with psql on my home box: it doesn't recognize the
> arrow keys nor the backspace and probably other keys as well.
> 
> I've compiled it from the sources, as I did on another server, where it
> works correctly. The termcaps should be the same on both the machines,
> so I'm a bit clueless now (the only difference is that the working one
> is using Mandrake 7.0, while my home box has 7.2)
> 
> Is there anything I could check or any switch to set?
> 

this is an issue I've also experienced with Mandrake 7.2, but at some page 
at postgre's website, this was listed as a known error.
Solution is to run 7.0.3, and it's easy to migrate.

Björn


---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster



[ADMIN] Sorry, too many clients already

2001-03-26 Thread Subra Radhakrishnan

Thanks everybody,

Majority of the mails I received asked me to start the
postmaster as follows and it seems to work so far:

/usr/local/pgsql/postmaster -i -B 256 -N 128 -D
/usr/local/pgsql/data/


Thanks again,

Subra


__
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/

---(end of broadcast)---
TIP 6: Have you searched our list archives?

http://www.postgresql.org/search.mpl