On 5/27/19 9:59 PM, Tom K wrote:
Hey Guy's,
I'm running Patroni w/ PostgreSQL 10, ETCD, Haproxy and Keepalived on 3
RHEL 7.6 VM's. Every now and then the underlying storage crashes taking
out the cluster. On recovery, PostgreSQL tends to come up while other
databases just blow up. That is
--> Function ' filter_id ' filters the ID's based on some conditions.
--> Input is set of ID's. (Not directly taking the input since there is no
provision to pass multiple rows to a function)
create function filter_id()
return table (id bigint)
begin
--> Assuming input table is already created #t
On 5/28/19 7:36 AM, RAJIN RAJ K wrote:
--> Function ' filter_id ' filters the ID's based on some conditions.
--> Input is set of ID's. (Not directly taking the input since there is
no provision to pass multiple rows to a function)
To be honest I cannot follow what you are trying to achieve bel
On 5/28/19 8:06 AM, RAJIN RAJ K wrote:
Please reply to list also.
Ccing list.
Thanks for the response.
CTE is not useful in my case. Here i want to pass the table to a
function and get the filtered results back from the function.
I tried few but not use full.
1. Pass table input --> Ref curs
Adrian, the current nagios alerting does the following:
postgres=# SELECT count(*) from pg_freespacemap_relations;
count
---
7
(1 row)
and this is the snippet from our config:
max_fsm_pages = 600
# (change requires restart)
max_fsm_relations =
What is the impact of fsm_relatiosn being maxed out?
From: Julie Nishimura
Sent: Tuesday, May 28, 2019 11:11 AM
To: Adrian Klaver; pgsql-general@lists.postgresql.org; pgsql-general
Subject: Re: POSTGRES_FSM_RELATIONS CRITICAL: DB control fsm relations used:
79569
On 28 May 2019 20:20:10 CEST, Julie Nishimura wrote:
>What is the impact of fsm_relatiosn being maxed out?
https://www.postgresql.org/docs/8.2/runtime-config-resource.html#RUNTIME-CONFIG-RESOURCE-FSM
Please no top-posting with fullquote.
Regards, Andreas
--
2ndQuadrant - The PostgreSQL Supp
On 5/28/19 11:20 AM, Julie Nishimura wrote:
What is the impact of fsm_relatiosn being maxed out?
It has been awhile since I thought about this as these settings are no
longer relevant as of 8.4:
https://www.postgresql.org/docs/8.4/release-8-4.html
"
Track free space in separate per-relati
Adrian, I am trying to avoid to do any tweaking to this legacy system that
nobody knows well (we inherited it recently).
Do you think it might help if we possibly drop old tables (I assume their
indices will be removed too), so the overall number of objects will go down?
Thanks a lot
__
On 2019-May-28, Julie Nishimura wrote:
> Adrian, I am trying to avoid to do any tweaking to this legacy system that
> nobody knows well (we inherited it recently).
> Do you think it might help if we possibly drop old tables (I assume their
> indices will be removed too), so the overall number of
On 5/28/19 2:53 PM, Julie Nishimura wrote:
Please post to list also.
Ccing list
I am sorry, did not mean to send it yet.
#--
# RESOURCE USAGE (except WAL)
#-
On Tue, May 28, 2019 at 9:53 AM Adrian Klaver
wrote:
> On 5/27/19 9:59 PM, Tom K wrote:
> > Hey Guy's,
> >
> >
> > I'm running Patroni w/ PostgreSQL 10, ETCD, Haproxy and Keepalived on 3
> > RHEL 7.6 VM's. Every now and then the underlying storage crashes taking
> > out the cluster. On recovery
12 matches
Mail list logo
