Re: Master-Slave error: the database system is starting up

2018-02-02 Thread Daniel Westermann
> http://www.postgresql-archive.org/Master-Slave-error-the-database-system-is-starting-up-td6004044.html > > >please execute "select * from pg_stat_replication;" and "select * from >pg_replication_slots ;" on the master and show us the output. >Please check also the logs on the standby. Did

Re: PG Sharding

2018-02-02 Thread Matej
I thought that this kind of solution had high latency and bad OLTP capabilities (low trans/second)? Analytics is not a high priority. BR 2018-02-01 19:01 GMT+01:00 Dan Wierenga : > > > On Wed, Jan 31, 2018 at 7:48 PM, Steven Lembark > wrote: > >> On Mon, 29 Jan 2018 15:34:18 +0100 >> Matej wro

FATAL: failed to create a backend connection

2018-02-02 Thread Vikas Sharma
Hi There, We are using Postgresql 9.3 with pgpool-II-93-3.5.10. One Master replicating to 3 Standbys using streaming replication. Pgpool is used for load balancing only. Lately we are seeing below on application servers. Caused by: org.postgresql.util.PSQLException: FATAL: failed to create a bac

Re: FATAL: failed to create a backend connection

2018-02-02 Thread Pavan Teja
On Feb 2, 2018 5:57 PM, "Vikas Sharma" wrote: Hi There, We are using Postgresql 9.3 with pgpool-II-93-3.5.10. One Master replicating to 3 Standbys using streaming replication. Pgpool is used for load balancing only. Lately we are seeing below on application servers. Caused by: org.postgresql.u

Increasing size of array items without locking

2018-02-02 Thread Rowan Seymour
Hi. I'm trying to increase the size of items in an array column on a large production database but it's locking the table. Is it correct than when you make a field larger, it shouldn't lock because it doesn't have to check the constraint on every row? For example expanding a VARCHAR column from 255

Ensure extension exists

2018-02-02 Thread Geoffrey Hoffman
I want to use the pgcrypto extension with a particular database. Does it exist permanently and survive a server reboot? Or if not, how do you ensure it runs when the server restarts? I ask because in my Docker development environment I have to run it every time I restart the container, and just

Re: Master-Slave error: the database system is starting up

2018-02-02 Thread Sargez
Hello Thenk you very much fot your interest! I am very grateful! There are my results: postgres=# select * from pg_stat_replication; pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | backend_xmin | state | sent_location

Postgres Planner "Inconsistency"?

2018-02-02 Thread Renzo Bertuzzi
Hi Can you help me with this situation please?? I have a strange problem with a query where the planner only uses and index if I use a constant value, but if I use a subquery it will prefer a seq scan. I have table "sample_table" with columns id serial primary key, and int_flag, with an index on

Re: Unexpected ErrorMessage reply to SSLRequest

2018-02-02 Thread Peter Eisentraut
On 1/31/18 17:01, Michał Muskała wrote: > I'm one of the maintainers of postgrex [1] - an Elixir client for > Postgresql. We implement the wire protocol and connect directly through > tcp. We got a bug report indicating we got an unexpected ErrorMessage > packet in reply to SSLRequest. What did th

Re: Ensure extension exists

2018-02-02 Thread Peter Eisentraut
On 2/1/18 19:47, Geoffrey Hoffman wrote: > I want to use the pgcrypto extension with a particular database. Does it > exist permanently and survive a server reboot? Or if not, how do you ensure > it runs when the server restarts? Once an extension is installed, it stays installed, just like any

Re: Ensure extension exists

2018-02-02 Thread geoff hoffman
On Feb 2, 2018, at 12:38 PM, Peter Eisentraut wrote: > > On 2/1/18 19:47, Geoffrey Hoffman wrote: >> I want to use the pgcrypto extension with a particular database. Does it >> exist permanently and survive a server reboot? Or if not, how do you ensure >> it runs when the server restarts? > >

Re: Postgres Planner "Inconsistency"?

2018-02-02 Thread Laurenz Albe
Renzo Bertuzzi wrote: > I have a strange problem with a query where the planner only uses and index > if I use a constant value, but if I use a subquery it will prefer a seq scan. > > I have table "sample_table" with columns id serial primary key, and int_flag, > with an index on int_flag. > I in