repmgr standby clone fails

From: Mariel Cherkassky <mariel(dot)cherkassky(at)gmail(dot)com>
To: pgsql-admin(at)lists(dot)postgresql(dot)org
Subject: repmgr standby clone fails
Date: 2019-02-18 14:15:02
Message-ID: CA+t6e1k3P6YC3iSan0oxpwqF-8uBxrySBGqKHpV4VhSYcNnPZg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

Hi,
I have 2 instances of postgres 9.6 and repmgr v4.2.
I registered my primary as a primary and I tried to clone my standby via
repmgr standby clone but I'm getting the following error every time :

WARNING: no record exists for designated upstream node 1
HINT: you'll need to create the replication slot ("repmgr_slot_2") manually
ERROR: no database connection available

I queried the repmgr.nodes ,repmgr.events and pg_replication_slots on both
nodes :

*Primary*
repmgr=# select * from repmgr.events;
-[ RECORD 1 ]---+------------------------------
node_id | 1
event | cluster_created
successful | t
event_timestamp | 2019-02-16 14:25:28.094412+00
details |

-[ RECORD 2 ]---+------------------------------
node_id | 1
event | primary_register
successful | t
event_timestamp | 2019-02-16 14:25:28.101808+00
details |

repmgr=# select * from pg_replication_slots ;
-[ RECORD 1 ]-------+--------------
slot_name | repmgr_slot_2
plugin |
slot_type | physical
datoid |
database |
active | f
active_pid |
xmin |
catalog_xmin |
restart_lsn | 419/50000035
confirmed_flush_lsn |

repmgr=# select * from repmgr.nodes ;
-[ RECORD 1
]----+-------------------------------------------------------------------------------------------------
node_id | 1
upstream_node_id |
active | t
node_name | nodeA
type | primary
location | default
priority | 100
conninfo | host=10.10.1.10 user=repmgr dbname=repmgr
connect_timeout=10
repluser | repmgr
slot_name | repmgr_slot_1
config_file | /etc/repmgr/9.6/repmgr.conf

*Secondary :*
repmgr=# select * from repmgr.events;
(Same as in primary)
-[ RECORD 1 ]---+------------------------------
node_id | 1
event | cluster_created
successful | t
event_timestamp | 2019-02-16 14:25:28.094412+00
details |

-[ RECORD 2 ]---+------------------------------
node_id | 1
event | primary_register
successful | t
event_timestamp | 2019-02-16 14:25:28.101808+00
details |

repmgr=# select * from repmgr.nodes;
-[ RECORD 1
]----+-------------------------------------------------------------------------------------------------
node_id | 1
upstream_node_id |
active | t
node_name | nodeA
type | primary
location | default
priority | 100
conninfo | host=10.10.2.10 user=repmgr dbname=repmgr
connect_timeout=10
repluser | repmgr
slot_name | repmgr_slot_1
config_file | /etc/repmgr/9.6/repmgr.conf

repmgr=#select count(*) from pg_replication_slots;
count
-------
1
(1 row)

my repmgr.conf contains :
use_replication_slots=true
node_name=nodeA
node_id=1
and more parameters...

Correct me if I'm wrong but the error means that the secondary cant find a
record for the primary db. However, as you can see repmgr`s tables are
populated with data about the primary. Any idea what might cause it ?
I didnt try it yet, but I thought about registering the primary again and
retrying the standby clone. Does someone familiar with the issue or with
other solution ?

Browse pgsql-admin by date

  From Date Subject
Next Message Debraj Manna 2019-02-18 14:21:09 Installing postgres10 on Ubuntu 16
Previous Message soumik.bhattacharjee 2019-02-18 09:01:10 RE: Table Drop Error