-
Notifications
You must be signed in to change notification settings - Fork 797
Description
Hi --
Environment: pgadmin 6.13 running in the official docker image, postgres 13 (also the official docker image) (note: I have seen this issue with earlier versions as well). This problem seems to have shown up in the last several months, first reported to me on August 9 (postgres 9.6 at the time, and the "latest" pgadmin) ... i.e. we were not experiencing it earlier in the year (Postgres 9.6 then, and whatever the "latest" pgadmin docker image was at the time, sorry I don't have specifics on that.)
##
postgres:
image: postgres:13
container_name: postgresdb
environment:
- POSTGRES_USER=qlik
- POSTGRES_PASSWORD=${PASSWORD}
- POSTGRES_DB=qlikdb
- PGDATA=/var/lib/postgresql/data
restart: always
volumes:
- postgres-data:/var/lib/postgresql
ports:
- 5432
networks:
- replnet
command: -c wal_level=logical -c max_replication_slots=3 -c max_wal_senders=3 -c wal_sender_timeout=60s
##
## Start PGADMIN, listening on port 5430. This will run on the replnet network
## and will allow us to browse the postgres database.
##
pgadmin:
image: dpage/pgadmin4:latest
container_name: pgadmin
depends_on:
- postgres
environment:
- PGADMIN_DEFAULT_EMAIL=pgadmin@qlik.com
- PGADMIN_DEFAULT_PASSWORD=${PASSWORD}
- PGADMIN_LISTEN_PORT=5430
- PGADMIN_CONFIG_ENABLE_PSQL=True
restart: always
volumes:
- pgadmin-data:/var/lib/pgadmin
- /testdrive/pgadmin.json:/pgadmin4/servers.json
ports:
- 5430:5430
networks:
- replnet
##I'm running into the above issue when I open a "Query Tool". Immediately upon opening the tool, I get a message "Not Acceptable", and if I hover over the connection icon at the top, it shows the message "Transaction ID not found in the session".
If I request completion help (CTRL-space), I then also see the "Transaction ID not found" in the Messages area:
Note: if I use the built-in queries from the left pane, they work without issue. The problem seems to be isolated to the query tool.
I've spent many hours trying to find a solution to this ... I've found some hint that there can be challenges with "shared" connections, but as best I could tell that issue was with Apache on Linux and not related to the gunicorn server used in the pgadmin docker image. I've also seen suggestions to disconnect / reconnect from the database. No luck there. And I've seen suggestions related to connection timeouts. That didn't seem to have an impact either.
Any guidance would be appreciated.