-
J james moved this topic from Support
-
J james marked this topic as a regular topic
-
-
When I try to connect nc -v mydb.mysql.database.azure.com 3306 from my Cloudron server to the Azure DB, it connects successfully.
I did not find any error's in Logs. Can you help me debug this?
-
Hello @webliska
nc
only checks the ports' availability.
Unfortunately, I can't provide full support for Azure Database for MySQL.
But, you should check out the following article https://learn.microsoft.com/en-us/azure/mysql/flexible-server/concepts-networking-publicFrom this page:
Only the IP addresses you allow have permission to access your Azure Database for MySQL Flexible Server instance. By default, no IP addresses are allowed. You can add IP addresses when initially setting up your server or after your server has been created.
The user that tries to connect is also important. Not every MySQL User is allowed to access the server from anywhere.
Also, this page: https://learn.microsoft.com/en-us/azure/mysql/how-to-create-users#create-a-nonadmin-userNote: on
CREATE USER 'username'@'host'
.
In many guides for'host'
the value%
is used to allow everything, which is considered a bad practice.
A good example is here: https://www.digitalocean.com/community/tutorials/how-to-allow-remote-access-to-mysql#step-2-create-a-remote-mysql-userMaybe these tips already help you resolve your issue.
-
Thanks for the links, but the MySQL connection itself is fully working — both inside and outside the container:
- mysql CLI connects fine using TLS (--ssl-mode=VERIFY_CA)
- Python’s MySQLdb driver inside the container runs SELECT NOW() successfully
- TLS handshake works (openssl s_client …)
So there is no network or credentials problem.
The actual issue is with Cloudron’s Redash package workers:
- supervisorctl status shows only workers for periodic emails default queues.
- Redis queues queries and schemas keep growing (LLEN rq:queue:queries > 0), but no worker consumes them.
- Redash UI stays stuck at “Query in queue…” because no worker for queries queue is running.
Normally, Redash workers should be started with:
QUEUES=queries,schemas,scheduled,emails,periodic,default
But the Cloudron package doesn’t expose Environment settings in the UI and prevents us from reading /proc/$pid/environ inside the container, so we can’t start a worker manually either (REDASH_COOKIE_SECRET missing).
This is 100% a packaging/configuration issue with the Cloudron Redash app, not with MySQL or Azure.
Please update the Redash app to:
- Start at least one worker for queries,schemas,scheduled,emails,periodic,default queues; OR
- Expose QUEUES and other REDASH_* env vars in the Cloudron App UI so we can start workers ourselves.
Until then, any query will remain stuck in the queue because no worker is available to process it.
So if you can help me with that.
Thanks!