No migration found when starting Torq from binary
yo-mr-white opened this issue · comments
I installed latest version (1.5.1) from binary (linux-arm64). When I want to start it (./torq-core -c .torq.conf start
), it successfully connects to the DB, but then I got this error:
{"level":"info","time":"2024-02-04T16:56:16+01:00","message":"Migrations might take a while. Please be patient."}
{"level":"error","error":"Running up migration: no migration found for version 0: read down for version 0 .: file does not exist","time":"2024-02-04T16:56:16+01:00","message":"Torq could not migrate the database."}
{"level":"debug","time":"2024-02-04T16:56:16+01:00","message":"RootService cancellation requested"}
{"level":"debug","time":"2024-02-04T16:56:16+01:00","message":"RootService updating current state to Inactive (due to failure)"}
{"level":"debug","time":"2024-02-04T16:56:16+01:00","message":"Cancel function called."}
{"level":"info","time":"2024-02-04T16:56:17+01:00","message":"Torq is dead."}
Any ideas how to solve it?
Are you sure the file .torq.conf
is readable and in the same folder as the binary you are running?
It's not a typo? ./torq-core -c ./torq.conf start
Yes, I am sure. The name and also access rights are correct. If I change the DB password in config file to something else, then I am getting different error, that the Torq couldn't connect to the DB, so there is no problem with reading the config file.
OK, thank you for reporting this. We will have to look into it. Potentially there was a problem including the migration files in the binary.
You should also have logs from the timescaledb right? No extra clues in there?
Same issue here, starting from scratch, with a vanilla and empty postgresql, on a raspbian.
Postgresql will not work, you need timescaledb. If it's also not working for timescaledb could you check timescaledb logs for me please.