Hi.
I've been struggling with a problem for four days now.
It shouldn't be a problem at all, as I've successfully completed the process several times over the past few years.
My Traccar runs in Docker - the database is on Microsoft AZURE ("Flexible Azure Database for MySQL Server"). Everything has been running smoothly for a long time - not a single Traccar update has ever caused any problems. My task now is to migrate the entire database from the aforementioned AZURE service to a MariaDB, which also runs in Docker.
I perform the following steps:
Complete dump of the existing DB:
mysqldump -h XXXXX.mysql.database.azure.com -u sven -p traccar > traccar_dump.sql
Stop the Docker container "traccar"
Adapt the traccar.xml to the new DB
Create an empty DB "traccar" in the new MariaDB.
4.1 Create a user with grant permissions for the "traccar" DB
(same username and password as before on Azure)
--> From here, I tested various solutions. None worked - however, the error messages were different.
5.1 Importing the dump using
mysql -h XXX.YYY.ionos.com -u sven -p traccar < traccar_dump.sql
Then I restart the Docker container "traccar."
Traccar doesn't start. The log keeps showing:
ERROR: Main method error - Table 'tc_attributes' already exists - SQLSyntaxErrorException (... < DatabaseModule:99 < <gener:-1 < *:-1 < ... < MainModule:139 < <gener:-1 < ...)
My second option also fails:
5.2 Before importing the dump again, I restart the Docker container. Traccar starts immediately and creates all tables in the new database (MariaDB). I stop the container and try importing the dump again (same command as above in 5.1). Then I start the Docker container and traccar starts. My devices are visible with the last known location. As soon as I address one of them and it connects to Traccar, the following error repeatedly appears in the log (as long as the device is online):
INFO: [T026e539c] connected 2025-07-13 13:32:26
INFO: [T026e539c: h02 < 80.187.116.15] *HQ,9171089463,V1,113453,A,5048.1872,N,01256.8069,E,031.80,265,130725,FFFFBBFF,262,01,35120,551#
2025-07-13 13:32:26
WARN: [T026e539c] error - Cannot invoke "Object.getClass()" because "object" is null - NullPointerException (CacheKey:22 < CacheGraph:95 < CacheManager:291 < *:325 < *:164 < ConnectionManager:157 < ...)
I've run out of ideas for a solution and my rudimentary SQL knowledge. What am I missing? Am I stupid?
The most important thing: At the end of the day, I still have a complete and functioning Traccar installation. I simply retrieve my previous traccar.xml and restart the container – and everything runs smoothly again. In the next few weeks, however, I would like to move more and more of my services away from AZURE – also for cost reasons.
There's no rush, and no data loss has occurred. But I think I can't avoid asking the community and Anton for help...
Thank you & best regards,
Sven
Hi.
I've been struggling with a problem for four days now.
It shouldn't be a problem at all, as I've successfully completed the process several times over the past few years.
My Traccar runs in Docker - the database is on Microsoft AZURE ("Flexible Azure Database for MySQL Server"). Everything has been running smoothly for a long time - not a single Traccar update has ever caused any problems. My task now is to migrate the entire database from the aforementioned AZURE service to a MariaDB, which also runs in Docker.
I perform the following steps:
Complete dump of the existing DB:
mysqldump -h XXXXX.mysql.database.azure.com -u sven -p traccar > traccar_dump.sqlStop the Docker container "traccar"
Adapt the traccar.xml to the new DB
Create an empty DB "traccar" in the new MariaDB.
4.1 Create a user with grant permissions for the "traccar" DB
(same username and password as before on Azure)
--> From here, I tested various solutions. None worked - however, the error messages were different.
5.1 Importing the dump using
mysql -h XXX.YYY.ionos.com -u sven -p traccar < traccar_dump.sqlThen I restart the Docker container "traccar."
Traccar doesn't start. The log keeps showing:
My second option also fails:
5.2 Before importing the dump again, I restart the Docker container. Traccar starts immediately and creates all tables in the new database (MariaDB). I stop the container and try importing the dump again (same command as above in 5.1). Then I start the Docker container and traccar starts. My devices are visible with the last known location. As soon as I address one of them and it connects to Traccar, the following error repeatedly appears in the log (as long as the device is online):
I've run out of ideas for a solution and my rudimentary SQL knowledge. What am I missing? Am I stupid?
The most important thing: At the end of the day, I still have a complete and functioning Traccar installation. I simply retrieve my previous traccar.xml and restart the container – and everything runs smoothly again. In the next few weeks, however, I would like to move more and more of my services away from AZURE – also for cost reasons.
There's no rush, and no data loss has occurred. But I think I can't avoid asking the community and Anton for help...
Thank you & best regards,
Sven