Process to follow when Metabase fails

MetabaseFix.md

Issue 1: Too big H2 database

This is mostly due to H2 database getting too big. So it doesn't load in 5 seconds. You can debug this using

docker logs b4f559dc829b --tail 100

If it is too big you will need to defrag it.

docker restart <containerID>; docker exec -it <containerID> /bin/bash;
wget http://repo2.maven.org/maven2/com/h2database/h2/1.4.196/h2-1.4.196.jar 
cd metabase.db;java -cp h2*.jar org.h2.tools.Shell -driver org.h2.Driver -url "jdbc:h2:./metabase.db" -user "" -password "";

This will open the SQL where running the following would defrag it.

shutdown defrag;

Now that you have a smaller database copy it to your docker host and keep it safe. Also now you need to spin up a new server and copy paste the older DB file to new one.

docker cp <oldContainerID>:/metabase.db .
docker run -d -p 3000:3000 -e "JAVA_TOOL_OPTIONS=-Xmx6g" metabase/metabase docker cp metabase.db <newContainerID>:/

Issue 2: Disk Filled

This is mostly due to two files.

1. Docker logs filling up the system. To delete them use this

truncate -s 0 /var/lib/docker/containers/*/*-json.log

2. Metabase internal DB filling up the system. To delete them,

docker exec -it <containerID> /bin/bash; cd metabase.db;

References

• https://www.metabase.com/docs/latest/troubleshooting-guide/running.html
• https://stackoverflow.com/questions/42510002/how-to-clear-the-logs-properly-for-a-docker-container
• metabase/metabase#6510
• metabase/metabase#6034