I am currently overriding the 60-database.sh startup file completely because in my case I may be restoring a database rather than creating a new one from scratch. Building from a specific SQL file might be a useful thing to be able to do in general though (e.g. for diagnosing a server where you have been given a dump) and I could also make use of that.
omego has an option --omerosql which does exactly that, but the question would be: How to expose that at container instantiation time. A simple way would be to designate a place where a SQL file could be placed using a volume mount (or other technique, in my case an earlier startup script might have fetched it to the prescribed location from S3), and if present will be used instead of creating a fresh database if that file exists.
Any thoughts?
I am currently overriding the
60-database.shstartup file completely because in my case I may be restoring a database rather than creating a new one from scratch. Building from a specific SQL file might be a useful thing to be able to do in general though (e.g. for diagnosing a server where you have been given a dump) and I could also make use of that.omegohas an option--omerosqlwhich does exactly that, but the question would be: How to expose that at container instantiation time. A simple way would be to designate a place where a SQL file could be placed using a volume mount (or other technique, in my case an earlier startup script might have fetched it to the prescribed location from S3), and if present will be used instead of creating a fresh database if that file exists.Any thoughts?