You can run a database in docker with persistent storage, but you won't get any benefit, only disadvantages and unnecessary complications. It is really not suitable for production environments, regardless of the fact you can. It will overcomplicate things pointlessly.
@@kehrin Tablespace files are binary sensitive, you can't patch the binary's of a database in an image, redeploy, then attache the persistence storage and then run as it was all ok. Oracle, for instance, will need the tablespace to be patched as well to not be corrupted - at least for some patches. So you have to patch the running containers and this defeats the purpose of a container. The benefit comes with a lot of smaller databases you do not patch, but a huge amount of them, then containers makes sense. Otherwise, you won't gain anything. Webserver and other servers where you can just reattach the persistence storage is whats it designed for. For message brokers, webserver etc., it makes sense. For 24/7 running databases, nope won't do. For high-performance databases nope, won't do either (VMware is also not suite for high-performance databases, not that VMware and docker is the same at all).
@@milomiller161 DBs need a nothing shared approach. A new container should have its own db files in its own space, and replicate data through the db's replication facilities
Impressed... Will explore
awesome!!!
Nice!!
the sound of the series in a 5.1 home theater is horrible, to fix it change your the set up to a stereo mode
You can run a database in docker with persistent storage, but you won't get any benefit, only disadvantages and unnecessary complications.
It is really not suitable for production environments, regardless of the fact you can. It will overcomplicate things pointlessly.
@@kehrin Tablespace files are binary sensitive, you can't patch the binary's of a database in an image, redeploy, then attache the persistence storage and then run as it was all ok.
Oracle, for instance, will need the tablespace to be patched as well to not be corrupted - at least for some patches.
So you have to patch the running containers and this defeats the purpose of a container.
The benefit comes with a lot of smaller databases you do not patch, but a huge amount of them, then containers makes sense. Otherwise, you won't gain anything. Webserver and other servers where you can just reattach the persistence storage is whats it designed for. For message brokers, webserver etc., it makes sense.
For 24/7 running databases, nope won't do. For high-performance databases nope, won't do either (VMware is also not suite for high-performance databases, not that VMware and docker is the same at all).
@@milomiller161 DBs need a nothing shared approach. A new container should have its own db files in its own space, and replicate data through the db's replication facilities