A quick a dirty way would be wrap the Liquidbase startup command Ina script that check every n seconds if postgre is up and only when it detects it up start Liquidbase. But all of this will be solved soon enough with the new feature.
On Friday, July 15, 2016, Derek Mahar <[email protected]> wrote: > Nevermind. I just learned that systemd cannot restart a oneshot service > on failure, so it can't restart Liquibase repeatedly until the PostgreSQL > server is available. I'll have to find some other way to compose these > containers. :( > > On Friday, 15 July 2016 12:17:52 UTC-4, Derek Mahar wrote: >> >> Is there another way to do this without running them in the same >> command? I have the following scenario which I mentioned in an earlier >> message: >> >> Service/Container 1: PostgreSQL Server >> Service/Container 2: Create database schema in PostgreSQL Server using >> Liquibase >> Service/Container 3: Application that connects to PostgreSQL Server >> >> Liquibase, in Service 2, will restart repeatedly until PostgreSQL is >> available to accept SQL commands. Service 3 will start only after Service >> 2 is complete. >> >> How can I connect these application containers using simple host name >> networking? >> >> Derek >> >> On Friday, 15 July 2016 12:06:02 UTC-4, Florian Koch wrote: >>> >>> Hi, >>> >>> Use rkt run ContainerA ContainerB >>> >>> >>> >>> >>> > Am 15.07.2016 um 18:03 schrieb Derek Mahar <[email protected]>: >>> > >>> > How do I link two containers by name using 'rkt run', similar to >>> 'docker run --link name:alias'? >>> > >>> > Thank you, >>> > >>> > Derek >>> >> -- Alessandro Puccetti
