Restart a service

Pre requisites

flex installation scripts

flex inventory

access to all flex hosts through ssh using an ssh key (you can use passwords too, please read ansible documentation for this)

Rolling restart methodology

Flex services are redundant, which means you can restart a service instance having the other instance(s) still providing the related service. We call this operation a rolling restart

Warning

Prior to restart another flex service instance you need to ensure the one you processed is healthy. To check this you can to a docker ps on the host and ensure the restarted service is set to healthy. If it is not you will need either to inspect the related logs or ask for support

Restarting

Identify the hosts running the element you want to restart. Please refer to inventory to find them

Restart commands (on each node). Please remind you must not restart a service instance until the current one is started and healthy

cd /flex  # (or the main installation folder you have defined to deploy flex)
docker compose restart <ServiceName>

Important

Most of flex services can be rolled restart with no user impact / visibility. The master service will have a visible impact for some users showing an http/503 error page. they will only need to reload the page to get back theirt access