VM : Why performing maintenance operations
This application is built to be very resilient and do not require a lot of maintenance operations. However, in some situation you might want to restart the docker to update your application version and benefit from the last patch releases.
In some cases, for instance if any error happened during the data processing, some of the files that are created for the operation will not be destroyed. They could end up in a situation where the files will accumulate and flood the volume allocated. This case would be easily fixed by stopping the app and restarting it.
When operate
Data processes failed
The first signal your application needs to be fixed is when some of your data processes launched from Salesforce end up with a status “Failed”. Before rushing to your WM be sure to check if the error description does not provide any information who could indicate the issue is somewhere else. For instance, if you built your own deduplication rules, some errors may occur and stop the process.
So, for instance if you check you Salesforce Lightning Application “DQE One”, in the Runs tab and you find any run failed such as follow, you might want to perform a restart of your VM.
Data processes pending
Sometime the job queue will get stuck on the same message and blocking the data processes. In this situation you should clear the fileshare allocated to the rabbitmq container and reload the application. This will fix the issue in almost every cases.
If your fileshare linked to the rabbitmq container is defined as public, you can simply access it through the azure portal interface and delete what he contains.
If your fileshare is defined as private, you will need to access the container to manually clear the rabbitmq directory. When the container, if running, proceeds as follows:
- Connect to your VM as an admin
sudo bash- Go to the repository where the docker-compose.yml file for DQE is located and down the docker
docker-compose down- Find the name of the volume of message queue
docker volume ls- The name of the volume must contain "rabbitmq" and "data" (for example : dqe_rabbitmq_data) and delete it
docker volume rm <name of the volume>- Up the docker-compose
docker-compose up
Related to