I am using swarm setup for internal load testing purpose where I spin custom containers on need basis and execute tests on them before the results are published and container is killed.
To make sure that a number of parallel containers can run for executing tests, I deployed swarm and have 1 manager with 2 workers now. A shell script sitting on the manager calls for a 'docker run' to spin a container. However, as there might be a case of the manager or one of the workers exhausted in resources, I want the run command for spinning container to be directed to one of the nodes based on their current resource utilisation. The requirement would be:
Direct the request on manager initially to Node 1 as First priority
If Node 1 CPU is > 75% - redirect the request to Node 2
If Node 2 CPU is also > 75% - (which is exceptional) redirect the request to Manager or fail creating a container.
Hi,
I am using swarm setup for internal load testing purpose where I spin custom containers on need basis and execute tests on them before the results are published and container is killed.
To make sure that a number of parallel containers can run for executing tests, I deployed swarm and have 1 manager with 2 workers now. A shell script sitting on the manager calls for a 'docker run' to spin a container. However, as there might be a case of the manager or one of the workers exhausted in resources, I want the run command for spinning container to be directed to one of the nodes based on their current resource utilisation. The requirement would be:
Thanks in advance for the assistance, Ganesh