You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When a node is shutdown it is still listed in the hub and has a status of UP. I believe I also tested this in the latest beta version and got the same results.
To Reproduce
Spin up a hub: java -jar selenium-server-4.0.0-alpha-7.jar hub --host localhost --subscribe-events tcp://localhost:4443 --publish-events tcp://localhost:4442 --registration-secret lifebeforedeath
Spin up a node: java -jar selenium-server-4.0.0-alpha-7.jar node --subscribe-events tcp://localhost:4443 --publish-events tcp://localhost:4442 --host localhost --detect-drivers true --registration-secret lifebeforedeath
Query the hub: curl -v -X POST -H "Content-Type: application/json" --data '{"query": "{ grid { nodes { status } } }"}' -s http://localhost:4444/graphql
you will see a status of "UP"
Kill/shutdown the node
Query the hub again
you will see the node listed and it still has a status of "UP"
Detailed steps to reproduce the behavior:
Expected behavior
I expect the node status to be down
Test script or set of commands reproducing this issue
What is the timeout for node health did I not wait long enough? Can time interval for the check for node status / health be configured?
Answered my own question: the hub will set the node to DOWN after 5minutes which seems quite long.
The interface also shows an error at this time: GraphQL error: Can't serialize value (/grid/nodes[0]/status) : Invalid input for Enum 'Status'. Unknown value 'DOWN'
The grapgql response also contains an error
curl -v -X POST -H "Content-Type: application/json" --data '{"query": "{ grid { nodes { status } } }"}' -s http://localhost:4444/graphql
Being able to configure this timeout to a smaller value would be crucial for large scale enterprise deployments and the errors in the interface, graphql are certainly defects.
🐛 Bug Report
When a node is shutdown it is still listed in the hub and has a status of UP. I believe I also tested this in the latest beta version and got the same results.
To Reproduce
Spin up a hub:
java -jar selenium-server-4.0.0-alpha-7.jar hub --host localhost --subscribe-events tcp://localhost:4443 --publish-events tcp://localhost:4442 --registration-secret lifebeforedeath
Spin up a node:
java -jar selenium-server-4.0.0-alpha-7.jar node --subscribe-events tcp://localhost:4443 --publish-events tcp://localhost:4442 --host localhost --detect-drivers true --registration-secret lifebeforedeath
Query the hub:
curl -v -X POST -H "Content-Type: application/json" --data '{"query": "{ grid { nodes { status } } }"}' -s http://localhost:4444/graphql
Kill/shutdown the node
Query the hub again
Detailed steps to reproduce the behavior:
Expected behavior
I expect the node status to be down
Test script or set of commands reproducing this issue
See steps to reproduce
Environment
OS: CentOS 8
Browser: Chrome Driver installed
Browser version: N/A
Browser Driver version: Latest
Language Bindings version: <N/A
Selenium Grid version (if applicable): 4.0.0-alpha-7
The text was updated successfully, but these errors were encountered: