moon
moon copied to clipboard
aerokube
Add request queues
Hello,
do I understand this correctly: there is no way, at the moment, to queue extraneous requests for moon?
So if I launch 5 tests in parallel on a grid, with only 4 available nodes, the 5th will be dropped and never played?
No, it will be queued and wait for available resources. This ticket is for changing internal implementation of queues...
чт, 4 апр. 2019 г., 10:24 Benjamin Audren [email protected]:
Hello,
do I understand this correctly: there is no way, at the moment, to queue extraneous requests for moon?
So if I launch 5 tests in parallel on a grid, with only 4 available nodes, the 5th will be dropped and never played?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/aerokube/moon/issues/88#issuecomment-479782048, or mute the thread https://github.com/notifications/unsubscribe-auth/ABWNbTxLsBl97DFHjvS-a8x4rUo8VAlLks5vdaiwgaJpZM4YYkM5 .
Ok - I must be doing something bad with my deployment then. I use the free 4 pods license for moon on my openshift cluster, and if I go over the 4 pods limit, my jobs are dropped instead of queued. Is there a setting somewhere to turn the queuing on? I could find nothing in the documentation.
Moon does not drop requests. There is route in OpenShift, you have to tune its timeout and set for ex 30 min...
чт, 4 апр. 2019 г., 10:35 Benjamin Audren [email protected]:
Ok - I must be doing something bad with my deployment then. I use the free 4 pods license for moon on my openshift cluster, and if I go over the 4 pods limit, my jobs are dropped instead of queued. Is there a setting somewhere to turn the queuing on? I could find nothing in the documentation.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/aerokube/moon/issues/88#issuecomment-479785437, or mute the thread https://github.com/notifications/unsubscribe-auth/ABWNbRRzZXE1jED-40RBopbMnL1ob025ks5vdatbgaJpZM4YYkM5 .
You are right, sorry about this! Setting a high timeout solves my issue. Thanks for the support!