Failures like https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/3812/pipeline happen from time to time, we should have a retry + backoff (wait 1 second, retry, wait 10 seconds, retry, wait 30 seconds, retry, give up if its still failing) when Jenkins does docker pull:
|
script: "docker pull ${image}", |
cc @Mousius @areusch
Failures like https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/3812/pipeline happen from time to time, we should have a retry + backoff (wait 1 second, retry, wait 10 seconds, retry, wait 30 seconds, retry, give up if its still failing) when Jenkins does
docker pull:tvm/ci/jenkins/Prepare.groovy.j2
Line 79 in d436501
cc @Mousius @areusch