Closed jovial closed 3 years ago
Clarification: It is the systemd in the containers that is too old to handle the capabilities present in the host kernel.
CI failure was: fatal: [testohpc-grp2-0]: FAILED! => {"changed": false, "msg": "Failure downloading http://repos.openhpc.community/OpenHPC/2/CentOS_8/x86_64/ohpc-release-2-1.el8.x86_64.rpm, Request failed: <urlopen error timed out>"}
re-running jobs.
Hmm, hit pip download issues this time:
pip._vendor.urllib3.exceptions.ProtocolError: ("Connection broken: ConnectionResetError(104, 'Connection reset by peer')", ConnectionResetError(104, 'Connection reset by peer'))
and a genuine failure on test12:
TASK [assert] ******************************************************************
fatal: [testohpc-login-0]: FAILED! => {
"assertion": "(jobid + '|0|wrap|compute|2|testohpc-compute-[0-1]|COMPLETED') in sacct.stdout",
"changed": false,
"evaluated_to": false,
"msg": "Didn't find expected output for 2 in sacct output: "
}
Weird test12 has passed now then?
Weird test12 has passed now then?
Yep, I guess we just got unlucky on that run!
For rationale, please see comment in change.