-
Bug
-
Resolution: Unresolved
-
Medium
-
None
-
None
The symptom stays the same since [0]:
Start iperf3 failed!
STDERR:
iperf3: error - unable to send control message: File descriptor in bad state
That is for ldpreload-ipef3. Udpquic error [1] is more verbose, but I see no hints on why connection failed:
"fail_descr": "ECHO_FAIL_QUIC_WRONG_CONNECT (37): Echo connect failed | ECHO_FAIL_TEST_ASSERT_TX_TOTAL (45): expected same (5368709120, got 0) : Invalid amount of data sent | ECHO_FAIL_MISSING_START_EVENT (41): Expected event sconnect to happen, but it did not! | ECHO_FAIL_MISSING_END_EVENT (42): Expected event lastbyte to happen, but it did not!"
I think this is some kind of initialization issue, as e810cq tests are always passing, and the run #94 passed also on cx6dx (also run #86 almost passed, the two failures were CSIT code errors).
Perhaps other jobs leave cx6dx interface in a state hoststack tests cannot recover from? Note that VPP (as opposed to hoststack) tests using cx6dx do not have this issue.
[0] https://s3-logs.fd.io/vex-yul-rot-jenkins-1/csit-vpp-perf-hoststack-daily-master-3n-icx/95/log.html.gz#s1-s1-s1-s1-s1-t1-k2-k3-k11-k8
[1] https://s3-logs.fd.io/vex-yul-rot-jenkins-1/csit-vpp-perf-hoststack-daily-master-3n-icx/104/log.html.gz#s1-s1-s1-s1-s4-t1-k2-k4-k12-k8