Open MaxYaremchuk opened 4 years ago
The bug has been fixed by the community under the latest branch 201911, hash dea38d1
Hi @Hedgehog-Guru, are we sure this issue was fixed in the above mentioned commit? It can still be reproduced in release 202211.
Sep 1 13:52:00.491344 INFO teamd#supervisord: teammgrd libteamdctl: cli_usock_process_msg: usock: Error message received: "NoSuchDev"
Sep 1 13:52:00.491822 INFO teamd#supervisord: teammgrd libteamdctl: cli_usock_process_msg: usock: Error message content: "No such device."
Sep 1 13:52:00.492043 INFO teamd#supervisord: teammgrd command call failed (Invalid argument)
Sep 1 13:52:00.537436 INFO teamd#supervisord: teammgrd Cannot find device "Ethernet48"
Sep 1 13:52:00.583468 ERR teamd#teammgrd: :- main: Runtime error: /usr/bin/teamdctl PortChannel2 port remove Ethernet48; /sbin/ip link set dev "Ethernet48" "down"; /sbin/ip link set dev "Ethernet48" mtu "9100" :
Sep 1 13:52:00.683021 INFO teamd#supervisord 2023-09-01 13:52:00,682 INFO exited: teammgrd (exit status 1; not expected)
Sep 1 13:52:01.693996 INFO teamd#supervisor-proc-exit-listener: Process 'teammgrd' exited unexpectedly. Terminating supervisor 'teamd'
Description The
show interfaces status
command returns traceback in case of adding a nonexistent port to a PortChannelMoreover, In case of removing nonexistent port from portchannel teamd container gets rebooted.
Steps to reproduce the issue
# config portchannel add PortChannel0001
# config portchannel member add PortChannel0001 Ethernet128
Pay attention, interface Ethernet128 is nonexistent in the system.
show interfaces status
Describe the results you received The system returns traceback:
No one docker has crashed:
# config portchannel member del PortChannel0001 Ethernet128
teamd container got rebooted with errors in syslog:Describe the results you expected The
config portchannel member add
command should prevent from assigning to a portchannel nonexistent interface.Output of
show version
Debug file sonic_dump_r-ethernet-sw230_20200117_194152.tar.gz