PR969 subport implement can not handle subport inside vrf situation #1510
Labels
Comments
|
@wendani - could you please look into the issue, thanks. |
1 task
|
Fixed in #1521. In vs test, we verify that if bound to (non-default) vrf, sub port interface is enslaved to vrf, and parent port not enslaved to vrf in linux kernel |
neethajohn
pushed a commit
that referenced
this issue
May 24, 2021
…-default vrf (#1521) Signed-off-by: Wenda Ni <wonda.ni@gmail.com> What I did Change in IntfMgr: Use alias (instead of subIntfAlias before the change) to hold complete sub interface name; use parentAlias (instead of alias before the change) to hold parent port name. Move sub interface creation to the first place before processing fields vrf, proxy arp, and garp. In doing above, a sub interface can receive the same treatment as other types of interfaces (interface type port, type vlan) on vrf, proxy arp, and garp fields, while maintaining an indicator (parentAlias not empty) to receive treatment specific to a sub interface (creation, deletion). Why I did it Fix #1277, #1510 How I verified it vs test. Extended all test cases to test a sub interface ingress linked to a non-default vrf.
qiluo-msft
pushed a commit
that referenced
this issue
May 25, 2021
…-default vrf (#1521) Signed-off-by: Wenda Ni <wonda.ni@gmail.com> What I did Change in IntfMgr: Use alias (instead of subIntfAlias before the change) to hold complete sub interface name; use parentAlias (instead of alias before the change) to hold parent port name. Move sub interface creation to the first place before processing fields vrf, proxy arp, and garp. In doing above, a sub interface can receive the same treatment as other types of interfaces (interface type port, type vlan) on vrf, proxy arp, and garp fields, while maintaining an indicator (parentAlias not empty) to receive treatment specific to a sub interface (creation, deletion). Why I did it Fix #1277, #1510 How I verified it vs test. Extended all test cases to test a sub interface ingress linked to a non-default vrf.
raphaelt-nvidia
pushed a commit
to raphaelt-nvidia/sonic-swss
that referenced
this issue
Oct 5, 2021
…-default vrf (sonic-net#1521) Signed-off-by: Wenda Ni <wonda.ni@gmail.com> What I did Change in IntfMgr: Use alias (instead of subIntfAlias before the change) to hold complete sub interface name; use parentAlias (instead of alias before the change) to hold parent port name. Move sub interface creation to the first place before processing fields vrf, proxy arp, and garp. In doing above, a sub interface can receive the same treatment as other types of interfaces (interface type port, type vlan) on vrf, proxy arp, and garp fields, while maintaining an indicator (parentAlias not empty) to receive treatment specific to a sub interface (creation, deletion). Why I did it Fix sonic-net#1277, sonic-net#1510 How I verified it vs test. Extended all test cases to test a sub interface ingress linked to a non-default vrf.
EdenGri
pushed a commit
to EdenGri/sonic-swss
that referenced
this issue
Feb 28, 2022
#### What I did Changed code to use use swsscommon instead of swsssdk in counterpoll #### How I did it Removed an explicit function call from sdsssdk module #### How to verify it run counterpoll tests
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
https://github.com/wendani/sonic-swss/blob/bdb92b6cabfe684d9adf0acc89aaa0e66875b9cb/cfgmgr/intfmgr.cpp#L386
https://github.com/wendani/sonic-swss/blob/bdb92b6cabfe684d9adf0acc89aaa0e66875b9cb/cfgmgr/intfmgr.cpp#L437
when I want to put PortChannel1.111 into vrf-100, it actully put PortChannel1 into Vrf-100 in linux(ip link set PortChannel1 master Vrf-100), not PortChannel1.111 into vrf-100
The text was updated successfully, but these errors were encountered: