You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Reboot with mgmt vrf enabled is causing force_mgmt routes to have priority over local table. I've verified on 202405 and above but likely a day 1 behavior.
Steps to reproduce the issue:
Initial Config
root@sonic:/home/admin# redis-cli -n 4 hgetall "MGMT_INTERFACE|eth0|10.245.21.113/22"
1) "forced_mgmt_routes@"
2) "10.245.0.0/16,......"
3) "gwaddr"
4) "10.245.20.1"
root@sonic:/home/admin# ip ru sh
1001: from all lookup local <Notice local rule has preference over the forced mgmt rule>
32764: from all to 10.245.0.0/16 lookup default
32765: from 10.245.21.113 lookup default
32766: from all lookup main
32767: from all lookup default
Configure mgmt vrf
root@sonic:/home/admin# config vrf add mgmt
root@sonic:/home/admin# ip ru sh
1000: from all lookup [l3mdev-table] <High priority rule to handle Vrf>
1001: from all lookup local <Here, local has priority over forced_mgmt route>
32764: from all to 10.245.0.0/16 lookup mgmt
32765: from 10.245.21.113 lookup mgmt
32766: from all lookup main
32767: from all lookup default
Save config and reboot
root@sonic:/home/admin# config save -y
root@sonic:/home/admin# reboot
<After reboot>
root@sonic:/home/admin# ip ru sh
1000: from all lookup [l3mdev-table]
32764: from all to 10.245.0.0/16 lookup mgmt <Forced mgmt rule took priority over local>
32765: from all lookup local
32765: from 10.245.21.113 lookup mgmt
32766: from all lookup main
32767: from all lookup default
Remove mgmt vrf and traffic to eth0 starts dropping
root@sonic:/home/admin# config vrf del mgmt
root@sonic:/home/admin# ip ru sh
1000: from all lookup [l3mdev-table]
32764: from all to 10.245.0.0/16 lookup default
32765: from all lookup local
32765: from 10.245.21.113 lookup default
32766: from all lookup main
32767: from all lookup default
root@sonic:/home/admin# ping 10.245.21.113
PING 10.245.21.113 (10.245.21.113) 56(84) bytes of data. <ping from inside the device is not working>
ping 10.245.21.113
PING 10.245.21.113 (10.245.21.113) 56(84) bytes of data.
icmp_seq=18 Destination Host Unreachable
Or can be reproduced by running syslog/test_syslog_source_ip.py::test_syslog_config_work_after_reboot setups with mgmt ip being matched with any of the forced_mgmt_route prefixes.
Describe the results you expected:
Traffic should work even after config
The text was updated successfully, but these errors were encountered:
Description
Reboot with mgmt vrf enabled is causing force_mgmt routes to have priority over local table. I've verified on 202405 and above but likely a day 1 behavior.
Steps to reproduce the issue:
Or can be reproduced by running
syslog/test_syslog_source_ip.py::test_syslog_config_work_after_reboot
setups with mgmt ip being matched with any of the forced_mgmt_route prefixes.Describe the results you expected:
Traffic should work even after config
The text was updated successfully, but these errors were encountered: