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
The issue was observed in stress/test_stress_routes.py in T2 testing. This is details of the issue that we debugged:
Test withdrew ebgp routes. One of routes in question is 192.169.67.128/25. We confirmed the route was removed from HW.
Test performed back-to-back route announce & withdraw: announce routes, wait for 5 seconds, and then withdraw routes again.
After the back-to-back route announce & withdraw, we observed 192.169.67.128/25 was installed in HW.
We checked in VM and confirmed VM did advertise 192.169.67.128/25 even the last action from test was route withdraw.
ARISTA18T1#show ip bgp neighbors 10.0.0.92 advertised-routes
BGP routing table information for VRF default
Router identifier 100.1.0.47, local AS number 65008
Route status codes: s - suppressed, * - valid, > - active, # - not installed, E - ECMP head, e - ECMP
S - Stale, c - Contributing to ECMP, b - backup, L - labeled-unicast
Origin codes: i - IGP, e - EGP, ? - incomplete
AS Path Attributes: Or-ID - Originator ID, C-LST - Cluster List, LL Nexthop - Link Local Nexthop
Network Next Hop Metric LocPref Weight Path
* > 100.1.0.47/32 10.0.0.93 - - - 65008 i
* > 192.169.67.128/25 10.0.0.93 - - - 65008 65504 i
ARISTA18T1#
We tried to increase the delay between back-to-back route announce & withdraw, and didn't see the issue. So it seems, if the delay is not long enough between back-to-back route announce & withdraw, VM may miss the route withdraw request/event.
Results you see
Test failed in T2 testing with the following error
E Failed: ipv4/ipv6 route used after is not equal to it used before
Results you expected to see
Test should pass.
Is it platform specific
generic
Relevant log output
Output of show version
Attach files (if any)
No response
The text was updated successfully, but these errors were encountered:
ysmanman
changed the title
[Bug]: [T2] VM failed to writhdraw routes
[Bug]: [T2] VM failed to writhdraw routes after back-to-back route announce & withdraw
Jan 16, 2025
ysmanman
changed the title
[Bug]: [T2] VM failed to writhdraw routes after back-to-back route announce & withdraw
[Bug]: [T2] VM didn't withdraw routes after back-to-back route announce & withdraw
Jan 16, 2025
Issue Description
The issue was observed in
stress/test_stress_routes.py
in T2 testing. This is details of the issue that we debugged:192.169.67.128/25
. We confirmed the route was removed from HW.192.169.67.128/25
was installed in HW.192.169.67.128/25
even the last action from test was route withdraw.We tried to increase the delay between back-to-back route announce & withdraw, and didn't see the issue. So it seems, if the delay is not long enough between back-to-back route announce & withdraw, VM may miss the route withdraw request/event.
Results you see
Test failed in T2 testing with the following error
Results you expected to see
Test should pass.
Is it platform specific
generic
Relevant log output
Output of
show version
Attach files (if any)
No response
The text was updated successfully, but these errors were encountered: