• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

openconfig / featureprofiles / 29729771018
9%
main: 9%

Build:
Build:
LAST BUILD BRANCH: gres
DEFAULT BRANCH: main
Ran 20 Jul 2026 01:42PM UTC
Jobs 1
Files 131
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

20 Jul 2026 08:59AM UTC coverage: 8.529% (-0.02%) from 8.548%
29729771018

Pull #5509

github

pjacakArista
CNTR: fix gNOI cache staleness after cold reboot and supervisor failover

The CNTR-1.8 cold-reboot test and CNTR-3 supervisor-failover tests fail
on Arista EOS because containerz state verification happens after control
plane disruption. The DUT may reboot or switch supervisors while Ondatra
still holds cached pre-disruption gNOI clients, containerz configuration
may not be saved to startup-config, the standby supervisor may need extra
time to rejoin, and Docker/containerz may restart during recovery.

CNTR-1.8, cold-reboot persistence: save the running configuration before
reboot when the platform requires an explicit config save. After the DUT
comes back, skip the normal containerz config push and build the containerz
client from fresh gNOI clients so the test does not restart the management
stack during warmup or reuse a stale pre-reboot gNOI connection.

CNTR-1.8, reboot detection: add WaitForReboot to poll short-timeout
DialGNMI and system time reads until the DUT is reachable again, then
return fresh BindingDUT().DialGNOI clients. This avoids hanging on stale
half-open connections and handles the case where the reboot completed
before the verification subtest started polling.

CNTR-3, switchover readiness: wait for the active supervisor's
switchover-ready leaf instead of the standby's leaf, and allow an
Arista-specific stabilization delay before declaring readiness failure.
The active supervisor reports whether the standby is ready to take over
more reliably than the standby's own state during standby mode.

CNTR-3, SwitchControlProcessor retry: use fresh gNOI clients when required
and retry SwitchControlProcessor on gRPC Unavailable while the DUT remains
reachable. Stop retrying once the device goes unreachable, because that
means the switchover is already executing and the recovery poller can take
over.

CNTR-3, post-switchover recovery: make waitForSwitchover poll short-timeout
DialGNMI before telemetry reads, return fresh gNOI clients after ... (continued)
Pull Request #5509: CNTR-1-3: fix gNOI cache staleness after cold reboot and supervisor failover

0 of 61 new or added lines in 4 files covered. (0.0%)

2305 of 27026 relevant lines covered (8.53%)

0.1 hits per line

Uncovered Changes

Lines Coverage ∆ File
23
31.66
-1.3% topologies/binding/binding.go
20
0.0
0.0% proto/metadata_go_proto/metadata.pb.go
12
0.0
0.0% internal/deviations/deviations.go
6
4.76
-0.2% internal/components/components.go
Jobs
ID Job ID Ran Files Coverage
1 29729771018.1 20 Jul 2026 01:42PM UTC 131
8.53
GitHub Action Run
Source Files on build 29729771018
  • Tree
  • List 131
  • Changed 4
  • Source Changed 0
  • Coverage Changed 4
Coverage ∆ File Lines Relevant Covered Missed Hits/Line
  • Back to Repo
  • Pull Request #5509
  • PR Base - main (#29709567751)
  • Delete
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc