test: the vip-upgrade epoch read retries when its etcd read times out (keel#133) #134
No reviewers
Labels
No labels
accessibility
bug
documentation
duplicate
enhancement
good first issue
help wanted
invalid
question
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
Keel-Linux/keel!134
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "test/flaky-vip-upgrade-epoch-133"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
In test_b_etcd_restarted_on_each_member_in_turn,
epoch()read the VIP claim with one agent_send. The "epoch" handler does one linearizable etcd read over the lossy link; when it outlasts its 2 s timeout the answer holds no "epoch" key and the test recorded epoch_before = None (assertIsNotNone then failed, with the VIP never moving — holders_seen "A" throughout).The ask now repeats while the answer holds no epoch, bounded (60 s), the same treatment upgrade-check already has in this file (healthy_when_asked, keel#94). Evidence in the issue: run 37966794940 job 113943040489 on 2026-10-09.
The vip-upgrade netns job is the coverage; a green vip-upgrade / trixie run here is the test plan.