-
Notifications
You must be signed in to change notification settings - Fork 35
sync: SVS v4 (mhash, PARTIAL, publish+pull) #190
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: psvs
Are you sure you want to change the base?
Changes from 6 commits
b88a364
fe5586b
e9522cc
1002285
5d8b1be
9340c00
56c2af0
67d93c2
aed3420
7f04d17
d27e022
1872c64
5f3a477
ea79483
e6ce78e
bbce410
a9a0de7
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Large diffs are not rendered by default.
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
|
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Was 3 retries not enough in testing?
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Good catch — and 3 isn't quite enough on the 53-node sprint topology. After this thread, the e2e test in CI started failing intermittently with 'fetch metadata failed with result: Timeout' (4.5s after the cat fired). Locally it passes reliably because DV's startup converges quickly, but CI's slower startup stretches the window in which a DV startup-Reset can wipe a producer prefix from a consumer's RIB just after the convergence check returns. So in a follow-up commit (9340c00), I bumped fetchMetadata and fetchDataByPrefix to Retries: 5 and Lifetime: 2s (~10s budget). That gives the consumer enough slack for DV to finish its post-startup re-announce cycle. Verified locally: all three scenarios (NDNd, NFD, NDNd-replay) pass. Verified in CI: 9340c00 is green. The seg fetcher (client_consume_seg.go) is still at maxRetries: 3 because the segment-fetch flake that originally motivated the bump is solved by the AIMD congestion window in 1002285, not by retry budget.
Collaborator
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. If it's a DV convergence issue, the CI should be updated have a long enough wait time for convergence, instead of modifying the underlying library. |
Uh oh!
There was an error while loading. Please reload this page.