Some coverage pages are missing

[tniessen]

Example:

• JavaScript: https://coverage.nodejs.org/coverage-ea9943158654bca9/index.html
• C++: https://coverage.nodejs.org/coverage-ea9943158654bca9/cxxcoverage.html

[mhdawson]

We do delete coverage data because we only have so much storage. I suspect this it what's happened in this case unless these are more recent.

[richardlau]

According to Jenkins the job has been broken for over a month failing due to the rsync process timing out: e.g. https://ci.nodejs.org/view/Node.js%20Daily/job/node-test-commit-linux-coverage-daily/1305/nodes=benchmark/console

04:16:23 + rsync -r out coveragedata:coverage-out
04:16:23 Build timed out (after 30 minutes). Marking the build as failed.
04:46:23 rsync: connection unexpectedly closed (96469308 bytes received so far) [sender]
04:46:23 rsync error: unexplained error (code 143) at io.c(226) [sender=3.1.1]

I'm guessing even the recent pages that appear to be there may be incomplete uploads.

This usually means it is time to prune older results, which is a manual process involving three different machines 😞. As I've stated in #nodejs-build in Slack, fixing coverage.nodejs.org is not high priority and the long term goal is to shift fully to codecov.io (https://github.com/nodejs/build/issues/2797).

[github-actions[bot]]

This issue is stale because it has been open many days with no activity. It will be closed soon unless the stale label is removed or a comment is made.

[tniessen]

@richardlau Should we close this now that codecov appears to be working for the most part?

[richardlau]

@tniessen Yes, let's close. I plan to land/deploy https://github.com/nodejs/build/pull/3145 after I'm back at work in two week's time and that is going to end up also removing the bits that update our own hosted coverage.

[richardlau]

FWIW I've created a CloudFlare Page Rule to permanently redirect coverage.nodejs.org to https://codecov.io/gh/nodejs/node: https://github.com/nodejs/build/pull/3145#discussion_r1197985902