Repository navigation
feat(schedules): stagger nongnu and gnu-alpha off gnu's slots - #12
Merged
Merged
Conversation
gnu, nongnu and gnu-alpha all read from FSF hosts. Each one gets its own slots, four hours apart: gnu at S3|S15, nongnu at S7|S19, gnu-alpha at S11|S23.
This was referenced Oct 15, 2026
This was referenced Oct 15, 2026
jshvn
added a commit
to katoptra/ctan
that referenced
this pull request
Oct 15, 2026
This repository does not contain the times of its runs. CLAUDE.md and docs/reference.md said the slot times of the scheduler and the cron expression of the healthcheck. When the scheduler moved a job, each of these lines had to change in a second repository. This change removes the times. CLAUDE.md says that only the scheduler has the times of the runs, and that the healthcheck cron has the same times as the slots of the mirror. katoptra/dispatch#12 adds the same rule to the README of the scheduler. katoptra/.github#9 adds it to CONTRIBUTING. Test plan: - [x] `task check`: only documentation changes; render.txt is unchanged - [x] No `HH:42` time or cron expression stays in the repository
jshvn
added a commit
to katoptra/gnu
that referenced
this pull request
Oct 15, 2026
This repository does not contain the times of its runs. CLAUDE.md said the slot times of the scheduler and the cron expression of the healthcheck. When the scheduler moved a job, each of these lines had to change in a second repository. This change removes the times. CLAUDE.md says that only the scheduler has the times of the runs, and that the healthcheck cron has the same times as the slots of the mirror. katoptra/dispatch#12 adds the same rule to the README of the scheduler. katoptra/.github#9 adds it to CONTRIBUTING. Test plan: - [x] `task check`: only documentation changes; render.txt is unchanged - [x] No `HH:42` time or cron expression stays in the repository
jshvn
added a commit
to katoptra/nongnu
that referenced
this pull request
Oct 15, 2026
This repository does not contain the times of its runs. CLAUDE.md said the slot times of the scheduler and the cron expression of the healthcheck. When the scheduler moved a job, each of these lines had to change in a second repository. This change removes the times. CLAUDE.md says that only the scheduler has the times of the runs, and that the healthcheck cron has the same times as the slots of the mirror. katoptra/dispatch#12 adds the same rule to the README of the scheduler. katoptra/.github#9 adds it to CONTRIBUTING. Test plan: - [x] `task check`: only documentation changes; render.txt is unchanged - [x] No `HH:42` time or cron expression stays in the repository
jshvn
added a commit
to katoptra/gnu-alpha
that referenced
this pull request
Oct 15, 2026
This repository does not contain the times of its runs. CLAUDE.md said the slot times of the scheduler and the cron expression of the healthcheck. When the scheduler moved a job, each of these lines had to change in a second repository. This change removes the times. CLAUDE.md says that only the scheduler has the times of the runs, and that the healthcheck cron has the same times as the slots of the mirror. katoptra/dispatch#12 adds the same rule to the README of the scheduler. katoptra/.github#9 adds it to CONTRIBUTING. Test plan: - [x] `task check`: only documentation changes; render.txt is unchanged - [x] No `HH:42` time or cron expression stays in the repository
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
gnu, nongnu and gnu-alpha all start at 03:42 and 15:42 UTC. All three read from FSF hosts in the same network (ftp.gnu.org and dl.sv.gnu.org). This change gives each one its own slots, four hours apart:
Four hours is the largest equal gap for three jobs that run two times a day. A run takes 0 to 2 minutes, so no two of these runs operate at the same time. The number of runs, listings and bytes for each day does not change.
gnu-alpha shares S11 with dropbox and all slots with ctan. Those jobs do not read from FSF hosts.
On the first tick after the deploy, each moved job fires one time if its latest new slot is after the slot that the state records. If not, the tick logs a "clock behind" warning for that job until its first new slot. A warning does not send a ping to
/fail.Change the healthchecks.io crons right after the deploy, not before: nongnu
42 7,19 * * *, gnu-alpha42 11,23 * * *. With the old cron, the missing 15:42 run causes an alert at 18:42.The README also says that this repository is the only location of the slots, and that the cron expression of each healthcheck has the same times as the slots of its job. The mirror repositories do not write their run times (katoptra/.github#9).
Test plan:
task check(gofmt, vet, tests)task targetsgives nongnu at 07:42, 19:42 and gnu-alpha at 11:42, 23:42