chore(deps): update vmactions/netbsd-vm action to v1.5.1 - #97
Merged
Merged
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR contains the following updates:
v1.5.0→v1.5.1Release Notes
vmactions/netbsd-vm (vmactions/netbsd-vm)
v1.5.1Compare Source
NetBSD on ubuntu-26.04 runners, and packages on 9.5 again
The action's test matrix now runs on ubuntu-26.04 and ubuntu-26.04-arm:
220 of 221 jobs pass, and the 9p job is skipped as before.
What failed on 26.04
With v1.5.0 on a 26.04 runner:
aarch64 jobs for 10.0 and 11.0, on both runner types, timed out in
the firmware. The serial log holds only the banner of 26.04's edk2
2025.11, which hangs under -cpu max:
tianocore/edk2#11962
9.0 - 9.3 aarch64 got past the firmware and then hung, booting
through ACPI on QEMU 10.2.1, right after the interrupt controller
attached. The last console line was
gicvthree0: ITS [#1] Collections table @ 0x49290000/0x10000, ...
Every riscv64 job stopped at
Error: QEMU binary 'qemu-system-riscv64' not found (searched PATH
and common install locations).
Since Ubuntu 25.10 the riscv64 emulator ships in a separate
qemu-system-riscv package, no longer in qemu-system-misc.
9.4, 9.5 and 10.1 aarch64 were not affected, nor were x86_64, sparc64
and microvm.
Fixes
and, for 2025.08 and later, boots with Ubuntu 24.04's edk2 2024.02
build instead, downloaded once and checked against a pinned sha256.
one applies on every host, and 9.x boots that way on QEMU 8.2.2 as
well.
for riscv64 on Ubuntu releases newer than 24.04.
The firmware and package fixes engage only on a newer firmware or Ubuntu
release than 24.04 ships; acpi=off is the one change that also reaches
ubuntu-24.04. Details:
https://github.com/anyvm-org/anyvm/releases/tag/v0.7.2
Fixed: packages on 9.5
Between 2026-09-17 and 2026-09-25 cdn.NetBSD.org and ftp.NetBSD.org
dropped every 9.x package tree. It moved to archive.NetBSD.org, whose
package downloads answer HTTP 402 with a bot-check form. No mirror lists
a 9.5 alias or any 9.5 quarterly, so on 9.5 the action had always kept
the image's own PKG_PATH, and its only live entry was ftp.NetBSD.org's
9.0_2026Q1 tree. With that gone, v1.5.0 fails every 9.5 job that
installs a package, on x86_64 and aarch64 alike:
pkg_add: no pkg found for 'curl', sorry.
For a mirror that lists neither a quarterly nor an alias for the running
release, the action now takes the newest quarterly of the release's own
branch, .0_YYYYQn. Quarterlies are built only for a branch's .0
release, and the minor releases' aliases point into those trees. On 9.5
that is the 9.0_2026Q2 tree on ftp.fr.NetBSD.org and ftp.jaist.ac.jp,
plus ftp.allbsd.org's newest 9.0 quarterly when it answers (2026Q2 on
aarch64, 2025Q4 on x86_64). Of the releases this action tests, only 9.5
reaches this fallback; a mirror whose listed alias does not answer is
still dropped, as before.
Fixed: a lost network probe no longer fails as "Unsupported OS"
vmactions/freebsd-vm#163: a job failed after three minutes with
Error: Unsupported OS: freebsd. Builder repository
anyvm-org/freebsd-builder not found.
although the image exists. A single failed probe of the release download
URL sent the run down the GitHub REST API. The action never passed a
token, so that request was unauthenticated and shared the 60/hour per-IP
limit with every other job on the runner IP; the 403 was retried for
three minutes and then reported as an unsupported OS. This action took
the same path. Two changes close it:
VM runtime, so the release lookup is authenticated (1000 requests
per hour per repository instead of 60 shared). Nothing to change in
your workflow; set it only if you want a different token.
only a 404 means the image is absent; a rate-limited 403 or a 429
stops the retry loop at once and reports the reset time; "the API
could not be asked" and "this repository does not exist" are
distinct outcomes, so a connectivity failure is reported as one; and
the Authorization header no longer follows a cross-host redirect.
Pins
BUILDER_VERSION 2.2.6 (unchanged)
ANYVM_VERSION 0.6.9 -> 0.7.2
Configuration
📅 Schedule: (UTC)
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about this update again.
This PR was generated by Mend Renovate. View the repository job log.