Repository navigation
qemu: restart the guest if its kernel panics during boot - #150
Merged
Merged
Conversation
Contributor
Author
|
Measured on GitHub-hosted runners (Azure, KVM nested), through the real ITF code (QemuProcess, QemuTarget, pre-test checks, this retry), unmodified QNX IFS, 4 parallel guests: 24,000 boots, 11 guest kernel panics (about 1 in 2,200), all 11 recovered by one restart, 0 failed boots. Without the retry, each of those 11 would have been a failed test. Run: https://github.com/castler/communication/actions/runs/41080480815 |
This was referenced Oct 30, 2026
LittleHuba
previously approved these changes
Oct 30, 2026
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Oct 30, 2026
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Oct 31, 2026
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Oct 31, 2026
castler
enabled auto-merge
October 31, 2026 20:12
castler
force-pushed
the
qemu-guest-panic-retry
branch
from
November 2, 2026 08:43
0cf282d to
adfbc31
Compare
The QNX kernel can panic during SMP bring-up (Shutdown[...] on the serial console) and never recovers. The pre-test checks then hang on an SSH banner that never arrives, and the ping check passes regardless. Detect the panic marker on the QEMU console, terminate QEMU so pending SSH connections fail fast, and boot again up to --qemu-boot-attempts (default 3) times. Other failures are not retried. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…hing the module Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
LittleHuba
approved these changes
Nov 2, 2026
github-project-automation
Bot
moved this from Backlog
to Done
in TST - Testing Community
etas-contrib-automation Bot
pushed a commit
to etas-contrib/score_communication
that referenced
this pull request
Nov 3, 2026
Bumps `score_itf` 0.6.0 -> 0.7.0. - eclipse-score/itf#150: restart the QNX guest when its kernel panics during boot (eclipse-score#1513, eclipse-score/qnx_sdp#9) - eclipse-score/itf#151: do not lose early docker exec output - eclipse-score/itf#144: SSH auth support
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.
The QNX 8 guest kernel intermittently panics during SMP bring-up (about 1 in 2000-4500 boots on nested-virtualization CI runners): the serial console prints
Shutdown[1] S/C/F=5/4/3 ... muonand the guest halts. Details: eclipse-score/communication#1513.Today the pre-test checks then hang on an SSH banner that never arrives (
banner_timeout=200), and the ping check passes regardless because it pings host loopback. The result is a flaky test failure.This PR:
Console/PipeConsoletake expression callbacks before the reader thread starts (no missed lines),Shutdown[) on the QEMU console and terminates QEMU, so pending SSH connections fail immediately,--qemu-boot-attemptstimes (default 3) if the pre-test checks fail after a panic. Other failures are not retried.Tests: new unit tests (
test_qemu_boot_retry). Also verified against a real QNX guest by temporarily using a marker that every boot prints: detect, terminate, restart x3, raise; and a normal QNX integration test in communication passes with this change.The kernel bug itself is not fixed here; this only makes the test infrastructure recover from it.