fangorn/sunshine-qemu
public
ref:f60daa42a1ef380cdae54c3b45472abcd7e1b4fc
fix(linux): say a VM is unreachable once, not several times a second
Capture, audio and input each retry the session while a VM is away, so a VM that stops mid-stream or hasn't
booted yet wrote "no QEMU owns [org.qemu]" at error level a few times a second: 673 lines in the twelve seconds
after killing a VM under test. The first failure is the one worth an error; the retries after it go to the debug
log until the VM answers again, which also resets the report for the next outage.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BPNw4PCgkEfhyCjQT19wsb
SHA:
f60daa42a1ef380cdae54c3b45472abcd7e1b4fc
Author:
Cole Christensen <cole.christensen@gmail.com>
Date:
2026-09-17 05:58
Parents:
4fe9720
3 files changed
+55
-1
| Type | ||
|---|---|---|
|
|
src/platform/linux/qemu/session.cpp | +19 −1 |
|
||
|
|
src/platform/linux/qemu/session.h | +17 −0 |
|
||
|
|
tests/unit/platform/linux/qemu/test_session.cpp | +19 −0 |
|
||