On Wed, Oct 14, 2026 at 10:18:45AM +0200, Alexander Mikhalitsyn wrote: > > - for this test, should we remove snapshot=on but instead create an > > explicit overlay that will be used on both src/dst qemu? > > This should solve the issue too, at the same time if test is R/O I'm > not sure if we really need this.
Then.. maybe no need to bother. :) I think we should better remove snapshot=on no matter what, then whoever sends a patch add migration blocker for it won't fail this test. Seems the /tmp mount is still needed then. For speed control, IMHO it's not good setting bw either too high or too low (similar to downtime); too high means converging too fast, too low means it won't converge. Maybe we can mimic what migration-test.c normally do in qtest? that is, we do: - migrate_ensure_non_converge(), - allow a few rounds of migration precopy to happen (e.g. query-migrate shows dirty-count >=2), then switch to, - migrate_ensure_converge(). Then make it generic in migrate_vms(). I think it can be helpful too for future migration functional tests when it's not always idle? Thanks, -- Peter Xu
