fix(test): remove wait_for_up from test_vmgenid

The `wait_for_up()` sends a dummy SSH command with a timeout of 10s, with the intention of blocking the integration test until a VM is responsive after booting. However, post-snapshot restore it runs into issue firecracker-microvm#4099, where on AMD inside the guest, sshd gets stuck inside a nanosleep syscall for longer than intended (and thus longer than the 10s timeout). Alternatively, we could pass `lapic=notscdeadline` in the kernel commandline arguments, but we do not have an easy way to _append_ to the default kernel command line from integration tests, and I'd like to avoid duplicating our default across a bunch of places. Signed-off-by: Patrick Roy <[email protected]>
roypat · Jul 3, 2024 · dc17a23 · dc17a23
1 parent ccd3b2d
commit dc17a23
Showing 1 changed file with 0 additions and 1 deletion.
diff --git a/tests/integration_tests/functional/test_snapshot_basic.py b/tests/integration_tests/functional/test_snapshot_basic.py
@@ -560,7 +560,6 @@ def test_vmgenid(guest_kernel_linux_6_1, rootfs, microvm_factory, snapshot_type)
         vm = microvm_factory.build()
         vm.spawn()
         copied_snapshot = vm.restore_from_snapshot(snapshot, resume=True)
-        vm.wait_for_up()
 
         # We should have as DMESG_VMGENID_RESUME messages as
         # snapshots we have resumed