mirror of
https://github.com/retspen/webvirtcloud
synced 2026-11-11 12:46:49 +00:00
fix(instances): a migration that does not finish in time is cancelled, not cut off
A VM that changes its memory faster than the network carries it never finishes a live migration without auto converge: measured with a guest rewriting about 1.2 GB/s over an 8.5 Gbit/s link. The migration then ran until gunicorn ended the request after 600 s. libvirt cancelled it when the request went away and the VM stayed on the source, but the user got an error page and nothing was logged. The app now cancels a migration that has not finished after 540 s (MIGRATION_TIME_LIMIT, below the 600 s of gunicorn.conf.py; a test keeps it below). The failure is reported and logged like any other: the migration was cancelled, the VM stays on the source host, and a VM like that needs auto converge. A timer that fires as the migration ends finishes before the compute locks are released, and an error the cancel did not cause keeps its own message.
This commit is contained in:
parent
1016f745c2
commit
043c24a208
4 changed files with 120 additions and 5 deletions
|
|
@ -2,17 +2,20 @@
|
|||
destination with the app's connection, re-reads the VM under the compute
|
||||
locks, and records a finished migration even when autostart fails."""
|
||||
|
||||
import re
|
||||
from pathlib import Path
|
||||
from unittest.mock import MagicMock, patch
|
||||
|
||||
from accounts.models import UserInstance
|
||||
from computes.models import Compute
|
||||
from django.conf import settings
|
||||
from django.contrib.auth import get_user_model
|
||||
from django.db import DatabaseError
|
||||
from django.test import TestCase
|
||||
from logs.models import Logs
|
||||
|
||||
from instances.models import Instance
|
||||
from instances.utils import migrate_instance
|
||||
from instances.utils import MIGRATION_TIME_LIMIT, migrate_instance
|
||||
from vrtManager import util
|
||||
|
||||
|
||||
|
|
@ -85,3 +88,10 @@ class MigrationBookkeepingTests(TestCase):
|
|||
found.refresh_from_db()
|
||||
self.assertTrue(found.is_template)
|
||||
self.assertEqual(UserInstance.objects.get(user=owner).instance_id, found.id)
|
||||
|
||||
def test_a_migration_is_cancelled_before_the_web_server_ends_the_request(self):
|
||||
self.migrate()
|
||||
self.assertEqual(self.moveto.call_args.kwargs["timeout"], MIGRATION_TIME_LIMIT)
|
||||
gunicorn = (Path(settings.BASE_DIR) / "gunicorn.conf.py").read_text()
|
||||
request_timeout = int(re.search(r"^timeout = (\d+)", gunicorn, re.M).group(1))
|
||||
self.assertLess(MIGRATION_TIME_LIMIT, request_timeout - 30) # time to answer and log
|
||||
|
|
|
|||
|
|
@ -150,6 +150,11 @@ def get_network_tuple(network_source_str):
|
|||
return network_source_pack[0], "net"
|
||||
|
||||
|
||||
# A migration that has not finished by then is cancelled: below the 600 s
|
||||
# after which gunicorn (gunicorn.conf.py) ends the request.
|
||||
MIGRATION_TIME_LIMIT = 540
|
||||
|
||||
|
||||
def migrate_instance(
|
||||
new_compute,
|
||||
instance,
|
||||
|
|
@ -194,6 +199,7 @@ def migrate_instance(
|
|||
autoconverge,
|
||||
compress,
|
||||
uri=new_compute.migration_uri,
|
||||
timeout=MIGRATION_TIME_LIMIT,
|
||||
)
|
||||
except Exception as error:
|
||||
msg = _("Instance migration to %(target)s failed: %(error)s") % {
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue