[libvirt] [PATCH 3/3] qemu: Ignore temporary job errors when checking migration status

[Jiri Denemark]

When qemu driver is polling for migration to finish (in
qemuMigrationWaitForCompletion), it may happen that another job allowed
during migration is running and if it does not finish within 30 seconds,
migration would be cancelled because of that. However, we can just
ignore the timeout and let the waiting loop try again later.

If an event fired at the end of migration is ever implemented in QEMU,
we can just wait for the event instead of polling for migration status
and libvirt will behave consistently, i.e., migration won't be cancelled
in case another job started during migration takes long time to finish.

For bug https://bugzilla.redhat.com/show_bug.cgi?id=1083238

Signed-off-by: Jiri Denemark <jdenemar at redhat.com>
---
 src/qemu/qemu_domain.c    | 10 ++++++----
 src/qemu/qemu_migration.c |  7 ++++---
 2 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
index 3df454f..2d8afd6 100644
--- a/src/qemu/qemu_domain.c
+++ b/src/qemu/qemu_domain.c
@@ -1287,8 +1287,9 @@ qemuDomainObjEnterMonitorInternal(virQEMUDriverPtr driver,
     qemuDomainObjPrivatePtr priv = obj->privateData;
 
     if (asyncJob != QEMU_ASYNC_JOB_NONE) {
-        if (qemuDomainObjBeginNestedJob(driver, obj, asyncJob) < 0)
-            return -1;
+        int ret;
+        if ((ret = qemuDomainObjBeginNestedJob(driver, obj, asyncJob)) < 0)
+            return ret;
         if (!virDomainObjIsActive(obj)) {
             virReportError(VIR_ERR_OPERATION_FAILED, "%s",
                            _("domain is no longer running"));
@@ -1367,8 +1368,9 @@ void qemuDomainObjExitMonitor(virQEMUDriverPtr driver,
  * with the same asyncJob.
  *
  * Returns 0 if job was started, in which case this must be followed with
- * qemuDomainObjExitMonitor(); or -1 if the job could not be
- * started (probably because the vm exited in the meantime).
+ * qemuDomainObjExitMonitor(); -2 if waiting for the nested job times out;
+ * or -1 if the job could not be started (probably because the vm exited
+ * in the meantime).
  */
 int
 qemuDomainObjEnterMonitorAsync(virQEMUDriverPtr driver,
diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c
index a9f7fea..f0df1a6 100644
--- a/src/qemu/qemu_migration.c
+++ b/src/qemu/qemu_migration.c
@@ -1714,8 +1714,9 @@ qemuMigrationUpdateJobStatus(virQEMUDriverPtr driver,
 
     ret = qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob);
     if (ret < 0) {
-        /* Guest already exited; nothing further to update.  */
-        return -1;
+        /* Guest already exited or waiting for the job timed out; nothing
+         * further to update. */
+        return ret;
     }
     ret = qemuMonitorGetMigrationStatus(priv->mon, &status);
 
@@ -1812,7 +1813,7 @@ qemuMigrationWaitForCompletion(virQEMUDriverPtr driver, virDomainObjPtr vm,
         /* Poll every 50ms for progress & to allow cancellation */
         struct timespec ts = { .tv_sec = 0, .tv_nsec = 50 * 1000 * 1000ull };
 
-        if (qemuMigrationUpdateJobStatus(driver, vm, job, asyncJob) < 0)
+        if (qemuMigrationUpdateJobStatus(driver, vm, job, asyncJob) == -1)
             goto cleanup;
 
         /* cancel migration if disk I/O error is emitted while migrating */
-- 
1.9.3