qemu: respond to NETDEV_STREAM_DISCONNECTED event

When a QEMU netdev is of type "stream", if the socket it uses for
connectivity to the host network gets closed, then QEMU will send a
NETDEV_STREAM_DISCONNECTED event. We know that any stream netdev we've
created is backed by a passt process, and if the socket was closed,
that means the passt process has disappeared.

When we receive this event, we can respond by starting a new passt
process with the same options (including socket path) we originally
used. If we have previously created the stream netdev device with a
"reconnect" option, then QEMU will automatically reconnect to this new
passt process. (If we hadn't used "reconnect", then QEMU will never
try to reconnect to the new passt process, so there's no point in
starting it.)

Note that NETDEV_STREAM_DISCONNECTED is an event sent for the netdev
(ie "host side") of the network device, and so it sends the
"netdev-id" to specify which device was disconnected. But libvirt's
virDomainNetDef (the object used to keep track of network devices) is
the internal representation of both the host-side "netdev", and the
guest side device, and virDomainNetDef doesn't directly keep track of
the netdev-id, only of the device's "alias" (which is the "id"
parameter of the *guest* side of the device). Fortunately, by convention
libvirt always names the host-side of devices as "host" + alias, so in
order to search for the affected NetDef, all we need to do is trim the
1st 4 characters from the netdev-id and look for the NetDef having
that resulting trimmed string as its alias. (Contrast this to
NIC_RX_FILTER_CHANGED, which is an event received for the guest side
of the device, and so directly contains the device alias.)

Resolves: https://bugzilla.redhat.com/2172098
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>

src/qemu/qemu_domain.c

@@ -11238,6 +11238,7 @@ qemuProcessEventFree(struct qemuProcessEvent *event)
         break;
     case QEMU_PROCESS_EVENT_WATCHDOG:
     case QEMU_PROCESS_EVENT_DEVICE_DELETED:
+    case QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED:
     case QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED:
     case QEMU_PROCESS_EVENT_SERIAL_CHANGED:
     case QEMU_PROCESS_EVENT_MONITOR_EOF:

src/qemu/qemu_domain.h

@@ -447,6 +447,7 @@ typedef enum {
     QEMU_PROCESS_EVENT_WATCHDOG = 0,
     QEMU_PROCESS_EVENT_GUESTPANIC,
     QEMU_PROCESS_EVENT_DEVICE_DELETED,
+    QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED,
     QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED,
     QEMU_PROCESS_EVENT_SERIAL_CHANGED,
     QEMU_PROCESS_EVENT_JOB_STATUS_CHANGE,

src/qemu/qemu_driver.c

@@ -40,6 +40,7 @@
 #include "qemu_hostdev.h"
 #include "qemu_hotplug.h"
 #include "qemu_monitor.h"
+#include "qemu_passt.h"
 #include "qemu_process.h"
 #include "qemu_migration.h"
 #include "qemu_migration_params.h"
@@ -3622,6 +3623,78 @@ processDeviceDeletedEvent(virQEMUDriver *driver,
 }
 
 
+static void
+processNetdevStreamDisconnectedEvent(virDomainObj *vm,
+                                     const char *netdevId)
+{
+    virDomainDeviceDef dev;
+    virDomainNetDef *def;
+    virQEMUCaps *qemuCaps = QEMU_DOMAIN_PRIVATE(vm)->qemuCaps;
+    const char *devAlias = STRSKIP(netdevId, "host");
+
+    /* The event sends us the "netdev-id", but we don't store the
+     * netdev-id in the NetDef and thus can't use it to find the
+     * correct NetDef. We *do* keep the device alias in the NetDef,
+     * and by convention the netdev-id is always "host" + devAlias, so
+     * we just need to remove "host" from the front of netdev-id to
+     * get the alias, which we can then use to find the proper NetDef.
+     */
+
+    if (!devAlias) {
+        VIR_WARN("Received NETDEV_STREAM_DISCONNECTED event for unrecognized netdev %s from domain %p %s",
+                  netdevId, vm, vm->def->name);
+        return;
+    }
+
+    VIR_DEBUG("Received NETDEV_STREAM_DISCONNECTED event for device %s from domain %p %s",
+              devAlias, vm, vm->def->name);
+
+    if (virDomainObjBeginJob(vm, VIR_JOB_QUERY) < 0)
+        return;
+
+    if (!virDomainObjIsActive(vm)) {
+        VIR_DEBUG("Domain is not running");
+        goto endjob;
+    }
+
+    if (virDomainDefFindDevice(vm->def, devAlias, &dev, true) < 0) {
+        VIR_WARN("NETDEV_STREAM_DISCONNECTED event received for non-existent device %s in domain %s",
+                 devAlias, vm->def->name);
+        goto endjob;
+    }
+    if (dev.type != VIR_DOMAIN_DEVICE_NET) {
+        VIR_WARN("NETDEV_STREAM_DISCONNECTED event received for non-network device %s in domain %s",
+                 devAlias, vm->def->name);
+        goto endjob;
+    }
+    def = dev.data.net;
+
+    if (def->backend.type != VIR_DOMAIN_NET_BACKEND_PASST) {
+        VIR_DEBUG("ignore NETDEV_STREAM_DISCONNECTED event for non-passt network device %s in domain %s",
+                  def->info.alias, vm->def->name);
+        goto endjob;
+    }
+
+    if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_NETDEV_STREAM_RECONNECT)) {
+        VIR_WARN("ignore NETDEV_STREAM_DISCONNECTED event for passt network device %s in domain %s - QEMU binary does not support reconnect",
+                  def->info.alias, vm->def->name);
+        goto endjob;
+    }
+
+    /* handle the event - restart the passt process with its original
+     * parameters
+     */
+    VIR_DEBUG("process NETDEV_STREAM_DISCONNECTED event for network device %s in domain %s",
+              def->info.alias, vm->def->name);
+
+    if (qemuPasstStart(vm, def) < 0)
+        goto endjob;
+
+ endjob:
+    virDomainObjEndJob(vm);
+}
+
+
 static void
 processNicRxFilterChangedEvent(virDomainObj *vm,
                                const char *devAlias)
@@ -3971,6 +4044,9 @@ static void qemuProcessEventHandler(void *data, void *opaque)
     case QEMU_PROCESS_EVENT_DEVICE_DELETED:
         processDeviceDeletedEvent(driver, vm, processEvent->data);
         break;
+    case QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED:
+        processNetdevStreamDisconnectedEvent(vm, processEvent->data);
+        break;
     case QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED:
         processNicRxFilterChangedEvent(vm, processEvent->data);
         break;

src/qemu/qemu_monitor.c

@@ -1265,6 +1265,17 @@ qemuMonitorEmitNicRxFilterChanged(qemuMonitor *mon,
 }
 
 
+void
+qemuMonitorEmitNetdevStreamDisconnected(qemuMonitor *mon,
+                                        const char *devAlias)
+{
+    VIR_DEBUG("mon=%p", mon);
+
+    QEMU_MONITOR_CALLBACK(mon, domainNetdevStreamDisconnected,
+                          mon->vm, devAlias);
+}
+
+
 void
 qemuMonitorEmitSerialChange(qemuMonitor *mon,
                             const char *devAlias,

src/qemu/qemu_monitor.h

@@ -250,6 +250,9 @@ typedef void (*qemuMonitorDomainDeviceUnplugErrCallback)(qemuMonitor *mon,
                                                          virDomainObj *vm,
                                                          const char *devPath,
                                                          const char *devAlias);
+typedef void (*qemuMonitorDomainNetdevStreamDisconnectedCallback)(qemuMonitor *mon,
+                                                                  virDomainObj *vm,
+                                                                  const char *devAlias);
 typedef void (*qemuMonitorDomainNicRxFilterChangedCallback)(qemuMonitor *mon,
                                                             virDomainObj *vm,
                                                             const char *devAlias);
@@ -397,6 +400,7 @@ struct _qemuMonitorCallbacks {
     qemuMonitorDomainMemoryFailureCallback domainMemoryFailure;
     qemuMonitorDomainMemoryDeviceSizeChange domainMemoryDeviceSizeChange;
     qemuMonitorDomainDeviceUnplugErrCallback domainDeviceUnplugError;
+    qemuMonitorDomainNetdevStreamDisconnectedCallback domainNetdevStreamDisconnected;
 };
 
 qemuMonitor *qemuMonitorOpen(virDomainObj *vm,
@@ -480,6 +484,8 @@ void qemuMonitorEmitDeviceDeleted(qemuMonitor *mon,
 void qemuMonitorEmitDeviceUnplugErr(qemuMonitor *mon,
                                     const char *devPath,
                                     const char *devAlias);
+void qemuMonitorEmitNetdevStreamDisconnected(qemuMonitor *mon,
+                                             const char *devAlias);
 void qemuMonitorEmitNicRxFilterChanged(qemuMonitor *mon,
                                        const char *devAlias);
 void qemuMonitorEmitSerialChange(qemuMonitor *mon,

src/qemu/qemu_monitor_json.c

@@ -84,6 +84,7 @@ static void qemuMonitorJSONHandleRdmaGidStatusChanged(qemuMonitor *mon, virJSONV
 static void qemuMonitorJSONHandleMemoryFailure(qemuMonitor *mon, virJSONValue *data);
 static void qemuMonitorJSONHandleMemoryDeviceSizeChange(qemuMonitor *mon, virJSONValue *data);
 static void qemuMonitorJSONHandleDeviceUnplugErr(qemuMonitor *mon, virJSONValue *data);
+static void qemuMonitorJSONHandleNetdevStreamDisconnected(qemuMonitor *mon, virJSONValue *data);
 
 typedef struct {
     const char *type;
@@ -106,6 +107,7 @@ static qemuEventHandler eventHandlers[] = {
     { "MEMORY_FAILURE", qemuMonitorJSONHandleMemoryFailure, },
     { "MIGRATION", qemuMonitorJSONHandleMigrationStatus, },
     { "MIGRATION_PASS", qemuMonitorJSONHandleMigrationPass, },
+    { "NETDEV_STREAM_DISCONNECTED", qemuMonitorJSONHandleNetdevStreamDisconnected, },
     { "NIC_RX_FILTER_CHANGED", qemuMonitorJSONHandleNicRxFilterChanged, },
     { "PR_MANAGER_STATUS_CHANGED", qemuMonitorJSONHandlePRManagerStatusChanged, },
     { "RDMA_GID_STATUS_CHANGED", qemuMonitorJSONHandleRdmaGidStatusChanged, },
@@ -1021,6 +1023,20 @@ qemuMonitorJSONHandleDeviceUnplugErr(qemuMonitor *mon, virJSONValue *data)
 }
 
 
+static void
+qemuMonitorJSONHandleNetdevStreamDisconnected(qemuMonitor *mon, virJSONValue *data)
+{
+    const char *name;
+
+    if (!(name = virJSONValueObjectGetString(data, "netdev-id"))) {
+        VIR_WARN("missing device in NETDEV_STREAM_DISCONNECTED event");
+        return;
+    }
+
+    qemuMonitorEmitNetdevStreamDisconnected(mon, name);
+}
+
+
 static void
 qemuMonitorJSONHandleNicRxFilterChanged(qemuMonitor *mon, virJSONValue *data)
 {

src/qemu/qemu_process.c

@@ -1361,6 +1361,23 @@ qemuProcessHandleBlockThreshold(qemuMonitor *mon G_GNUC_UNUSED,
 }
 
 
+static void
+qemuProcessHandleNetdevStreamDisconnected(qemuMonitor *mon G_GNUC_UNUSED,
+                                          virDomainObj *vm,
+                                          const char *devAlias)
+{
+    virObjectLock(vm);
+
+    VIR_DEBUG("Device %s Netdev Stream Disconnected in domain %p %s",
+              devAlias, vm, vm->def->name);
+
+    qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED,
+                           0, 0, g_strdup(devAlias));
+
+    virObjectUnlock(vm);
+}
+
+
 static void
 qemuProcessHandleNicRxFilterChanged(qemuMonitor *mon G_GNUC_UNUSED,
                                     virDomainObj *vm,
@@ -1802,6 +1819,7 @@ static qemuMonitorCallbacks monitorCallbacks = {
     .domainMemoryFailure = qemuProcessHandleMemoryFailure,
     .domainMemoryDeviceSizeChange = qemuProcessHandleMemoryDeviceSizeChange,
     .domainDeviceUnplugError = qemuProcessHandleDeviceUnplugErr,
+    .domainNetdevStreamDisconnected = qemuProcessHandleNetdevStreamDisconnected,
 };
 
 static void