2016-10-27 08:42:52 +02:00
|
|
|
/*
|
|
|
|
* COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
|
|
|
|
* (a.k.a. Fault Tolerance or Continuous Replication)
|
|
|
|
*
|
|
|
|
* Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
|
|
|
|
* Copyright (c) 2016 FUJITSU LIMITED
|
|
|
|
* Copyright (c) 2016 Intel Corporation
|
|
|
|
*
|
|
|
|
* This work is licensed under the terms of the GNU GPL, version 2 or
|
|
|
|
* later. See the COPYING file in the top-level directory.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "qemu/osdep.h"
|
2016-10-27 08:42:54 +02:00
|
|
|
#include "sysemu/sysemu.h"
|
2018-02-01 12:18:31 +01:00
|
|
|
#include "qapi/error.h"
|
2018-02-11 10:36:01 +01:00
|
|
|
#include "qapi/qapi-commands-migration.h"
|
2017-04-24 20:07:27 +02:00
|
|
|
#include "migration.h"
|
2017-04-20 18:52:18 +02:00
|
|
|
#include "qemu-file.h"
|
2017-04-20 14:48:46 +02:00
|
|
|
#include "savevm.h"
|
2016-10-27 08:42:52 +02:00
|
|
|
#include "migration/colo.h"
|
2017-04-21 14:31:22 +02:00
|
|
|
#include "block.h"
|
2016-10-27 08:42:59 +02:00
|
|
|
#include "io/channel-buffer.h"
|
2016-10-27 08:42:54 +02:00
|
|
|
#include "trace.h"
|
2016-10-27 08:42:56 +02:00
|
|
|
#include "qemu/error-report.h"
|
Include qemu/main-loop.h less
In my "build everything" tree, changing qemu/main-loop.h triggers a
recompile of some 5600 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h). It includes block/aio.h,
which in turn includes qemu/event_notifier.h, qemu/notify.h,
qemu/processor.h, qemu/qsp.h, qemu/queue.h, qemu/thread-posix.h,
qemu/thread.h, qemu/timer.h, and a few more.
Include qemu/main-loop.h only where it's needed. Touching it now
recompiles only some 1700 objects. For block/aio.h and
qemu/event_notifier.h, these numbers drop from 5600 to 2800. For the
others, they shrink only slightly.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190812052359.30071-21-armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
2019-08-12 07:23:50 +02:00
|
|
|
#include "qemu/main-loop.h"
|
2019-08-12 07:23:46 +02:00
|
|
|
#include "qemu/rcu.h"
|
COLO: Add 'x-colo-lost-heartbeat' command to trigger failover
We leave users to choose whatever heartbeat solution they want,
if the heartbeat is lost, or other errors they detect, they can use
experimental command 'x_colo_lost_heartbeat' to tell COLO to do failover,
COLO will do operations accordingly.
For example, if the command is sent to the Primary side,
the Primary side will exit COLO mode, does cleanup work,
and then, PVM will take over the service work. If sent to the Secondary side,
the Secondary side will run failover work, then takes over PVM's service work.
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Eric Blake <eblake@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Amit Shah <amit@amitshah.net>
2016-10-27 08:43:03 +02:00
|
|
|
#include "migration/failover.h"
|
2020-02-24 07:54:10 +01:00
|
|
|
#include "migration/ram.h"
|
2018-12-13 13:37:24 +01:00
|
|
|
#ifdef CONFIG_REPLICATION
|
2021-05-17 13:36:56 +02:00
|
|
|
#include "block/replication.h"
|
2018-12-13 13:37:24 +01:00
|
|
|
#endif
|
2018-09-03 06:38:45 +02:00
|
|
|
#include "net/colo-compare.h"
|
|
|
|
#include "net/colo.h"
|
2018-09-03 06:38:46 +02:00
|
|
|
#include "block/block.h"
|
2018-09-03 06:38:51 +02:00
|
|
|
#include "qapi/qapi-events-migration.h"
|
2018-09-03 06:38:53 +02:00
|
|
|
#include "qapi/qmp/qerror.h"
|
2018-09-03 06:38:54 +02:00
|
|
|
#include "sysemu/cpus.h"
|
2019-08-12 07:23:59 +02:00
|
|
|
#include "sysemu/runstate.h"
|
2018-09-03 06:38:58 +02:00
|
|
|
#include "net/filter.h"
|
2016-10-27 08:42:52 +02:00
|
|
|
|
2017-01-17 13:57:44 +01:00
|
|
|
static bool vmstate_loading;
|
2018-09-03 06:38:45 +02:00
|
|
|
static Notifier packets_compare_notifier;
|
2017-01-17 13:57:44 +01:00
|
|
|
|
2019-03-22 11:13:33 +01:00
|
|
|
/* User need to know colo mode after COLO failover */
|
|
|
|
static COLOMode last_colo_mode;
|
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
#define COLO_BUFFER_BASE_SIZE (4 * 1024 * 1024)
|
|
|
|
|
2016-10-27 08:42:54 +02:00
|
|
|
bool migration_in_colo_state(void)
|
|
|
|
{
|
|
|
|
MigrationState *s = migrate_get_current();
|
|
|
|
|
|
|
|
return (s->state == MIGRATION_STATUS_COLO);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:55 +02:00
|
|
|
bool migration_incoming_in_colo_state(void)
|
|
|
|
{
|
|
|
|
MigrationIncomingState *mis = migration_incoming_get_current();
|
|
|
|
|
|
|
|
return mis && (mis->state == MIGRATION_STATUS_COLO);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:05 +02:00
|
|
|
static bool colo_runstate_is_stopped(void)
|
|
|
|
{
|
|
|
|
return runstate_check(RUN_STATE_COLO) || !runstate_is_running();
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:06 +02:00
|
|
|
static void secondary_vm_do_failover(void)
|
|
|
|
{
|
2018-11-01 03:12:26 +01:00
|
|
|
/* COLO needs enable block-replication */
|
|
|
|
#ifdef CONFIG_REPLICATION
|
2016-10-27 08:43:06 +02:00
|
|
|
int old_state;
|
|
|
|
MigrationIncomingState *mis = migration_incoming_get_current();
|
2018-09-03 06:38:46 +02:00
|
|
|
Error *local_err = NULL;
|
2016-10-27 08:43:06 +02:00
|
|
|
|
2017-01-17 13:57:44 +01:00
|
|
|
/* Can not do failover during the process of VM's loading VMstate, Or
|
|
|
|
* it will break the secondary VM.
|
|
|
|
*/
|
|
|
|
if (vmstate_loading) {
|
|
|
|
old_state = failover_set_state(FAILOVER_STATUS_ACTIVE,
|
|
|
|
FAILOVER_STATUS_RELAUNCH);
|
|
|
|
if (old_state != FAILOVER_STATUS_ACTIVE) {
|
|
|
|
error_report("Unknown error while do failover for secondary VM,"
|
2017-08-24 10:46:08 +02:00
|
|
|
"old_state: %s", FailoverStatus_str(old_state));
|
2017-01-17 13:57:44 +01:00
|
|
|
}
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:06 +02:00
|
|
|
migrate_set_state(&mis->state, MIGRATION_STATUS_COLO,
|
|
|
|
MIGRATION_STATUS_COMPLETED);
|
|
|
|
|
2018-09-03 06:38:46 +02:00
|
|
|
replication_stop_all(true, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_report_err(local_err);
|
2020-03-24 16:36:28 +01:00
|
|
|
local_err = NULL;
|
2018-09-03 06:38:46 +02:00
|
|
|
}
|
|
|
|
|
2018-09-03 06:38:58 +02:00
|
|
|
/* Notify all filters of all NIC to do checkpoint */
|
|
|
|
colo_notify_filters_event(COLO_EVENT_FAILOVER, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_report_err(local_err);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:06 +02:00
|
|
|
if (!autostart) {
|
|
|
|
error_report("\"-S\" qemu option will be ignored in secondary side");
|
|
|
|
/* recover runstate to normal migration finish state */
|
|
|
|
autostart = true;
|
|
|
|
}
|
2017-01-17 13:57:43 +01:00
|
|
|
/*
|
|
|
|
* Make sure COLO incoming thread not block in recv or send,
|
|
|
|
* If mis->from_src_file and mis->to_src_file use the same fd,
|
|
|
|
* The second shutdown() will return -1, we ignore this value,
|
|
|
|
* It is harmless.
|
|
|
|
*/
|
|
|
|
if (mis->from_src_file) {
|
|
|
|
qemu_file_shutdown(mis->from_src_file);
|
|
|
|
}
|
|
|
|
if (mis->to_src_file) {
|
|
|
|
qemu_file_shutdown(mis->to_src_file);
|
|
|
|
}
|
2016-10-27 08:43:06 +02:00
|
|
|
|
|
|
|
old_state = failover_set_state(FAILOVER_STATUS_ACTIVE,
|
|
|
|
FAILOVER_STATUS_COMPLETED);
|
|
|
|
if (old_state != FAILOVER_STATUS_ACTIVE) {
|
|
|
|
error_report("Incorrect state (%s) while doing failover for "
|
2017-08-24 10:46:08 +02:00
|
|
|
"secondary VM", FailoverStatus_str(old_state));
|
2016-10-27 08:43:06 +02:00
|
|
|
return;
|
|
|
|
}
|
2017-01-17 13:57:43 +01:00
|
|
|
/* Notify COLO incoming thread that failover work is finished */
|
|
|
|
qemu_sem_post(&mis->colo_incoming_sem);
|
2019-03-22 11:13:30 +01:00
|
|
|
|
2016-10-27 08:43:06 +02:00
|
|
|
/* For Secondary VM, jump to incoming co */
|
|
|
|
if (mis->migration_incoming_co) {
|
|
|
|
qemu_coroutine_enter(mis->migration_incoming_co);
|
|
|
|
}
|
2018-11-01 03:12:26 +01:00
|
|
|
#else
|
|
|
|
abort();
|
|
|
|
#endif
|
2016-10-27 08:43:06 +02:00
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:05 +02:00
|
|
|
static void primary_vm_do_failover(void)
|
|
|
|
{
|
2018-11-01 03:12:26 +01:00
|
|
|
#ifdef CONFIG_REPLICATION
|
2016-10-27 08:43:05 +02:00
|
|
|
MigrationState *s = migrate_get_current();
|
|
|
|
int old_state;
|
2018-09-03 06:38:46 +02:00
|
|
|
Error *local_err = NULL;
|
2016-10-27 08:43:05 +02:00
|
|
|
|
|
|
|
migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
|
|
|
|
MIGRATION_STATUS_COMPLETED);
|
2018-09-03 06:38:59 +02:00
|
|
|
/*
|
|
|
|
* kick COLO thread which might wait at
|
|
|
|
* qemu_sem_wait(&s->colo_checkpoint_sem).
|
|
|
|
*/
|
2021-11-01 08:56:59 +01:00
|
|
|
colo_checkpoint_notify(s);
|
2016-10-27 08:43:05 +02:00
|
|
|
|
2017-01-17 13:57:43 +01:00
|
|
|
/*
|
|
|
|
* Wake up COLO thread which may blocked in recv() or send(),
|
|
|
|
* The s->rp_state.from_dst_file and s->to_dst_file may use the
|
|
|
|
* same fd, but we still shutdown the fd for twice, it is harmless.
|
|
|
|
*/
|
|
|
|
if (s->to_dst_file) {
|
|
|
|
qemu_file_shutdown(s->to_dst_file);
|
|
|
|
}
|
|
|
|
if (s->rp_state.from_dst_file) {
|
|
|
|
qemu_file_shutdown(s->rp_state.from_dst_file);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:05 +02:00
|
|
|
old_state = failover_set_state(FAILOVER_STATUS_ACTIVE,
|
|
|
|
FAILOVER_STATUS_COMPLETED);
|
|
|
|
if (old_state != FAILOVER_STATUS_ACTIVE) {
|
|
|
|
error_report("Incorrect state (%s) while doing failover for Primary VM",
|
2017-08-24 10:46:08 +02:00
|
|
|
FailoverStatus_str(old_state));
|
2016-10-27 08:43:05 +02:00
|
|
|
return;
|
|
|
|
}
|
2018-09-03 06:38:46 +02:00
|
|
|
|
|
|
|
replication_stop_all(true, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_report_err(local_err);
|
|
|
|
local_err = NULL;
|
|
|
|
}
|
|
|
|
|
2017-01-17 13:57:43 +01:00
|
|
|
/* Notify COLO thread that failover work is finished */
|
|
|
|
qemu_sem_post(&s->colo_exit_sem);
|
2018-11-01 03:12:26 +01:00
|
|
|
#else
|
|
|
|
abort();
|
|
|
|
#endif
|
2016-10-27 08:43:05 +02:00
|
|
|
}
|
|
|
|
|
2018-09-03 06:38:47 +02:00
|
|
|
COLOMode get_colo_mode(void)
|
|
|
|
{
|
|
|
|
if (migration_in_colo_state()) {
|
|
|
|
return COLO_MODE_PRIMARY;
|
|
|
|
} else if (migration_incoming_in_colo_state()) {
|
|
|
|
return COLO_MODE_SECONDARY;
|
|
|
|
} else {
|
2018-09-03 06:38:52 +02:00
|
|
|
return COLO_MODE_NONE;
|
2018-09-03 06:38:47 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-04-26 11:07:28 +02:00
|
|
|
void colo_do_failover(void)
|
2016-10-27 08:43:05 +02:00
|
|
|
{
|
|
|
|
/* Make sure VM stopped while failover happened. */
|
|
|
|
if (!colo_runstate_is_stopped()) {
|
|
|
|
vm_stop_force_state(RUN_STATE_COLO);
|
|
|
|
}
|
|
|
|
|
2021-11-01 08:57:04 +01:00
|
|
|
switch (last_colo_mode = get_colo_mode()) {
|
2019-03-22 11:13:32 +01:00
|
|
|
case COLO_MODE_PRIMARY:
|
2016-10-27 08:43:05 +02:00
|
|
|
primary_vm_do_failover();
|
2019-03-22 11:13:32 +01:00
|
|
|
break;
|
|
|
|
case COLO_MODE_SECONDARY:
|
2016-10-27 08:43:06 +02:00
|
|
|
secondary_vm_do_failover();
|
2019-03-22 11:13:32 +01:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
error_report("colo_do_failover failed because the colo mode"
|
|
|
|
" could not be obtained");
|
2016-10-27 08:43:05 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-12-13 13:37:24 +01:00
|
|
|
#ifdef CONFIG_REPLICATION
|
2017-02-28 20:01:56 +01:00
|
|
|
void qmp_xen_set_replication(bool enable, bool primary,
|
|
|
|
bool has_failover, bool failover,
|
|
|
|
Error **errp)
|
|
|
|
{
|
|
|
|
ReplicationMode mode = primary ?
|
|
|
|
REPLICATION_MODE_PRIMARY :
|
|
|
|
REPLICATION_MODE_SECONDARY;
|
|
|
|
|
|
|
|
if (has_failover && enable) {
|
|
|
|
error_setg(errp, "Parameter 'failover' is only for"
|
|
|
|
" stopping replication");
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (enable) {
|
|
|
|
replication_start_all(mode, errp);
|
|
|
|
} else {
|
|
|
|
if (!has_failover) {
|
|
|
|
failover = NULL;
|
|
|
|
}
|
|
|
|
replication_stop_all(failover, failover ? NULL : errp);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-02-25 03:46:04 +01:00
|
|
|
ReplicationStatus *qmp_query_xen_replication_status(Error **errp)
|
|
|
|
{
|
|
|
|
Error *err = NULL;
|
|
|
|
ReplicationStatus *s = g_new0(ReplicationStatus, 1);
|
|
|
|
|
|
|
|
replication_get_error_all(&err);
|
|
|
|
if (err) {
|
|
|
|
s->error = true;
|
|
|
|
s->desc = g_strdup(error_get_pretty(err));
|
|
|
|
} else {
|
|
|
|
s->error = false;
|
|
|
|
}
|
|
|
|
|
|
|
|
error_free(err);
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
void qmp_xen_colo_do_checkpoint(Error **errp)
|
|
|
|
{
|
2020-04-22 15:07:16 +02:00
|
|
|
Error *err = NULL;
|
|
|
|
|
|
|
|
replication_do_checkpoint_all(&err);
|
|
|
|
if (err) {
|
|
|
|
error_propagate(errp, err);
|
|
|
|
return;
|
|
|
|
}
|
2019-06-09 18:44:33 +02:00
|
|
|
/* Notify all filters of all NIC to do checkpoint */
|
|
|
|
colo_notify_filters_event(COLO_EVENT_CHECKPOINT, errp);
|
2017-02-25 03:46:04 +01:00
|
|
|
}
|
2018-12-13 13:37:24 +01:00
|
|
|
#endif
|
2017-02-25 03:46:04 +01:00
|
|
|
|
2018-09-03 06:38:53 +02:00
|
|
|
COLOStatus *qmp_query_colo_status(Error **errp)
|
|
|
|
{
|
|
|
|
COLOStatus *s = g_new0(COLOStatus, 1);
|
|
|
|
|
|
|
|
s->mode = get_colo_mode();
|
2019-03-22 11:13:33 +01:00
|
|
|
s->last_mode = last_colo_mode;
|
2018-09-03 06:38:53 +02:00
|
|
|
|
|
|
|
switch (failover_get_state()) {
|
|
|
|
case FAILOVER_STATUS_NONE:
|
|
|
|
s->reason = COLO_EXIT_REASON_NONE;
|
|
|
|
break;
|
2019-03-22 11:13:30 +01:00
|
|
|
case FAILOVER_STATUS_COMPLETED:
|
2018-09-03 06:38:53 +02:00
|
|
|
s->reason = COLO_EXIT_REASON_REQUEST;
|
|
|
|
break;
|
|
|
|
default:
|
2019-03-22 11:13:31 +01:00
|
|
|
if (migration_in_colo_state()) {
|
|
|
|
s->reason = COLO_EXIT_REASON_PROCESSING;
|
|
|
|
} else {
|
|
|
|
s->reason = COLO_EXIT_REASON_ERROR;
|
|
|
|
}
|
2018-09-03 06:38:53 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:57 +02:00
|
|
|
static void colo_send_message(QEMUFile *f, COLOMessage msg,
|
|
|
|
Error **errp)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (msg >= COLO_MESSAGE__MAX) {
|
|
|
|
error_setg(errp, "%s: Invalid message", __func__);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
qemu_put_be32(f, msg);
|
|
|
|
qemu_fflush(f);
|
|
|
|
|
|
|
|
ret = qemu_file_get_error(f);
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg_errno(errp, -ret, "Can't send COLO message");
|
|
|
|
}
|
2017-08-24 10:46:08 +02:00
|
|
|
trace_colo_send_message(COLOMessage_str(msg));
|
2016-10-27 08:42:57 +02:00
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
static void colo_send_message_value(QEMUFile *f, COLOMessage msg,
|
|
|
|
uint64_t value, Error **errp)
|
|
|
|
{
|
|
|
|
Error *local_err = NULL;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
colo_send_message(f, msg, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
qemu_put_be64(f, value);
|
|
|
|
qemu_fflush(f);
|
|
|
|
|
|
|
|
ret = qemu_file_get_error(f);
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg_errno(errp, -ret, "Failed to send value for message:%s",
|
2017-08-24 10:46:08 +02:00
|
|
|
COLOMessage_str(msg));
|
2016-10-27 08:42:59 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:57 +02:00
|
|
|
static COLOMessage colo_receive_message(QEMUFile *f, Error **errp)
|
|
|
|
{
|
|
|
|
COLOMessage msg;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
msg = qemu_get_be32(f);
|
|
|
|
ret = qemu_file_get_error(f);
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg_errno(errp, -ret, "Can't receive COLO message");
|
|
|
|
return msg;
|
|
|
|
}
|
|
|
|
if (msg >= COLO_MESSAGE__MAX) {
|
|
|
|
error_setg(errp, "%s: Invalid message", __func__);
|
|
|
|
return msg;
|
|
|
|
}
|
2017-08-24 10:46:08 +02:00
|
|
|
trace_colo_receive_message(COLOMessage_str(msg));
|
2016-10-27 08:42:57 +02:00
|
|
|
return msg;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void colo_receive_check_message(QEMUFile *f, COLOMessage expect_msg,
|
|
|
|
Error **errp)
|
|
|
|
{
|
|
|
|
COLOMessage msg;
|
|
|
|
Error *local_err = NULL;
|
|
|
|
|
|
|
|
msg = colo_receive_message(f, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
if (msg != expect_msg) {
|
|
|
|
error_setg(errp, "Unexpected COLO message %d, expected %d",
|
|
|
|
msg, expect_msg);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:00 +02:00
|
|
|
static uint64_t colo_receive_message_value(QEMUFile *f, uint32_t expect_msg,
|
|
|
|
Error **errp)
|
|
|
|
{
|
|
|
|
Error *local_err = NULL;
|
|
|
|
uint64_t value;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
colo_receive_check_message(f, expect_msg, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
value = qemu_get_be64(f);
|
|
|
|
ret = qemu_file_get_error(f);
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg_errno(errp, -ret, "Failed to get value for COLO message: %s",
|
2017-08-24 10:46:08 +02:00
|
|
|
COLOMessage_str(expect_msg));
|
2016-10-27 08:43:00 +02:00
|
|
|
}
|
|
|
|
return value;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
static int colo_do_checkpoint_transaction(MigrationState *s,
|
|
|
|
QIOChannelBuffer *bioc,
|
|
|
|
QEMUFile *fb)
|
2016-10-27 08:42:57 +02:00
|
|
|
{
|
|
|
|
Error *local_err = NULL;
|
2016-10-27 08:42:59 +02:00
|
|
|
int ret = -1;
|
2016-10-27 08:42:57 +02:00
|
|
|
|
|
|
|
colo_send_message(s->to_dst_file, COLO_MESSAGE_CHECKPOINT_REQUEST,
|
|
|
|
&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
colo_receive_check_message(s->rp_state.from_dst_file,
|
|
|
|
COLO_MESSAGE_CHECKPOINT_REPLY, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:59 +02:00
|
|
|
/* Reset channel-buffer directly */
|
|
|
|
qio_channel_io_seek(QIO_CHANNEL(bioc), 0, 0, NULL);
|
|
|
|
bioc->usage = 0;
|
2016-10-27 08:42:57 +02:00
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
qemu_mutex_lock_iothread();
|
2016-10-27 08:43:05 +02:00
|
|
|
if (failover_get_state() != FAILOVER_STATUS_NONE) {
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:59 +02:00
|
|
|
vm_stop_force_state(RUN_STATE_COLO);
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
trace_colo_vm_state_change("run", "stop");
|
2016-10-27 08:43:05 +02:00
|
|
|
/*
|
|
|
|
* Failover request bh could be called after vm_stop_force_state(),
|
|
|
|
* So we need check failover_request_is_active() again.
|
|
|
|
*/
|
|
|
|
if (failover_get_state() != FAILOVER_STATUS_NONE) {
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:59 +02:00
|
|
|
qemu_mutex_lock_iothread();
|
2018-11-01 03:12:26 +01:00
|
|
|
|
|
|
|
#ifdef CONFIG_REPLICATION
|
2018-09-03 06:38:46 +02:00
|
|
|
replication_do_checkpoint_all(&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
goto out;
|
|
|
|
}
|
2018-11-01 03:12:26 +01:00
|
|
|
#else
|
|
|
|
abort();
|
|
|
|
#endif
|
2016-10-27 08:42:57 +02:00
|
|
|
|
|
|
|
colo_send_message(s->to_dst_file, COLO_MESSAGE_VMSTATE_SEND, &local_err);
|
|
|
|
if (local_err) {
|
2018-09-03 06:38:54 +02:00
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
/* Note: device state is saved into buffer */
|
|
|
|
ret = qemu_save_device_state(fb);
|
|
|
|
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
if (ret < 0) {
|
2016-10-27 08:42:57 +02:00
|
|
|
goto out;
|
|
|
|
}
|
2021-11-09 04:04:54 +01:00
|
|
|
|
|
|
|
if (migrate_auto_converge()) {
|
|
|
|
mig_throttle_counter_reset();
|
|
|
|
}
|
2018-09-03 06:38:54 +02:00
|
|
|
/*
|
|
|
|
* Only save VM's live state, which not including device state.
|
|
|
|
* TODO: We may need a timeout mechanism to prevent COLO process
|
|
|
|
* to be blocked here.
|
|
|
|
*/
|
|
|
|
qemu_savevm_live_state(s->to_dst_file);
|
|
|
|
|
|
|
|
qemu_fflush(fb);
|
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
/*
|
|
|
|
* We need the size of the VMstate data in Secondary side,
|
|
|
|
* With which we can decide how much data should be read.
|
|
|
|
*/
|
|
|
|
colo_send_message_value(s->to_dst_file, COLO_MESSAGE_VMSTATE_SIZE,
|
|
|
|
bioc->usage, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:57 +02:00
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
qemu_put_buffer(s->to_dst_file, bioc->data, bioc->usage);
|
|
|
|
qemu_fflush(s->to_dst_file);
|
|
|
|
ret = qemu_file_get_error(s->to_dst_file);
|
|
|
|
if (ret < 0) {
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:57 +02:00
|
|
|
|
|
|
|
colo_receive_check_message(s->rp_state.from_dst_file,
|
|
|
|
COLO_MESSAGE_VMSTATE_RECEIVED, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2020-05-11 13:11:01 +02:00
|
|
|
qemu_event_reset(&s->colo_checkpoint_event);
|
|
|
|
colo_notify_compares_event(NULL, COLO_EVENT_CHECKPOINT, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:57 +02:00
|
|
|
colo_receive_check_message(s->rp_state.from_dst_file,
|
|
|
|
COLO_MESSAGE_VMSTATE_LOADED, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
ret = 0;
|
|
|
|
|
|
|
|
qemu_mutex_lock_iothread();
|
|
|
|
vm_start();
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
trace_colo_vm_state_change("stop", "run");
|
2016-10-27 08:42:57 +02:00
|
|
|
|
|
|
|
out:
|
|
|
|
if (local_err) {
|
|
|
|
error_report_err(local_err);
|
|
|
|
}
|
2016-10-27 08:42:59 +02:00
|
|
|
return ret;
|
2016-10-27 08:42:57 +02:00
|
|
|
}
|
|
|
|
|
2018-09-03 06:38:45 +02:00
|
|
|
static void colo_compare_notify_checkpoint(Notifier *notifier, void *data)
|
|
|
|
{
|
|
|
|
colo_checkpoint_notify(data);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:54 +02:00
|
|
|
static void colo_process_checkpoint(MigrationState *s)
|
|
|
|
{
|
2016-10-27 08:42:59 +02:00
|
|
|
QIOChannelBuffer *bioc;
|
|
|
|
QEMUFile *fb = NULL;
|
2016-10-27 08:42:57 +02:00
|
|
|
Error *local_err = NULL;
|
|
|
|
int ret;
|
|
|
|
|
2021-11-01 08:57:04 +01:00
|
|
|
if (get_colo_mode() != COLO_MODE_PRIMARY) {
|
2019-03-22 11:13:33 +01:00
|
|
|
error_report("COLO mode must be COLO_MODE_PRIMARY");
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:04 +02:00
|
|
|
failover_init_state();
|
|
|
|
|
2016-10-27 08:42:56 +02:00
|
|
|
s->rp_state.from_dst_file = qemu_file_get_return_path(s->to_dst_file);
|
|
|
|
if (!s->rp_state.from_dst_file) {
|
|
|
|
error_report("Open QEMUFile from_dst_file failed");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2018-09-03 06:38:45 +02:00
|
|
|
packets_compare_notifier.notify = colo_compare_notify_checkpoint;
|
|
|
|
colo_compare_register_notifier(&packets_compare_notifier);
|
|
|
|
|
2016-10-27 08:42:57 +02:00
|
|
|
/*
|
|
|
|
* Wait for Secondary finish loading VM states and enter COLO
|
|
|
|
* restore.
|
|
|
|
*/
|
|
|
|
colo_receive_check_message(s->rp_state.from_dst_file,
|
|
|
|
COLO_MESSAGE_CHECKPOINT_READY, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:59 +02:00
|
|
|
bioc = qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE);
|
2022-06-20 13:02:05 +02:00
|
|
|
fb = qemu_file_new_output(QIO_CHANNEL(bioc));
|
2016-10-27 08:42:59 +02:00
|
|
|
object_unref(OBJECT(bioc));
|
2016-10-27 08:42:57 +02:00
|
|
|
|
2016-10-27 08:42:54 +02:00
|
|
|
qemu_mutex_lock_iothread();
|
2018-11-01 03:12:26 +01:00
|
|
|
#ifdef CONFIG_REPLICATION
|
2018-09-03 06:38:46 +02:00
|
|
|
replication_start_all(REPLICATION_MODE_PRIMARY, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
goto out;
|
|
|
|
}
|
2018-11-01 03:12:26 +01:00
|
|
|
#else
|
|
|
|
abort();
|
|
|
|
#endif
|
2018-09-03 06:38:46 +02:00
|
|
|
|
2016-10-27 08:42:54 +02:00
|
|
|
vm_start();
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
trace_colo_vm_state_change("stop", "run");
|
|
|
|
|
2021-11-10 18:41:56 +01:00
|
|
|
timer_mod(s->colo_delay_timer, qemu_clock_get_ms(QEMU_CLOCK_HOST) +
|
|
|
|
s->parameters.x_checkpoint_delay);
|
2017-01-17 13:57:42 +01:00
|
|
|
|
2016-10-27 08:42:57 +02:00
|
|
|
while (s->state == MIGRATION_STATUS_COLO) {
|
2016-10-27 08:43:05 +02:00
|
|
|
if (failover_get_state() != FAILOVER_STATUS_NONE) {
|
|
|
|
error_report("failover request");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2020-05-11 13:10:44 +02:00
|
|
|
qemu_event_wait(&s->colo_checkpoint_event);
|
2016-10-27 08:43:02 +02:00
|
|
|
|
2018-09-03 06:38:59 +02:00
|
|
|
if (s->state != MIGRATION_STATUS_COLO) {
|
|
|
|
goto out;
|
|
|
|
}
|
2016-10-27 08:42:59 +02:00
|
|
|
ret = colo_do_checkpoint_transaction(s, bioc, fb);
|
2016-10-27 08:42:57 +02:00
|
|
|
if (ret < 0) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
2016-10-27 08:42:54 +02:00
|
|
|
|
2016-10-27 08:42:56 +02:00
|
|
|
out:
|
2016-10-27 08:42:57 +02:00
|
|
|
/* Throw the unreported error message after exited from loop */
|
|
|
|
if (local_err) {
|
|
|
|
error_report_err(local_err);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:59 +02:00
|
|
|
if (fb) {
|
|
|
|
qemu_fclose(fb);
|
|
|
|
}
|
|
|
|
|
2018-09-03 06:38:51 +02:00
|
|
|
/*
|
|
|
|
* There are only two reasons we can get here, some error happened
|
|
|
|
* or the user triggered failover.
|
|
|
|
*/
|
|
|
|
switch (failover_get_state()) {
|
2019-03-22 11:13:30 +01:00
|
|
|
case FAILOVER_STATUS_COMPLETED:
|
2018-09-03 06:38:51 +02:00
|
|
|
qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
|
|
|
|
COLO_EXIT_REASON_REQUEST);
|
|
|
|
break;
|
|
|
|
default:
|
2019-03-22 11:13:31 +01:00
|
|
|
qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
|
|
|
|
COLO_EXIT_REASON_ERROR);
|
2018-09-03 06:38:51 +02:00
|
|
|
}
|
|
|
|
|
2017-01-17 13:57:43 +01:00
|
|
|
/* Hope this not to be too long to wait here */
|
|
|
|
qemu_sem_wait(&s->colo_exit_sem);
|
|
|
|
qemu_sem_destroy(&s->colo_exit_sem);
|
2018-09-03 06:38:45 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* It is safe to unregister notifier after failover finished.
|
|
|
|
* Besides, colo_delay_timer and colo_checkpoint_sem can't be
|
2020-09-17 09:50:21 +02:00
|
|
|
* released before unregister notifier, or there will be use-after-free
|
2018-09-03 06:38:45 +02:00
|
|
|
* error.
|
|
|
|
*/
|
|
|
|
colo_compare_unregister_notifier(&packets_compare_notifier);
|
|
|
|
timer_free(s->colo_delay_timer);
|
2020-05-11 13:10:44 +02:00
|
|
|
qemu_event_destroy(&s->colo_checkpoint_event);
|
2018-09-03 06:38:45 +02:00
|
|
|
|
2017-01-17 13:57:43 +01:00
|
|
|
/*
|
|
|
|
* Must be called after failover BH is completed,
|
|
|
|
* Or the failover BH may shutdown the wrong fd that
|
|
|
|
* re-used by other threads after we release here.
|
|
|
|
*/
|
2016-10-27 08:42:56 +02:00
|
|
|
if (s->rp_state.from_dst_file) {
|
|
|
|
qemu_fclose(s->rp_state.from_dst_file);
|
colo: fixed 'Segmentation fault' when the simplex mode PVM poweroff
The GDB statck is as follows:
Program terminated with signal SIGSEGV, Segmentation fault.
0 object_class_dynamic_cast (class=0x55c8f5d2bf50, typename=0x55c8f2f7379e "qio-channel") at qom/object.c:832
if (type->class->interfaces &&
[Current thread is 1 (Thread 0x7f756e97eb00 (LWP 1811577))]
(gdb) bt
0 object_class_dynamic_cast (class=0x55c8f5d2bf50, typename=0x55c8f2f7379e "qio-channel") at qom/object.c:832
1 0x000055c8f2c3dd14 in object_dynamic_cast (obj=0x55c8f543ac00, typename=0x55c8f2f7379e "qio-channel") at qom/object.c:763
2 0x000055c8f2c3ddce in object_dynamic_cast_assert (obj=0x55c8f543ac00, typename=0x55c8f2f7379e "qio-channel",
file=0x55c8f2f73780 "migration/qemu-file-channel.c", line=117, func=0x55c8f2f73800 <__func__.18724> "channel_shutdown") at qom/object.c:786
3 0x000055c8f2bbc6ac in channel_shutdown (opaque=0x55c8f543ac00, rd=true, wr=true, errp=0x0) at migration/qemu-file-channel.c:117
4 0x000055c8f2bba56e in qemu_file_shutdown (f=0x7f7558070f50) at migration/qemu-file.c:67
5 0x000055c8f2ba5373 in migrate_fd_cancel (s=0x55c8f4ccf3f0) at migration/migration.c:1699
6 0x000055c8f2ba1992 in migration_shutdown () at migration/migration.c:187
7 0x000055c8f29a5b77 in main (argc=69, argv=0x7fff3e9e8c08, envp=0x7fff3e9e8e38) at vl.c:4512
The root cause is that we still want to shutdown the from_dst_file in
migrate_fd_cancel() after qemu_close in colo_process_checkpoint().
So, we should set the s->rp_state.from_dst_file = NULL after
qemu_close().
Signed-off-by: Lei Rao <lei.rao@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2021-11-01 08:57:02 +01:00
|
|
|
s->rp_state.from_dst_file = NULL;
|
2016-10-27 08:42:56 +02:00
|
|
|
}
|
2016-10-27 08:42:54 +02:00
|
|
|
}
|
|
|
|
|
2017-01-17 13:57:42 +01:00
|
|
|
void colo_checkpoint_notify(void *opaque)
|
|
|
|
{
|
|
|
|
MigrationState *s = opaque;
|
|
|
|
int64_t next_notify_time;
|
|
|
|
|
2020-05-11 13:10:44 +02:00
|
|
|
qemu_event_set(&s->colo_checkpoint_event);
|
2017-01-17 13:57:42 +01:00
|
|
|
s->colo_checkpoint_time = qemu_clock_get_ms(QEMU_CLOCK_HOST);
|
|
|
|
next_notify_time = s->colo_checkpoint_time +
|
|
|
|
s->parameters.x_checkpoint_delay;
|
|
|
|
timer_mod(s->colo_delay_timer, next_notify_time);
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:54 +02:00
|
|
|
void migrate_start_colo_process(MigrationState *s)
|
|
|
|
{
|
|
|
|
qemu_mutex_unlock_iothread();
|
2020-05-11 13:10:44 +02:00
|
|
|
qemu_event_init(&s->colo_checkpoint_event, false);
|
2017-01-17 13:57:42 +01:00
|
|
|
s->colo_delay_timer = timer_new_ms(QEMU_CLOCK_HOST,
|
|
|
|
colo_checkpoint_notify, s);
|
|
|
|
|
2017-01-17 13:57:43 +01:00
|
|
|
qemu_sem_init(&s->colo_exit_sem, 0);
|
2016-10-27 08:42:54 +02:00
|
|
|
colo_process_checkpoint(s);
|
|
|
|
qemu_mutex_lock_iothread();
|
|
|
|
}
|
2016-10-27 08:42:55 +02:00
|
|
|
|
2020-02-24 07:54:08 +01:00
|
|
|
static void colo_incoming_process_checkpoint(MigrationIncomingState *mis,
|
|
|
|
QEMUFile *fb, QIOChannelBuffer *bioc, Error **errp)
|
|
|
|
{
|
|
|
|
uint64_t total_size;
|
|
|
|
uint64_t value;
|
|
|
|
Error *local_err = NULL;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
qemu_mutex_lock_iothread();
|
|
|
|
vm_stop_force_state(RUN_STATE_COLO);
|
|
|
|
qemu_mutex_unlock_iothread();
|
2021-12-14 18:19:45 +01:00
|
|
|
trace_colo_vm_state_change("run", "stop");
|
2020-02-24 07:54:08 +01:00
|
|
|
|
|
|
|
/* FIXME: This is unnecessary for periodic checkpoint mode */
|
|
|
|
colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_REPLY,
|
|
|
|
&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
colo_receive_check_message(mis->from_src_file,
|
|
|
|
COLO_MESSAGE_VMSTATE_SEND, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
qemu_mutex_lock_iothread();
|
2020-05-11 13:10:48 +02:00
|
|
|
cpu_synchronize_all_states();
|
2020-02-24 07:54:08 +01:00
|
|
|
ret = qemu_loadvm_state_main(mis->from_src_file, mis);
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg(errp, "Load VM's live state (ram) error");
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
value = colo_receive_message_value(mis->from_src_file,
|
|
|
|
COLO_MESSAGE_VMSTATE_SIZE, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Read VM device state data into channel buffer,
|
|
|
|
* It's better to re-use the memory allocated.
|
|
|
|
* Here we need to handle the channel buffer directly.
|
|
|
|
*/
|
|
|
|
if (value > bioc->capacity) {
|
|
|
|
bioc->capacity = value;
|
|
|
|
bioc->data = g_realloc(bioc->data, bioc->capacity);
|
|
|
|
}
|
|
|
|
total_size = qemu_get_buffer(mis->from_src_file, bioc->data, value);
|
|
|
|
if (total_size != value) {
|
|
|
|
error_setg(errp, "Got %" PRIu64 " VMState data, less than expected"
|
|
|
|
" %" PRIu64, total_size, value);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
bioc->usage = total_size;
|
|
|
|
qio_channel_io_seek(QIO_CHANNEL(bioc), 0, 0, NULL);
|
|
|
|
|
|
|
|
colo_send_message(mis->to_src_file, COLO_MESSAGE_VMSTATE_RECEIVED,
|
|
|
|
&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
qemu_mutex_lock_iothread();
|
|
|
|
vmstate_loading = true;
|
2020-05-11 13:10:51 +02:00
|
|
|
colo_flush_ram_cache();
|
2020-02-24 07:54:08 +01:00
|
|
|
ret = qemu_load_device_state(fb);
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg(errp, "COLO: load device state failed");
|
2020-05-11 13:10:55 +02:00
|
|
|
vmstate_loading = false;
|
2020-02-24 07:54:08 +01:00
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CONFIG_REPLICATION
|
|
|
|
replication_get_error_all(&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
2020-05-11 13:10:55 +02:00
|
|
|
vmstate_loading = false;
|
2020-02-24 07:54:08 +01:00
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* discard colo disk buffer */
|
|
|
|
replication_do_checkpoint_all(&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
2020-05-11 13:10:55 +02:00
|
|
|
vmstate_loading = false;
|
2020-02-24 07:54:08 +01:00
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
abort();
|
|
|
|
#endif
|
|
|
|
/* Notify all filters of all NIC to do checkpoint */
|
|
|
|
colo_notify_filters_event(COLO_EVENT_CHECKPOINT, &local_err);
|
|
|
|
|
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
2020-05-11 13:10:55 +02:00
|
|
|
vmstate_loading = false;
|
2020-02-24 07:54:08 +01:00
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
vmstate_loading = false;
|
|
|
|
vm_start();
|
|
|
|
qemu_mutex_unlock_iothread();
|
2021-12-14 18:19:45 +01:00
|
|
|
trace_colo_vm_state_change("stop", "run");
|
2020-02-24 07:54:08 +01:00
|
|
|
|
|
|
|
if (failover_get_state() == FAILOVER_STATUS_RELAUNCH) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
colo_send_message(mis->to_src_file, COLO_MESSAGE_VMSTATE_LOADED,
|
|
|
|
&local_err);
|
2020-07-22 10:40:47 +02:00
|
|
|
error_propagate(errp, local_err);
|
2020-02-24 07:54:08 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
static void colo_wait_handle_message(MigrationIncomingState *mis,
|
|
|
|
QEMUFile *fb, QIOChannelBuffer *bioc, Error **errp)
|
2016-10-27 08:42:57 +02:00
|
|
|
{
|
|
|
|
COLOMessage msg;
|
|
|
|
Error *local_err = NULL;
|
|
|
|
|
2020-02-24 07:54:08 +01:00
|
|
|
msg = colo_receive_message(mis->from_src_file, &local_err);
|
2016-10-27 08:42:57 +02:00
|
|
|
if (local_err) {
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
switch (msg) {
|
|
|
|
case COLO_MESSAGE_CHECKPOINT_REQUEST:
|
2020-02-24 07:54:08 +01:00
|
|
|
colo_incoming_process_checkpoint(mis, fb, bioc, errp);
|
2016-10-27 08:42:57 +02:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
error_setg(errp, "Got unknown COLO message: %d", msg);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-11-11 03:11:33 +01:00
|
|
|
void colo_shutdown(void)
|
|
|
|
{
|
|
|
|
MigrationIncomingState *mis = NULL;
|
|
|
|
MigrationState *s = NULL;
|
|
|
|
|
|
|
|
switch (get_colo_mode()) {
|
|
|
|
case COLO_MODE_PRIMARY:
|
|
|
|
s = migrate_get_current();
|
|
|
|
qemu_event_set(&s->colo_checkpoint_event);
|
|
|
|
qemu_sem_post(&s->colo_exit_sem);
|
|
|
|
break;
|
|
|
|
case COLO_MODE_SECONDARY:
|
|
|
|
mis = migration_incoming_get_current();
|
|
|
|
qemu_sem_post(&mis->colo_incoming_sem);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:42:55 +02:00
|
|
|
void *colo_process_incoming_thread(void *opaque)
|
|
|
|
{
|
|
|
|
MigrationIncomingState *mis = opaque;
|
2016-10-27 08:43:00 +02:00
|
|
|
QEMUFile *fb = NULL;
|
|
|
|
QIOChannelBuffer *bioc = NULL; /* Cache incoming device state */
|
2016-10-27 08:42:57 +02:00
|
|
|
Error *local_err = NULL;
|
2016-10-27 08:42:55 +02:00
|
|
|
|
2018-08-06 15:29:29 +02:00
|
|
|
rcu_register_thread();
|
2017-01-17 13:57:43 +01:00
|
|
|
qemu_sem_init(&mis->colo_incoming_sem, 0);
|
|
|
|
|
2016-10-27 08:42:55 +02:00
|
|
|
migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
|
|
|
|
MIGRATION_STATUS_COLO);
|
|
|
|
|
2021-11-01 08:57:04 +01:00
|
|
|
if (get_colo_mode() != COLO_MODE_SECONDARY) {
|
2019-03-22 11:13:33 +01:00
|
|
|
error_report("COLO mode must be COLO_MODE_SECONDARY");
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:04 +02:00
|
|
|
failover_init_state();
|
|
|
|
|
2016-10-27 08:42:56 +02:00
|
|
|
mis->to_src_file = qemu_file_get_return_path(mis->from_src_file);
|
|
|
|
if (!mis->to_src_file) {
|
|
|
|
error_report("COLO incoming thread: Open QEMUFile to_src_file failed");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* Note: the communication between Primary side and Secondary side
|
|
|
|
* should be sequential, we set the fd to unblocked in migration incoming
|
|
|
|
* coroutine, and here we are in the COLO incoming thread, so it is ok to
|
|
|
|
* set the fd back to blocked.
|
|
|
|
*/
|
|
|
|
qemu_file_set_blocking(mis->from_src_file, true);
|
|
|
|
|
2020-02-24 07:54:10 +01:00
|
|
|
colo_incoming_start_dirty_log();
|
|
|
|
|
2016-10-27 08:43:00 +02:00
|
|
|
bioc = qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE);
|
2022-06-20 13:02:05 +02:00
|
|
|
fb = qemu_file_new_input(QIO_CHANNEL(bioc));
|
2016-10-27 08:43:00 +02:00
|
|
|
object_unref(OBJECT(bioc));
|
|
|
|
|
2018-09-03 06:38:45 +02:00
|
|
|
qemu_mutex_lock_iothread();
|
2018-11-01 03:12:26 +01:00
|
|
|
#ifdef CONFIG_REPLICATION
|
2018-09-03 06:38:46 +02:00
|
|
|
replication_start_all(REPLICATION_MODE_SECONDARY, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
goto out;
|
|
|
|
}
|
2018-11-01 03:12:26 +01:00
|
|
|
#else
|
|
|
|
abort();
|
|
|
|
#endif
|
2018-09-03 06:38:45 +02:00
|
|
|
vm_start();
|
|
|
|
qemu_mutex_unlock_iothread();
|
2021-12-14 18:19:45 +01:00
|
|
|
trace_colo_vm_state_change("stop", "run");
|
2018-09-03 06:38:45 +02:00
|
|
|
|
2016-10-27 08:42:57 +02:00
|
|
|
colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_READY,
|
|
|
|
&local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mis->state == MIGRATION_STATUS_COLO) {
|
2020-02-24 07:54:08 +01:00
|
|
|
colo_wait_handle_message(mis, fb, bioc, &local_err);
|
2016-10-27 08:42:57 +02:00
|
|
|
if (local_err) {
|
2020-02-24 07:54:08 +01:00
|
|
|
error_report_err(local_err);
|
|
|
|
break;
|
2016-10-27 08:42:57 +02:00
|
|
|
}
|
2020-05-11 13:10:55 +02:00
|
|
|
|
|
|
|
if (failover_get_state() == FAILOVER_STATUS_RELAUNCH) {
|
|
|
|
failover_set_state(FAILOVER_STATUS_RELAUNCH,
|
|
|
|
FAILOVER_STATUS_NONE);
|
|
|
|
failover_request_active(NULL);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:06 +02:00
|
|
|
if (failover_get_state() != FAILOVER_STATUS_NONE) {
|
|
|
|
error_report("failover request");
|
2020-02-24 07:54:08 +01:00
|
|
|
break;
|
2016-10-27 08:42:57 +02:00
|
|
|
}
|
|
|
|
}
|
2016-10-27 08:42:55 +02:00
|
|
|
|
2016-10-27 08:42:56 +02:00
|
|
|
out:
|
2019-03-22 11:13:31 +01:00
|
|
|
/*
|
|
|
|
* There are only two reasons we can get here, some error happened
|
|
|
|
* or the user triggered failover.
|
|
|
|
*/
|
2018-09-03 06:38:51 +02:00
|
|
|
switch (failover_get_state()) {
|
2019-03-22 11:13:30 +01:00
|
|
|
case FAILOVER_STATUS_COMPLETED:
|
2018-09-03 06:38:51 +02:00
|
|
|
qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
|
|
|
|
COLO_EXIT_REASON_REQUEST);
|
|
|
|
break;
|
|
|
|
default:
|
2019-03-22 11:13:31 +01:00
|
|
|
qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
|
|
|
|
COLO_EXIT_REASON_ERROR);
|
2018-09-03 06:38:51 +02:00
|
|
|
}
|
|
|
|
|
2016-10-27 08:43:00 +02:00
|
|
|
if (fb) {
|
|
|
|
qemu_fclose(fb);
|
|
|
|
}
|
|
|
|
|
2017-01-17 13:57:43 +01:00
|
|
|
/* Hope this not to be too long to loop here */
|
|
|
|
qemu_sem_wait(&mis->colo_incoming_sem);
|
|
|
|
qemu_sem_destroy(&mis->colo_incoming_sem);
|
2016-10-27 08:42:55 +02:00
|
|
|
|
2018-08-06 15:29:29 +02:00
|
|
|
rcu_unregister_thread();
|
2016-10-27 08:42:55 +02:00
|
|
|
return NULL;
|
|
|
|
}
|