2 * COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
3 * (a.k.a. Fault Tolerance or Continuous Replication)
5 * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
6 * Copyright (c) 2016 FUJITSU LIMITED
7 * Copyright (c) 2016 Intel Corporation
9 * This work is licensed under the terms of the GNU GPL, version 2 or
10 * later. See the COPYING file in the top-level directory.
13 #include "qemu/osdep.h"
14 #include "sysemu/sysemu.h"
15 #include "qapi/error.h"
16 #include "qapi/qapi-commands-migration.h"
17 #include "qemu-file-channel.h"
18 #include "migration.h"
19 #include "qemu-file.h"
21 #include "migration/colo.h"
23 #include "io/channel-buffer.h"
25 #include "qemu/error-report.h"
26 #include "migration/failover.h"
27 #include "replication.h"
29 static bool vmstate_loading
;
31 #define COLO_BUFFER_BASE_SIZE (4 * 1024 * 1024)
33 bool migration_in_colo_state(void)
35 MigrationState
*s
= migrate_get_current();
37 return (s
->state
== MIGRATION_STATUS_COLO
);
40 bool migration_incoming_in_colo_state(void)
42 MigrationIncomingState
*mis
= migration_incoming_get_current();
44 return mis
&& (mis
->state
== MIGRATION_STATUS_COLO
);
47 static bool colo_runstate_is_stopped(void)
49 return runstate_check(RUN_STATE_COLO
) || !runstate_is_running();
52 static void secondary_vm_do_failover(void)
55 MigrationIncomingState
*mis
= migration_incoming_get_current();
57 /* Can not do failover during the process of VM's loading VMstate, Or
58 * it will break the secondary VM.
60 if (vmstate_loading
) {
61 old_state
= failover_set_state(FAILOVER_STATUS_ACTIVE
,
62 FAILOVER_STATUS_RELAUNCH
);
63 if (old_state
!= FAILOVER_STATUS_ACTIVE
) {
64 error_report("Unknown error while do failover for secondary VM,"
65 "old_state: %s", FailoverStatus_str(old_state
));
70 migrate_set_state(&mis
->state
, MIGRATION_STATUS_COLO
,
71 MIGRATION_STATUS_COMPLETED
);
74 error_report("\"-S\" qemu option will be ignored in secondary side");
75 /* recover runstate to normal migration finish state */
79 * Make sure COLO incoming thread not block in recv or send,
80 * If mis->from_src_file and mis->to_src_file use the same fd,
81 * The second shutdown() will return -1, we ignore this value,
84 if (mis
->from_src_file
) {
85 qemu_file_shutdown(mis
->from_src_file
);
87 if (mis
->to_src_file
) {
88 qemu_file_shutdown(mis
->to_src_file
);
91 old_state
= failover_set_state(FAILOVER_STATUS_ACTIVE
,
92 FAILOVER_STATUS_COMPLETED
);
93 if (old_state
!= FAILOVER_STATUS_ACTIVE
) {
94 error_report("Incorrect state (%s) while doing failover for "
95 "secondary VM", FailoverStatus_str(old_state
));
98 /* Notify COLO incoming thread that failover work is finished */
99 qemu_sem_post(&mis
->colo_incoming_sem
);
100 /* For Secondary VM, jump to incoming co */
101 if (mis
->migration_incoming_co
) {
102 qemu_coroutine_enter(mis
->migration_incoming_co
);
106 static void primary_vm_do_failover(void)
108 MigrationState
*s
= migrate_get_current();
111 migrate_set_state(&s
->state
, MIGRATION_STATUS_COLO
,
112 MIGRATION_STATUS_COMPLETED
);
115 * Wake up COLO thread which may blocked in recv() or send(),
116 * The s->rp_state.from_dst_file and s->to_dst_file may use the
117 * same fd, but we still shutdown the fd for twice, it is harmless.
119 if (s
->to_dst_file
) {
120 qemu_file_shutdown(s
->to_dst_file
);
122 if (s
->rp_state
.from_dst_file
) {
123 qemu_file_shutdown(s
->rp_state
.from_dst_file
);
126 old_state
= failover_set_state(FAILOVER_STATUS_ACTIVE
,
127 FAILOVER_STATUS_COMPLETED
);
128 if (old_state
!= FAILOVER_STATUS_ACTIVE
) {
129 error_report("Incorrect state (%s) while doing failover for Primary VM",
130 FailoverStatus_str(old_state
));
133 /* Notify COLO thread that failover work is finished */
134 qemu_sem_post(&s
->colo_exit_sem
);
137 void colo_do_failover(MigrationState
*s
)
139 /* Make sure VM stopped while failover happened. */
140 if (!colo_runstate_is_stopped()) {
141 vm_stop_force_state(RUN_STATE_COLO
);
144 if (get_colo_mode() == COLO_MODE_PRIMARY
) {
145 primary_vm_do_failover();
147 secondary_vm_do_failover();
151 void qmp_xen_set_replication(bool enable
, bool primary
,
152 bool has_failover
, bool failover
,
155 #ifdef CONFIG_REPLICATION
156 ReplicationMode mode
= primary
?
157 REPLICATION_MODE_PRIMARY
:
158 REPLICATION_MODE_SECONDARY
;
160 if (has_failover
&& enable
) {
161 error_setg(errp
, "Parameter 'failover' is only for"
162 " stopping replication");
167 replication_start_all(mode
, errp
);
172 replication_stop_all(failover
, failover
? NULL
: errp
);
179 ReplicationStatus
*qmp_query_xen_replication_status(Error
**errp
)
181 #ifdef CONFIG_REPLICATION
183 ReplicationStatus
*s
= g_new0(ReplicationStatus
, 1);
185 replication_get_error_all(&err
);
189 s
->desc
= g_strdup(error_get_pretty(err
));
201 void qmp_xen_colo_do_checkpoint(Error
**errp
)
203 #ifdef CONFIG_REPLICATION
204 replication_do_checkpoint_all(errp
);
210 static void colo_send_message(QEMUFile
*f
, COLOMessage msg
,
215 if (msg
>= COLO_MESSAGE__MAX
) {
216 error_setg(errp
, "%s: Invalid message", __func__
);
219 qemu_put_be32(f
, msg
);
222 ret
= qemu_file_get_error(f
);
224 error_setg_errno(errp
, -ret
, "Can't send COLO message");
226 trace_colo_send_message(COLOMessage_str(msg
));
229 static void colo_send_message_value(QEMUFile
*f
, COLOMessage msg
,
230 uint64_t value
, Error
**errp
)
232 Error
*local_err
= NULL
;
235 colo_send_message(f
, msg
, &local_err
);
237 error_propagate(errp
, local_err
);
240 qemu_put_be64(f
, value
);
243 ret
= qemu_file_get_error(f
);
245 error_setg_errno(errp
, -ret
, "Failed to send value for message:%s",
246 COLOMessage_str(msg
));
250 static COLOMessage
colo_receive_message(QEMUFile
*f
, Error
**errp
)
255 msg
= qemu_get_be32(f
);
256 ret
= qemu_file_get_error(f
);
258 error_setg_errno(errp
, -ret
, "Can't receive COLO message");
261 if (msg
>= COLO_MESSAGE__MAX
) {
262 error_setg(errp
, "%s: Invalid message", __func__
);
265 trace_colo_receive_message(COLOMessage_str(msg
));
269 static void colo_receive_check_message(QEMUFile
*f
, COLOMessage expect_msg
,
273 Error
*local_err
= NULL
;
275 msg
= colo_receive_message(f
, &local_err
);
277 error_propagate(errp
, local_err
);
280 if (msg
!= expect_msg
) {
281 error_setg(errp
, "Unexpected COLO message %d, expected %d",
286 static uint64_t colo_receive_message_value(QEMUFile
*f
, uint32_t expect_msg
,
289 Error
*local_err
= NULL
;
293 colo_receive_check_message(f
, expect_msg
, &local_err
);
295 error_propagate(errp
, local_err
);
299 value
= qemu_get_be64(f
);
300 ret
= qemu_file_get_error(f
);
302 error_setg_errno(errp
, -ret
, "Failed to get value for COLO message: %s",
303 COLOMessage_str(expect_msg
));
308 static int colo_do_checkpoint_transaction(MigrationState
*s
,
309 QIOChannelBuffer
*bioc
,
312 Error
*local_err
= NULL
;
315 colo_send_message(s
->to_dst_file
, COLO_MESSAGE_CHECKPOINT_REQUEST
,
321 colo_receive_check_message(s
->rp_state
.from_dst_file
,
322 COLO_MESSAGE_CHECKPOINT_REPLY
, &local_err
);
326 /* Reset channel-buffer directly */
327 qio_channel_io_seek(QIO_CHANNEL(bioc
), 0, 0, NULL
);
330 qemu_mutex_lock_iothread();
331 if (failover_get_state() != FAILOVER_STATUS_NONE
) {
332 qemu_mutex_unlock_iothread();
335 vm_stop_force_state(RUN_STATE_COLO
);
336 qemu_mutex_unlock_iothread();
337 trace_colo_vm_state_change("run", "stop");
339 * Failover request bh could be called after vm_stop_force_state(),
340 * So we need check failover_request_is_active() again.
342 if (failover_get_state() != FAILOVER_STATUS_NONE
) {
346 /* Disable block migration */
347 migrate_set_block_enabled(false, &local_err
);
348 qemu_savevm_state_header(fb
);
349 qemu_savevm_state_setup(fb
);
350 qemu_mutex_lock_iothread();
351 qemu_savevm_state_complete_precopy(fb
, false, false);
352 qemu_mutex_unlock_iothread();
356 colo_send_message(s
->to_dst_file
, COLO_MESSAGE_VMSTATE_SEND
, &local_err
);
361 * We need the size of the VMstate data in Secondary side,
362 * With which we can decide how much data should be read.
364 colo_send_message_value(s
->to_dst_file
, COLO_MESSAGE_VMSTATE_SIZE
,
365 bioc
->usage
, &local_err
);
370 qemu_put_buffer(s
->to_dst_file
, bioc
->data
, bioc
->usage
);
371 qemu_fflush(s
->to_dst_file
);
372 ret
= qemu_file_get_error(s
->to_dst_file
);
377 colo_receive_check_message(s
->rp_state
.from_dst_file
,
378 COLO_MESSAGE_VMSTATE_RECEIVED
, &local_err
);
383 colo_receive_check_message(s
->rp_state
.from_dst_file
,
384 COLO_MESSAGE_VMSTATE_LOADED
, &local_err
);
391 qemu_mutex_lock_iothread();
393 qemu_mutex_unlock_iothread();
394 trace_colo_vm_state_change("stop", "run");
398 error_report_err(local_err
);
403 static void colo_process_checkpoint(MigrationState
*s
)
405 QIOChannelBuffer
*bioc
;
407 int64_t current_time
= qemu_clock_get_ms(QEMU_CLOCK_HOST
);
408 Error
*local_err
= NULL
;
411 failover_init_state();
413 s
->rp_state
.from_dst_file
= qemu_file_get_return_path(s
->to_dst_file
);
414 if (!s
->rp_state
.from_dst_file
) {
415 error_report("Open QEMUFile from_dst_file failed");
420 * Wait for Secondary finish loading VM states and enter COLO
423 colo_receive_check_message(s
->rp_state
.from_dst_file
,
424 COLO_MESSAGE_CHECKPOINT_READY
, &local_err
);
428 bioc
= qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE
);
429 fb
= qemu_fopen_channel_output(QIO_CHANNEL(bioc
));
430 object_unref(OBJECT(bioc
));
432 qemu_mutex_lock_iothread();
434 qemu_mutex_unlock_iothread();
435 trace_colo_vm_state_change("stop", "run");
437 timer_mod(s
->colo_delay_timer
,
438 current_time
+ s
->parameters
.x_checkpoint_delay
);
440 while (s
->state
== MIGRATION_STATUS_COLO
) {
441 if (failover_get_state() != FAILOVER_STATUS_NONE
) {
442 error_report("failover request");
446 qemu_sem_wait(&s
->colo_checkpoint_sem
);
448 ret
= colo_do_checkpoint_transaction(s
, bioc
, fb
);
455 /* Throw the unreported error message after exited from loop */
457 error_report_err(local_err
);
464 timer_del(s
->colo_delay_timer
);
466 /* Hope this not to be too long to wait here */
467 qemu_sem_wait(&s
->colo_exit_sem
);
468 qemu_sem_destroy(&s
->colo_exit_sem
);
470 * Must be called after failover BH is completed,
471 * Or the failover BH may shutdown the wrong fd that
472 * re-used by other threads after we release here.
474 if (s
->rp_state
.from_dst_file
) {
475 qemu_fclose(s
->rp_state
.from_dst_file
);
479 void colo_checkpoint_notify(void *opaque
)
481 MigrationState
*s
= opaque
;
482 int64_t next_notify_time
;
484 qemu_sem_post(&s
->colo_checkpoint_sem
);
485 s
->colo_checkpoint_time
= qemu_clock_get_ms(QEMU_CLOCK_HOST
);
486 next_notify_time
= s
->colo_checkpoint_time
+
487 s
->parameters
.x_checkpoint_delay
;
488 timer_mod(s
->colo_delay_timer
, next_notify_time
);
491 void migrate_start_colo_process(MigrationState
*s
)
493 qemu_mutex_unlock_iothread();
494 qemu_sem_init(&s
->colo_checkpoint_sem
, 0);
495 s
->colo_delay_timer
= timer_new_ms(QEMU_CLOCK_HOST
,
496 colo_checkpoint_notify
, s
);
498 qemu_sem_init(&s
->colo_exit_sem
, 0);
499 migrate_set_state(&s
->state
, MIGRATION_STATUS_ACTIVE
,
500 MIGRATION_STATUS_COLO
);
501 colo_process_checkpoint(s
);
502 qemu_mutex_lock_iothread();
505 static void colo_wait_handle_message(QEMUFile
*f
, int *checkpoint_request
,
509 Error
*local_err
= NULL
;
511 msg
= colo_receive_message(f
, &local_err
);
513 error_propagate(errp
, local_err
);
518 case COLO_MESSAGE_CHECKPOINT_REQUEST
:
519 *checkpoint_request
= 1;
522 *checkpoint_request
= 0;
523 error_setg(errp
, "Got unknown COLO message: %d", msg
);
528 void *colo_process_incoming_thread(void *opaque
)
530 MigrationIncomingState
*mis
= opaque
;
532 QIOChannelBuffer
*bioc
= NULL
; /* Cache incoming device state */
535 Error
*local_err
= NULL
;
537 qemu_sem_init(&mis
->colo_incoming_sem
, 0);
539 migrate_set_state(&mis
->state
, MIGRATION_STATUS_ACTIVE
,
540 MIGRATION_STATUS_COLO
);
542 failover_init_state();
544 mis
->to_src_file
= qemu_file_get_return_path(mis
->from_src_file
);
545 if (!mis
->to_src_file
) {
546 error_report("COLO incoming thread: Open QEMUFile to_src_file failed");
550 * Note: the communication between Primary side and Secondary side
551 * should be sequential, we set the fd to unblocked in migration incoming
552 * coroutine, and here we are in the COLO incoming thread, so it is ok to
553 * set the fd back to blocked.
555 qemu_file_set_blocking(mis
->from_src_file
, true);
557 bioc
= qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE
);
558 fb
= qemu_fopen_channel_input(QIO_CHANNEL(bioc
));
559 object_unref(OBJECT(bioc
));
561 colo_send_message(mis
->to_src_file
, COLO_MESSAGE_CHECKPOINT_READY
,
567 while (mis
->state
== MIGRATION_STATUS_COLO
) {
570 colo_wait_handle_message(mis
->from_src_file
, &request
, &local_err
);
575 if (failover_get_state() != FAILOVER_STATUS_NONE
) {
576 error_report("failover request");
580 /* FIXME: This is unnecessary for periodic checkpoint mode */
581 colo_send_message(mis
->to_src_file
, COLO_MESSAGE_CHECKPOINT_REPLY
,
587 colo_receive_check_message(mis
->from_src_file
,
588 COLO_MESSAGE_VMSTATE_SEND
, &local_err
);
593 value
= colo_receive_message_value(mis
->from_src_file
,
594 COLO_MESSAGE_VMSTATE_SIZE
, &local_err
);
600 * Read VM device state data into channel buffer,
601 * It's better to re-use the memory allocated.
602 * Here we need to handle the channel buffer directly.
604 if (value
> bioc
->capacity
) {
605 bioc
->capacity
= value
;
606 bioc
->data
= g_realloc(bioc
->data
, bioc
->capacity
);
608 total_size
= qemu_get_buffer(mis
->from_src_file
, bioc
->data
, value
);
609 if (total_size
!= value
) {
610 error_report("Got %" PRIu64
" VMState data, less than expected"
611 " %" PRIu64
, total_size
, value
);
614 bioc
->usage
= total_size
;
615 qio_channel_io_seek(QIO_CHANNEL(bioc
), 0, 0, NULL
);
617 colo_send_message(mis
->to_src_file
, COLO_MESSAGE_VMSTATE_RECEIVED
,
623 qemu_mutex_lock_iothread();
624 qemu_system_reset(SHUTDOWN_CAUSE_NONE
);
625 vmstate_loading
= true;
626 if (qemu_loadvm_state(fb
) < 0) {
627 error_report("COLO: loadvm failed");
628 qemu_mutex_unlock_iothread();
632 vmstate_loading
= false;
633 qemu_mutex_unlock_iothread();
635 if (failover_get_state() == FAILOVER_STATUS_RELAUNCH
) {
636 failover_set_state(FAILOVER_STATUS_RELAUNCH
,
637 FAILOVER_STATUS_NONE
);
638 failover_request_active(NULL
);
642 colo_send_message(mis
->to_src_file
, COLO_MESSAGE_VMSTATE_LOADED
,
650 vmstate_loading
= false;
651 /* Throw the unreported error message after exited from loop */
653 error_report_err(local_err
);
660 /* Hope this not to be too long to loop here */
661 qemu_sem_wait(&mis
->colo_incoming_sem
);
662 qemu_sem_destroy(&mis
->colo_incoming_sem
);
663 /* Must be called after failover BH is completed */
664 if (mis
->to_src_file
) {
665 qemu_fclose(mis
->to_src_file
);
667 migration_incoming_exit_colo();