{"id":807512,"url":"http://patchwork.ozlabs.org/api/1.0/patches/807512/?format=json","project":{"id":14,"url":"http://patchwork.ozlabs.org/api/1.0/projects/14/?format=json","name":"QEMU Development","link_name":"qemu-devel","list_id":"qemu-devel.nongnu.org","list_email":"qemu-devel@nongnu.org","web_url":"","scm_url":"","webscm_url":""},"msgid":"<1504081950-2528-10-git-send-email-peterx@redhat.com>","date":"2017-08-30T08:32:06","name":"[RFC,v2,09/33] migration: implement \"postcopy-pause\" src logic","commit_ref":null,"pull_url":null,"state":"new","archived":false,"hash":"ae3515ac9d7239c17feb2124569dff01c31a46c3","submitter":{"id":67717,"url":"http://patchwork.ozlabs.org/api/1.0/people/67717/?format=json","name":"Peter Xu","email":"peterx@redhat.com"},"delegate":null,"mbox":"http://patchwork.ozlabs.org/project/qemu-devel/patch/1504081950-2528-10-git-send-email-peterx@redhat.com/mbox/","series":[{"id":552,"url":"http://patchwork.ozlabs.org/api/1.0/series/552/?format=json","date":"2017-08-30T08:31:59","name":"Migration: postcopy failure recovery","version":2,"mbox":"http://patchwork.ozlabs.org/series/552/mbox/"}],"check":"pending","checks":"http://patchwork.ozlabs.org/api/patches/807512/checks/","tags":{},"headers":{"Return-Path":"<qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org>","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=<UNKNOWN>)","ext-mx03.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx03.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=peterx@redhat.com"],"Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xhzlL3FN3z9t2Q\n\tfor <incoming@patchwork.ozlabs.org>;\n\tWed, 30 Aug 2017 18:51:58 +1000 (AEST)","from localhost ([::1]:49096 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t<qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org>)\n\tid 1dmyj6-0006E8-06\n\tfor incoming@patchwork.ozlabs.org; Wed, 30 Aug 2017 04:51:56 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:34061)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from <peterx@redhat.com>) id 1dmyR5-00078n-2y\n\tfor qemu-devel@nongnu.org; Wed, 30 Aug 2017 04:33:20 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from <peterx@redhat.com>) id 1dmyR3-0003ID-Ng\n\tfor qemu-devel@nongnu.org; Wed, 30 Aug 2017 04:33:19 -0400","from mx1.redhat.com ([209.132.183.28]:47414)\n\tby eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32)\n\t(Exim 4.71) (envelope-from <peterx@redhat.com>) id 1dmyR3-0003Hs-FH\n\tfor qemu-devel@nongnu.org; Wed, 30 Aug 2017 04:33:17 -0400","from smtp.corp.redhat.com\n\t(int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby mx1.redhat.com (Postfix) with ESMTPS id 7FF1E80F6C;\n\tWed, 30 Aug 2017 08:33:16 +0000 (UTC)","from pxdev.xzpeter.org.com (dhcp-14-103.nay.redhat.com\n\t[10.66.14.103])\n\tby smtp.corp.redhat.com (Postfix) with ESMTP id 7C022871DD;\n\tWed, 30 Aug 2017 08:33:13 +0000 (UTC)"],"DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 7FF1E80F6C","From":"Peter Xu <peterx@redhat.com>","To":"qemu-devel@nongnu.org","Date":"Wed, 30 Aug 2017 16:32:06 +0800","Message-Id":"<1504081950-2528-10-git-send-email-peterx@redhat.com>","In-Reply-To":"<1504081950-2528-1-git-send-email-peterx@redhat.com>","References":"<1504081950-2528-1-git-send-email-peterx@redhat.com>","X-Scanned-By":"MIMEDefang 2.79 on 10.5.11.15","X-Greylist":"Sender IP whitelisted, not delayed by milter-greylist-4.5.16\n\t(mx1.redhat.com [10.5.110.27]);\n\tWed, 30 Aug 2017 08:33:16 +0000 (UTC)","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic]\n\t[fuzzy]","X-Received-From":"209.132.183.28","Subject":"[Qemu-devel] [RFC v2 09/33] migration: implement \"postcopy-pause\"\n\tsrc logic","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"<qemu-devel.nongnu.org>","List-Unsubscribe":"<https://lists.nongnu.org/mailman/options/qemu-devel>,\n\t<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>","List-Archive":"<http://lists.nongnu.org/archive/html/qemu-devel/>","List-Post":"<mailto:qemu-devel@nongnu.org>","List-Help":"<mailto:qemu-devel-request@nongnu.org?subject=help>","List-Subscribe":"<https://lists.nongnu.org/mailman/listinfo/qemu-devel>,\n\t<mailto:qemu-devel-request@nongnu.org?subject=subscribe>","Cc":"Laurent Vivier <lvivier@redhat.com>,\n\tAndrea Arcangeli <aarcange@redhat.com>, \n\tJuan Quintela <quintela@redhat.com>,\n\tAlexey Perevalov <a.perevalov@samsung.com>, peterx@redhat.com,\n\t\"Dr . David Alan Gilbert\" <dgilbert@redhat.com>","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t<qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org>"},"content":"Now when network down for postcopy, the source side will not fail the\nmigration. Instead we convert the status into this new paused state, and\nwe will try to wait for a rescue in the future.\n\nIf a recovery is detected, migration_thread() will reset its local\nvariables to prepare for that.\n\nSigned-off-by: Peter Xu <peterx@redhat.com>\n---\n migration/migration.c  | 98 +++++++++++++++++++++++++++++++++++++++++++++++---\n migration/migration.h  |  3 ++\n migration/trace-events |  1 +\n 3 files changed, 98 insertions(+), 4 deletions(-)","diff":"diff --git a/migration/migration.c b/migration/migration.c\nindex f6130db..8d26ea8 100644\n--- a/migration/migration.c\n+++ b/migration/migration.c\n@@ -993,6 +993,8 @@ static void migrate_fd_cleanup(void *opaque)\n \n     notifier_list_notify(&migration_state_notifiers, s);\n     block_cleanup_parameters(s);\n+\n+    qemu_sem_destroy(&s->postcopy_pause_sem);\n }\n \n void migrate_fd_error(MigrationState *s, const Error *error)\n@@ -1136,6 +1138,7 @@ MigrationState *migrate_init(void)\n     s->migration_thread_running = false;\n     error_free(s->error);\n     s->error = NULL;\n+    qemu_sem_init(&s->postcopy_pause_sem, 0);\n \n     migrate_set_state(&s->state, MIGRATION_STATUS_NONE, MIGRATION_STATUS_SETUP);\n \n@@ -1938,6 +1941,80 @@ bool migrate_colo_enabled(void)\n     return s->enabled_capabilities[MIGRATION_CAPABILITY_X_COLO];\n }\n \n+typedef enum MigThrError {\n+    /* No error detected */\n+    MIG_THR_ERR_NONE = 0,\n+    /* Detected error, but resumed successfully */\n+    MIG_THR_ERR_RECOVERED = 1,\n+    /* Detected fatal error, need to exit */\n+    MIG_THR_ERR_FATAL = 2,\n+} MigThrError;\n+\n+/*\n+ * We don't return until we are in a safe state to continue current\n+ * postcopy migration.  Returns MIG_THR_ERR_RECOVERED if recovered, or\n+ * MIG_THR_ERR_FATAL if unrecovery failure happened.\n+ */\n+static MigThrError postcopy_pause(MigrationState *s)\n+{\n+    assert(s->state == MIGRATION_STATUS_POSTCOPY_ACTIVE);\n+    migrate_set_state(&s->state, MIGRATION_STATUS_POSTCOPY_ACTIVE,\n+                      MIGRATION_STATUS_POSTCOPY_PAUSED);\n+\n+    /* Current channel is possibly broken. Release it. */\n+    assert(s->to_dst_file);\n+    qemu_file_shutdown(s->to_dst_file);\n+    qemu_fclose(s->to_dst_file);\n+    s->to_dst_file = NULL;\n+\n+    error_report(\"Detected IO failure for postcopy. \"\n+                 \"Migration paused.\");\n+\n+    /*\n+     * We wait until things fixed up. Then someone will setup the\n+     * status back for us.\n+     */\n+    while (s->state == MIGRATION_STATUS_POSTCOPY_PAUSED) {\n+        qemu_sem_wait(&s->postcopy_pause_sem);\n+    }\n+\n+    trace_postcopy_pause_continued();\n+\n+    return MIG_THR_ERR_RECOVERED;\n+}\n+\n+static MigThrError migration_detect_error(MigrationState *s)\n+{\n+    int ret;\n+\n+    /* Try to detect any file errors */\n+    ret = qemu_file_get_error(s->to_dst_file);\n+\n+    if (!ret) {\n+        /* Everything is fine */\n+        return MIG_THR_ERR_NONE;\n+    }\n+\n+    if (s->state == MIGRATION_STATUS_POSTCOPY_ACTIVE && ret == -EIO) {\n+        /*\n+         * For postcopy, we allow the network to be down for a\n+         * while. After that, it can be continued by a\n+         * recovery phase.\n+         */\n+        return postcopy_pause(s);\n+    } else {\n+        /*\n+         * For precopy (or postcopy with error outside IO), we fail\n+         * with no time.\n+         */\n+        migrate_set_state(&s->state, s->state, MIGRATION_STATUS_FAILED);\n+        trace_migration_thread_file_err();\n+\n+        /* Time to stop the migration, now. */\n+        return MIG_THR_ERR_FATAL;\n+    }\n+}\n+\n /*\n  * Master migration thread on the source VM.\n  * It drives the migration and pumps the data down the outgoing channel.\n@@ -1962,6 +2039,7 @@ static void *migration_thread(void *opaque)\n     /* The active state we expect to be in; ACTIVE or POSTCOPY_ACTIVE */\n     enum MigrationStatus current_active_state = MIGRATION_STATUS_ACTIVE;\n     bool enable_colo = migrate_colo_enabled();\n+    MigThrError thr_error;\n \n     rcu_register_thread();\n \n@@ -2034,12 +2112,24 @@ static void *migration_thread(void *opaque)\n             }\n         }\n \n-        if (qemu_file_get_error(s->to_dst_file)) {\n-            migrate_set_state(&s->state, current_active_state,\n-                              MIGRATION_STATUS_FAILED);\n-            trace_migration_thread_file_err();\n+        /*\n+         * Try to detect any kind of failures, and see whether we\n+         * should stop the migration now.\n+         */\n+        thr_error = migration_detect_error(s);\n+        if (thr_error == MIG_THR_ERR_FATAL) {\n+            /* Stop migration */\n             break;\n+        } else if (thr_error == MIG_THR_ERR_RECOVERED) {\n+            /*\n+             * Just recovered from a e.g. network failure, reset all\n+             * the local variables. This is important to avoid\n+             * breaking transferred_bytes and bandwidth calculation\n+             */\n+            initial_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);\n+            initial_bytes = 0;\n         }\n+\n         current_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);\n         if (current_time >= initial_time + BUFFER_DELAY) {\n             uint64_t transferred_bytes = qemu_ftell(s->to_dst_file) -\ndiff --git a/migration/migration.h b/migration/migration.h\nindex 70e3094..0c957c9 100644\n--- a/migration/migration.h\n+++ b/migration/migration.h\n@@ -149,6 +149,9 @@ struct MigrationState\n     bool send_configuration;\n     /* Whether we send section footer during migration */\n     bool send_section_footer;\n+\n+    /* Needed by postcopy-pause state */\n+    QemuSemaphore postcopy_pause_sem;\n };\n \n void migrate_set_state(int *state, int old_state, int new_state);\ndiff --git a/migration/trace-events b/migration/trace-events\nindex d2910a6..907564b 100644\n--- a/migration/trace-events\n+++ b/migration/trace-events\n@@ -98,6 +98,7 @@ migration_thread_setup_complete(void) \"\"\n open_return_path_on_source(void) \"\"\n open_return_path_on_source_continue(void) \"\"\n postcopy_start(void) \"\"\n+postcopy_pause_continued(void) \"\"\n postcopy_start_set_run(void) \"\"\n source_return_path_thread_bad_end(void) \"\"\n source_return_path_thread_end(void) \"\"\n","prefixes":["RFC","v2","09/33"]}