qemu_migration.c 92.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
/*
 * qemu_migration.c: QEMU migration handling
 *
 * Copyright (C) 2006-2011 Red Hat, Inc.
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307  USA
 *
 */

#include <config.h>

#include <sys/time.h>
25 26
#include <gnutls/gnutls.h>
#include <gnutls/x509.h>
27 28 29 30 31 32 33

#include "qemu_migration.h"
#include "qemu_monitor.h"
#include "qemu_domain.h"
#include "qemu_process.h"
#include "qemu_capabilities.h"
#include "qemu_audit.h"
34
#include "qemu_cgroup.h"
35 36 37 38 39 40 41 42

#include "logging.h"
#include "virterror_internal.h"
#include "memory.h"
#include "util.h"
#include "files.h"
#include "datatypes.h"
#include "fdstream.h"
43
#include "uuid.h"
44
#include "locking/domain_lock.h"
45

46 47 48

#define VIR_FROM_THIS VIR_FROM_QEMU

49
enum qemuMigrationCookieFlags {
50
    QEMU_MIGRATION_COOKIE_FLAG_GRAPHICS,
51
    QEMU_MIGRATION_COOKIE_FLAG_LOCKSTATE,
52 53 54 55 56 57 58

    QEMU_MIGRATION_COOKIE_FLAG_LAST
};

VIR_ENUM_DECL(qemuMigrationCookieFlag);
VIR_ENUM_IMPL(qemuMigrationCookieFlag,
              QEMU_MIGRATION_COOKIE_FLAG_LAST,
59
              "graphics", "lockstate");
60 61 62

enum qemuMigrationCookieFeatures {
    QEMU_MIGRATION_COOKIE_GRAPHICS  = (1 << QEMU_MIGRATION_COOKIE_FLAG_GRAPHICS),
63
    QEMU_MIGRATION_COOKIE_LOCKSTATE = (1 << QEMU_MIGRATION_COOKIE_FLAG_LOCKSTATE),
64 65 66 67 68 69 70 71 72 73 74 75
};

typedef struct _qemuMigrationCookieGraphics qemuMigrationCookieGraphics;
typedef qemuMigrationCookieGraphics *qemuMigrationCookieGraphicsPtr;
struct _qemuMigrationCookieGraphics {
    int type;
    int port;
    int tlsPort;
    char *listen;
    char *tlsSubject;
};

76 77 78 79
typedef struct _qemuMigrationCookie qemuMigrationCookie;
typedef qemuMigrationCookie *qemuMigrationCookiePtr;
struct _qemuMigrationCookie {
    int flags;
80
    int flagsMandatory;
81 82

    /* Host properties */
83 84 85 86
    unsigned char localHostuuid[VIR_UUID_BUFLEN];
    unsigned char remoteHostuuid[VIR_UUID_BUFLEN];
    char *localHostname;
    char *remoteHostname;
87 88 89 90

    /* Guest properties */
    unsigned char uuid[VIR_UUID_BUFLEN];
    char *name;
91

92 93 94 95
    /* If (flags & QEMU_MIGRATION_COOKIE_LOCKSTATE) */
    char *lockState;
    char *lockDriver;

96 97
    /* If (flags & QEMU_MIGRATION_COOKIE_GRAPHICS) */
    qemuMigrationCookieGraphicsPtr graphics;
98 99
};

100 101 102 103 104 105 106 107 108
static void qemuMigrationCookieGraphicsFree(qemuMigrationCookieGraphicsPtr grap)
{
    if (!grap)
        return;
    VIR_FREE(grap->listen);
    VIR_FREE(grap->tlsSubject);
    VIR_FREE(grap);
}

109 110 111 112 113 114

static void qemuMigrationCookieFree(qemuMigrationCookiePtr mig)
{
    if (!mig)
        return;

115 116 117
    if (mig->flags & QEMU_MIGRATION_COOKIE_GRAPHICS)
        qemuMigrationCookieGraphicsFree(mig->graphics);

118 119
    VIR_FREE(mig->localHostname);
    VIR_FREE(mig->remoteHostname);
120
    VIR_FREE(mig->name);
121 122
    VIR_FREE(mig->lockState);
    VIR_FREE(mig->lockDriver);
123 124 125 126
    VIR_FREE(mig);
}


127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
static char *
qemuDomainExtractTLSSubject(const char *certdir)
{
    char *certfile = NULL;
    char *subject = NULL;
    char *pemdata = NULL;
    gnutls_datum_t pemdatum;
    gnutls_x509_crt_t cert;
    int ret;
    size_t subjectlen;

    if (virAsprintf(&certfile, "%s/server-cert.pem", certdir) < 0)
        goto no_memory;

    if (virFileReadAll(certfile, 8192, &pemdata) < 0) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("unable to read server cert %s"), certfile);
        goto error;
    }

    ret = gnutls_x509_crt_init(&cert);
    if (ret < 0) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("cannot initialize cert object: %s"),
                        gnutls_strerror(ret));
        goto error;
    }

    pemdatum.data = (unsigned char *)pemdata;
    pemdatum.size = strlen(pemdata);

    ret = gnutls_x509_crt_import(cert, &pemdatum, GNUTLS_X509_FMT_PEM);
    if (ret < 0) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("cannot load cert data from %s: %s"),
                        certfile, gnutls_strerror(ret));
        goto error;
    }

    subjectlen = 1024;
    if (VIR_ALLOC_N(subject, subjectlen+1) < 0)
        goto no_memory;

    gnutls_x509_crt_get_dn(cert, subject, &subjectlen);
    subject[subjectlen] = '\0';

    VIR_FREE(certfile);
    VIR_FREE(pemdata);

    return subject;

no_memory:
    virReportOOMError();
error:
    VIR_FREE(certfile);
    VIR_FREE(pemdata);
    return NULL;
}


static qemuMigrationCookieGraphicsPtr
qemuMigrationCookieGraphicsAlloc(struct qemud_driver *driver,
                                 virDomainGraphicsDefPtr def)
{
    qemuMigrationCookieGraphicsPtr mig = NULL;
    const char *listenAddr;

    if (VIR_ALLOC(mig) < 0)
        goto no_memory;

    mig->type = def->type;
    if (mig->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC) {
        mig->port = def->data.vnc.port;
        listenAddr = def->data.vnc.listenAddr;
        if (!listenAddr)
            listenAddr = driver->vncListen;

        if (driver->vncTLS &&
            !(mig->tlsSubject = qemuDomainExtractTLSSubject(driver->vncTLSx509certdir)))
            goto error;
    } else {
        mig->port = def->data.spice.port;
        if (driver->spiceTLS)
            mig->tlsPort = def->data.spice.tlsPort;
        else
            mig->tlsPort = -1;
        listenAddr = def->data.spice.listenAddr;
        if (!listenAddr)
            listenAddr = driver->spiceListen;

        if (driver->spiceTLS &&
            !(mig->tlsSubject = qemuDomainExtractTLSSubject(driver->spiceTLSx509certdir)))
            goto error;
    }
    if (!(mig->listen = strdup(listenAddr)))
        goto no_memory;

    return mig;

no_memory:
    virReportOOMError();
error:
    qemuMigrationCookieGraphicsFree(mig);
    return NULL;
}


234 235 236 237 238 239 240 241 242 243 244 245
static qemuMigrationCookiePtr
qemuMigrationCookieNew(virDomainObjPtr dom)
{
    qemuMigrationCookiePtr mig = NULL;

    if (VIR_ALLOC(mig) < 0)
        goto no_memory;

    if (!(mig->name = strdup(dom->def->name)))
        goto no_memory;
    memcpy(mig->uuid, dom->def->uuid, VIR_UUID_BUFLEN);

246
    if (!(mig->localHostname = virGetHostname(NULL)))
247
        goto no_memory;
248
    if (virGetHostUUID(mig->localHostuuid) < 0) {
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("Unable to obtain host UUID"));
        goto error;
    }

    return mig;

no_memory:
    virReportOOMError();
error:
    qemuMigrationCookieFree(mig);
    return NULL;
}


264 265 266 267 268 269 270 271 272 273 274 275 276
static int
qemuMigrationCookieAddGraphics(qemuMigrationCookiePtr mig,
                               struct qemud_driver *driver,
                               virDomainObjPtr dom)
{
    if (mig->flags & QEMU_MIGRATION_COOKIE_GRAPHICS) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("Migration graphics data already present"));
        return -1;
    }

    if (dom->def->ngraphics == 1 &&
        (dom->def->graphics[0]->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC ||
277 278 279 280 281 282
         dom->def->graphics[0]->type == VIR_DOMAIN_GRAPHICS_TYPE_SPICE)) {
        if (!(mig->graphics =
              qemuMigrationCookieGraphicsAlloc(driver, dom->def->graphics[0])))
            return -1;
        mig->flags |= QEMU_MIGRATION_COOKIE_GRAPHICS;
    }
283 284 285 286 287

    return 0;
}


288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
static int
qemuMigrationCookieAddLockstate(qemuMigrationCookiePtr mig,
                                struct qemud_driver *driver,
                                virDomainObjPtr dom)
{
    qemuDomainObjPrivatePtr priv = dom->privateData;

    if (mig->flags & QEMU_MIGRATION_COOKIE_LOCKSTATE) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("Migration lockstate data already present"));
        return -1;
    }

    if (virDomainObjGetState(dom, NULL) == VIR_DOMAIN_PAUSED) {
        if (priv->lockState &&
            !(mig->lockState = strdup(priv->lockState)))
            return -1;
    } else {
        if (virDomainLockProcessInquire(driver->lockManager, dom, &mig->lockState) < 0)
            return -1;
    }

    if (!(mig->lockDriver = strdup(virLockManagerPluginGetName(driver->lockManager)))) {
        VIR_FREE(mig->lockState);
        return -1;
    }

    mig->flags |= QEMU_MIGRATION_COOKIE_LOCKSTATE;
    mig->flagsMandatory |= QEMU_MIGRATION_COOKIE_LOCKSTATE;

    return 0;
}



323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
static void qemuMigrationCookieGraphicsXMLFormat(virBufferPtr buf,
                                                 qemuMigrationCookieGraphicsPtr grap)
{
    virBufferAsprintf(buf, "  <graphics type='%s' port='%d' listen='%s'",
                      virDomainGraphicsTypeToString(grap->type),
                      grap->port, grap->listen);
    if (grap->type == VIR_DOMAIN_GRAPHICS_TYPE_SPICE)
        virBufferAsprintf(buf, " tlsPort='%d'", grap->tlsPort);
    if (grap->tlsSubject) {
        virBufferAddLit(buf, ">\n");
        virBufferEscapeString(buf, "    <cert info='subject' value='%s'/>\n", grap->tlsSubject);
        virBufferAddLit(buf, "  </graphics>\n");
    } else {
        virBufferAddLit(buf, "/>\n");
    }
}


341 342 343 344 345
static void qemuMigrationCookieXMLFormat(virBufferPtr buf,
                                         qemuMigrationCookiePtr mig)
{
    char uuidstr[VIR_UUID_STRING_BUFLEN];
    char hostuuidstr[VIR_UUID_STRING_BUFLEN];
346
    int i;
347 348

    virUUIDFormat(mig->uuid, uuidstr);
349
    virUUIDFormat(mig->localHostuuid, hostuuidstr);
350 351 352 353

    virBufferAsprintf(buf, "<qemu-migration>\n");
    virBufferEscapeString(buf, "  <name>%s</name>\n", mig->name);
    virBufferAsprintf(buf, "  <uuid>%s</uuid>\n", uuidstr);
354
    virBufferEscapeString(buf, "  <hostname>%s</hostname>\n", mig->localHostname);
355
    virBufferAsprintf(buf, "  <hostuuid>%s</hostuuid>\n", hostuuidstr);
356

357 358 359 360 361 362
    for (i = 0 ; i < QEMU_MIGRATION_COOKIE_FLAG_LAST ; i++) {
        if (mig->flagsMandatory & (1 << i))
            virBufferAsprintf(buf, "  <feature name='%s'/>\n",
                              qemuMigrationCookieFlagTypeToString(i));
    }

363 364
    if ((mig->flags & QEMU_MIGRATION_COOKIE_GRAPHICS) &&
        mig->graphics)
365 366
        qemuMigrationCookieGraphicsXMLFormat(buf, mig->graphics);

367 368 369 370 371 372 373 374 375
    if ((mig->flags & QEMU_MIGRATION_COOKIE_LOCKSTATE) &&
        mig->lockState) {
        virBufferAsprintf(buf, "  <lockstate driver='%s'>\n",
                          mig->lockDriver);
        virBufferAsprintf(buf, "    <leases>%s</leases>\n",
                          mig->lockState);
        virBufferAddLit(buf, "  </lockstate>\n");
    }

376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394
    virBufferAddLit(buf, "</qemu-migration>\n");
}


static char *qemuMigrationCookieXMLFormatStr(qemuMigrationCookiePtr mig)
{
    virBuffer buf = VIR_BUFFER_INITIALIZER;

    qemuMigrationCookieXMLFormat(&buf, mig);

    if (virBufferError(&buf)) {
        virReportOOMError();
        return NULL;
    }

    return virBufferContentAndReset(&buf);
}


395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
static qemuMigrationCookieGraphicsPtr
qemuMigrationCookieGraphicsXMLParse(xmlXPathContextPtr ctxt)
{
    qemuMigrationCookieGraphicsPtr grap;
    char *tmp;

    if (VIR_ALLOC(grap) < 0)
        goto no_memory;

    if (!(tmp = virXPathString("string(./graphics/@type)", ctxt))) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing type attribute in migration data"));
        goto error;
    }
    if ((grap->type = virDomainGraphicsTypeFromString(tmp)) < 0) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("unknown graphics type %s"), tmp);
        VIR_FREE(tmp);
        goto error;
    }
    if (virXPathInt("string(./graphics/@port)", ctxt, &grap->port) < 0) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing port attribute in migration data"));
        goto error;
    }
    if (grap->type == VIR_DOMAIN_GRAPHICS_TYPE_SPICE) {
        if (virXPathInt("string(./graphics/@tlsPort)", ctxt, &grap->tlsPort) < 0) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR,
                            "%s", _("missing tlsPort attribute in migration data"));
            goto error;
        }
    }
    if (!(grap->listen = virXPathString("string(./graphics/@listen)", ctxt))) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing listen attribute in migration data"));
        goto error;
    }
    /* Optional */
433
    grap->tlsSubject = virXPathString("string(./graphics/cert[@info='subject']/@value)", ctxt);
434 435 436 437 438 439 440 441 442 443 444 445


    return grap;

no_memory:
    virReportOOMError();
error:
    qemuMigrationCookieGraphicsFree(grap);
    return NULL;
}


446 447 448
static int
qemuMigrationCookieXMLParse(qemuMigrationCookiePtr mig,
                            xmlXPathContextPtr ctxt,
449
                            int flags)
450 451 452
{
    char uuidstr[VIR_UUID_STRING_BUFLEN];
    char *tmp;
453 454
    xmlNodePtr *nodes = NULL;
    int i, n;
455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490

    /* We don't store the uuid, name, hostname, or hostuuid
     * values. We just compare them to local data to do some
     * sanity checking on migration operation
     */

    /* Extract domain name */
    if (!(tmp = virXPathString("string(./name[1])", ctxt))) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing name element in migration data"));
        goto error;
    }
    if (STRNEQ(tmp, mig->name)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Incoming cookie data had unexpected name %s vs %s"),
                        tmp, mig->name);
        goto error;
    }
    VIR_FREE(tmp);

    /* Extract domain uuid */
    tmp = virXPathString("string(./uuid[1])", ctxt);
    if (!tmp) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing uuid element in migration data"));
        goto error;
    }
    virUUIDFormat(mig->uuid, uuidstr);
    if (STRNEQ(tmp, uuidstr)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Incoming cookie data had unexpected UUID %s vs %s"),
                        tmp, uuidstr);
    }
    VIR_FREE(tmp);

    /* Check & forbid "localhost" migration */
491
    if (!(mig->remoteHostname = virXPathString("string(./hostname[1])", ctxt))) {
492 493 494 495
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing hostname element in migration data"));
        goto error;
    }
496
    if (STREQ(mig->remoteHostname, mig->localHostname)) {
497 498
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Attempt to migrate guest to the same host %s"),
499
                        mig->remoteHostname);
500 501 502 503 504 505 506 507
        goto error;
    }

    if (!(tmp = virXPathString("string(./hostuuid[1])", ctxt))) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("missing hostuuid element in migration data"));
        goto error;
    }
508 509 510 511 512 513
    if (virUUIDParse(tmp, mig->remoteHostuuid) < 0) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        "%s", _("malformed hostuuid element in migration data"));
        goto error;
    }
    if (memcmp(mig->remoteHostuuid, mig->localHostuuid, VIR_UUID_BUFLEN) == 0) {
514 515 516 517 518 519 520
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Attempt to migrate guest to the same host %s"),
                        tmp);
        goto error;
    }
    VIR_FREE(tmp);

521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552
    /* Check to ensure all mandatory features from XML are also
     * present in 'flags' */
    if ((n = virXPathNodeSet("./features", ctxt, &nodes)) < 0)
        goto error;

    for (i = 0 ; i < n ; i++) {
        int val;
        char *str = virXMLPropString(nodes[i], "name");
        if (!str) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR,
                            "%s", _("missing feature name"));
            goto error;
        }

        if ((val = qemuMigrationCookieFlagTypeFromString(str)) < 0) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR,
                            _("Unknown migration cookie feature %s"),
                            str);
            VIR_FREE(str);
            goto error;
        }

        if ((flags & (1 << val)) == 0) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR,
                            _("Unsupported migration cookie feature %s"),
                            str);
            VIR_FREE(str);
        }
        VIR_FREE(str);
    }
    VIR_FREE(nodes);

553 554 555 556 557
    if ((flags & QEMU_MIGRATION_COOKIE_GRAPHICS) &&
        virXPathBoolean("count(./graphics) > 0", ctxt) &&
        (!(mig->graphics = qemuMigrationCookieGraphicsXMLParse(ctxt))))
        goto error;

558 559 560 561 562 563 564 565 566 567 568 569 570
    if ((flags & QEMU_MIGRATION_COOKIE_LOCKSTATE) &&
        virXPathBoolean("count(./lockstate) > 0", ctxt)) {
        mig->lockDriver = virXPathString("string(./lockstate[1]/@driver)", ctxt);
        if (!mig->lockDriver) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                            _("Missing lock driver name in migration cookie"));
            goto error;
        }
        mig->lockState = virXPathString("string(./lockstate[1]/leases[1])", ctxt);
        if (mig->lockState && STREQ(mig->lockState, ""))
            VIR_FREE(mig->lockState);
    }

571 572 573 574
    return 0;

error:
    VIR_FREE(tmp);
575
    VIR_FREE(nodes);
576 577 578 579 580 581 582 583 584 585 586
    return -1;
}


static int
qemuMigrationCookieXMLParseStr(qemuMigrationCookiePtr mig,
                               const char *xml,
                               int flags)
{
    xmlDocPtr doc = NULL;
    xmlXPathContextPtr ctxt = NULL;
587
    int ret = -1;
588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612

    VIR_DEBUG("xml=%s", NULLSTR(xml));

    if (!(doc = virXMLParseString(xml, "qemumigration.xml")))
        goto cleanup;

    if ((ctxt = xmlXPathNewContext(doc)) == NULL) {
        virReportOOMError();
        goto cleanup;
    }

    ctxt->node = xmlDocGetRootElement(doc);

    ret = qemuMigrationCookieXMLParse(mig, ctxt, flags);

cleanup:
    xmlXPathFreeContext(ctxt);
    xmlFreeDoc(doc);

    return ret;
}


static int
qemuMigrationBakeCookie(qemuMigrationCookiePtr mig,
613 614
                        struct qemud_driver *driver,
                        virDomainObjPtr dom,
615 616
                        char **cookieout,
                        int *cookieoutlen,
617
                        int flags)
618
{
619 620
    if (!cookieout || !cookieoutlen)
        return 0;
621 622 623

    *cookieoutlen = 0;

624 625 626 627
    if (flags & QEMU_MIGRATION_COOKIE_GRAPHICS &&
        qemuMigrationCookieAddGraphics(mig, driver, dom) < 0)
        return -1;

628 629 630 631
    if (flags & QEMU_MIGRATION_COOKIE_LOCKSTATE &&
        qemuMigrationCookieAddLockstate(mig, driver, dom) < 0)
        return -1;

632 633 634 635 636 637 638 639 640 641 642 643
    if (!(*cookieout = qemuMigrationCookieXMLFormatStr(mig)))
        return -1;

    *cookieoutlen = strlen(*cookieout) + 1;

    VIR_DEBUG("cookielen=%d cookie=%s", *cookieoutlen, *cookieout);

    return 0;
}


static qemuMigrationCookiePtr
644 645
qemuMigrationEatCookie(struct qemud_driver *driver,
                       virDomainObjPtr dom,
646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670
                       const char *cookiein,
                       int cookieinlen,
                       int flags)
{
    qemuMigrationCookiePtr mig = NULL;

    /* Parse & validate incoming cookie (if any) */
    if (cookiein && cookieinlen &&
        cookiein[cookieinlen-1] != '\0') {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("Migration cookie was not NULL terminated"));
        goto error;
    }

    VIR_DEBUG("cookielen=%d cookie='%s'", cookieinlen, NULLSTR(cookiein));

    if (!(mig = qemuMigrationCookieNew(dom)))
        return NULL;

    if (cookiein && cookieinlen &&
        qemuMigrationCookieXMLParseStr(mig,
                                       cookiein,
                                       flags) < 0)
        goto error;

671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688
    if (mig->flags & QEMU_MIGRATION_COOKIE_LOCKSTATE) {
        if (!mig->lockDriver) {
            if (virLockManagerPluginUsesState(driver->lockManager)) {
                qemuReportError(VIR_ERR_INTERNAL_ERROR,
                                _("Missing %s lock state for migration cookie"),
                                virLockManagerPluginGetName(driver->lockManager));
                goto error;
            }
        } else if (STRNEQ(mig->lockDriver,
                          virLockManagerPluginGetName(driver->lockManager))) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR,
                            _("Source host lock driver %s different from target %s"),
                            mig->lockDriver,
                            virLockManagerPluginGetName(driver->lockManager));
            goto error;
        }
    }

689 690 691 692 693 694
    return mig;

error:
    qemuMigrationCookieFree(mig);
    return NULL;
}
695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715

bool
qemuMigrationIsAllowed(virDomainDefPtr def)
{
    if (def->nhostdevs > 0) {
        qemuReportError(VIR_ERR_OPERATION_INVALID,
            "%s", _("Domain with assigned host devices cannot be migrated"));
        return false;
    }

    return true;
}

/** qemuMigrationSetOffline
 * Pause domain for non-live migration.
 */
int
qemuMigrationSetOffline(struct qemud_driver *driver,
                        virDomainObjPtr vm)
{
    int ret;
716
    VIR_DEBUG("driver=%p vm=%p", driver, vm);
J
Jiri Denemark 已提交
717
    ret = qemuProcessStopCPUs(driver, vm, VIR_DOMAIN_PAUSED_MIGRATION);
718 719 720 721 722 723 724 725 726 727 728 729 730 731
    if (ret == 0) {
        virDomainEventPtr event;

        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_SUSPENDED,
                                         VIR_DOMAIN_EVENT_SUSPENDED_MIGRATED);
        if (event)
            qemuDomainEventQueue(driver, event);
    }

    return ret;
}


732 733 734
static int
qemuMigrationProcessJobSignals(struct qemud_driver *driver,
                               virDomainObjPtr vm,
735 736
                               const char *job,
                               bool cleanup)
737 738 739 740 741 742 743
{
    qemuDomainObjPrivatePtr priv = vm->privateData;
    int ret = -1;

    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, _("%s: %s"),
                        job, _("guest unexpectedly quit"));
744 745
        if (cleanup)
            priv->jobSignals = 0;
746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
        return -1;
    }

    if (priv->jobSignals & QEMU_JOB_SIGNAL_CANCEL) {
        priv->jobSignals ^= QEMU_JOB_SIGNAL_CANCEL;
        VIR_DEBUG("Cancelling job at client request");
        qemuDomainObjEnterMonitorWithDriver(driver, vm);
        ret = qemuMonitorMigrateCancel(priv->mon);
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        if (ret < 0) {
            VIR_WARN("Unable to cancel job");
        }
    } else if (priv->jobSignals & QEMU_JOB_SIGNAL_SUSPEND) {
        priv->jobSignals ^= QEMU_JOB_SIGNAL_SUSPEND;
        VIR_DEBUG("Pausing domain for non-live migration");
        if (qemuMigrationSetOffline(driver, vm) < 0)
            VIR_WARN("Unable to pause domain");
    } else if (priv->jobSignals & QEMU_JOB_SIGNAL_MIGRATE_DOWNTIME) {
        unsigned long long ms = priv->jobSignalsData.migrateDowntime;

        priv->jobSignals ^= QEMU_JOB_SIGNAL_MIGRATE_DOWNTIME;
        priv->jobSignalsData.migrateDowntime = 0;
        VIR_DEBUG("Setting migration downtime to %llums", ms);
        qemuDomainObjEnterMonitorWithDriver(driver, vm);
        ret = qemuMonitorSetMigrationDowntime(priv->mon, ms);
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        if (ret < 0)
            VIR_WARN("Unable to set migration downtime");
    } else if (priv->jobSignals & QEMU_JOB_SIGNAL_MIGRATE_SPEED) {
        unsigned long bandwidth = priv->jobSignalsData.migrateBandwidth;

        priv->jobSignals ^= QEMU_JOB_SIGNAL_MIGRATE_SPEED;
        priv->jobSignalsData.migrateBandwidth = 0;
        VIR_DEBUG("Setting migration bandwidth to %luMbs", bandwidth);
        qemuDomainObjEnterMonitorWithDriver(driver, vm);
        ret = qemuMonitorSetMigrationSpeed(priv->mon, bandwidth);
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        if (ret < 0)
            VIR_WARN("Unable to set migration speed");
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812
    } else if (priv->jobSignals & QEMU_JOB_SIGNAL_BLKSTAT) {
        qemuDomainObjEnterMonitorWithDriver(driver, vm);
        ret = qemuMonitorGetBlockStatsInfo(priv->mon,
                              priv->jobSignalsData.statDevName,
                              &priv->jobSignalsData.blockStat->rd_req,
                              &priv->jobSignalsData.blockStat->rd_bytes,
                              &priv->jobSignalsData.blockStat->wr_req,
                              &priv->jobSignalsData.blockStat->wr_bytes,
                              &priv->jobSignalsData.blockStat->errs);
        qemuDomainObjExitMonitorWithDriver(driver, vm);

        *priv->jobSignalsData.statRetCode = ret;
        priv->jobSignals ^= QEMU_JOB_SIGNAL_BLKSTAT;

        if (ret < 0)
            VIR_WARN("Unable to get block statistics");
    } else if (priv->jobSignals & QEMU_JOB_SIGNAL_BLKINFO) {
        qemuDomainObjEnterMonitorWithDriver(driver, vm);
        ret = qemuMonitorGetBlockExtent(priv->mon,
                           priv->jobSignalsData.infoDevName,
                           &priv->jobSignalsData.blockInfo->allocation);
        qemuDomainObjExitMonitorWithDriver(driver, vm);

        *priv->jobSignalsData.infoRetCode = ret;
        priv->jobSignals ^= QEMU_JOB_SIGNAL_BLKINFO;

        if (ret < 0)
            VIR_WARN("Unable to get block information");
813 814 815 816 817 818 819 820 821 822 823 824
    } else {
        ret = 0;
    }

    return ret;
}


static int
qemuMigrationUpdateJobStatus(struct qemud_driver *driver,
                             virDomainObjPtr vm,
                             const char *job)
825
{
826
    qemuDomainObjPrivatePtr priv = vm->privateData;
827 828 829 830 831
    int ret = -1;
    int status;
    unsigned long long memProcessed;
    unsigned long long memRemaining;
    unsigned long long memTotal;
832 833 834 835 836 837 838 839 840 841 842 843 844 845 846

    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, _("%s: %s"),
                        job, _("guest unexpectedly quit"));
        return -1;
    }

    qemuDomainObjEnterMonitorWithDriver(driver, vm);
    ret = qemuMonitorGetMigrationStatus(priv->mon,
                                        &status,
                                        &memProcessed,
                                        &memRemaining,
                                        &memTotal);
    qemuDomainObjExitMonitorWithDriver(driver, vm);

J
Jiri Denemark 已提交
847
    if (ret < 0 || virTimeMs(&priv->jobInfo.timeElapsed) < 0) {
848 849 850
        priv->jobInfo.type = VIR_DOMAIN_JOB_FAILED;
        return -1;
    }
J
Jiri Denemark 已提交
851
    priv->jobInfo.timeElapsed -= priv->jobStart;
852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896

    switch (status) {
    case QEMU_MONITOR_MIGRATION_STATUS_INACTIVE:
        priv->jobInfo.type = VIR_DOMAIN_JOB_NONE;
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        _("%s: %s"), job, _("is not active"));
        break;

    case QEMU_MONITOR_MIGRATION_STATUS_ACTIVE:
        priv->jobInfo.dataTotal = memTotal;
        priv->jobInfo.dataRemaining = memRemaining;
        priv->jobInfo.dataProcessed = memProcessed;

        priv->jobInfo.memTotal = memTotal;
        priv->jobInfo.memRemaining = memRemaining;
        priv->jobInfo.memProcessed = memProcessed;

        ret = 0;
        break;

    case QEMU_MONITOR_MIGRATION_STATUS_COMPLETED:
        priv->jobInfo.type = VIR_DOMAIN_JOB_COMPLETED;
        ret = 0;
        break;

    case QEMU_MONITOR_MIGRATION_STATUS_ERROR:
        priv->jobInfo.type = VIR_DOMAIN_JOB_FAILED;
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        _("%s: %s"), job, _("unexpectedly failed"));
        break;

    case QEMU_MONITOR_MIGRATION_STATUS_CANCELLED:
        priv->jobInfo.type = VIR_DOMAIN_JOB_CANCELLED;
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        _("%s: %s"), job, _("canceled by client"));
        break;
    }

    return ret;
}


int
qemuMigrationWaitForCompletion(struct qemud_driver *driver, virDomainObjPtr vm)
{
897
    qemuDomainObjPrivatePtr priv = vm->privateData;
898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
    const char *job;

    switch (priv->jobActive) {
    case QEMU_JOB_MIGRATION_OUT:
        job = _("migration job");
        break;
    case QEMU_JOB_SAVE:
        job = _("domain save job");
        break;
    case QEMU_JOB_DUMP:
        job = _("domain core dump job");
        break;
    default:
        job = _("job");
    }
913 914 915 916 917 918

    priv->jobInfo.type = VIR_DOMAIN_JOB_UNBOUNDED;

    while (priv->jobInfo.type == VIR_DOMAIN_JOB_UNBOUNDED) {
        /* Poll every 50ms for progress & to allow cancellation */
        struct timespec ts = { .tv_sec = 0, .tv_nsec = 50 * 1000 * 1000ull };
919 920 921
        while (priv->jobSignals) {
            if (qemuMigrationProcessJobSignals(driver, vm, job, false) < 0)
                goto cleanup;
922 923
        }

924
        virCondSignal(&priv->signalCond);
925

926
        if (qemuMigrationUpdateJobStatus(driver, vm, job) < 0)
927 928 929 930 931 932 933 934 935 936 937 938 939
            goto cleanup;


        virDomainObjUnlock(vm);
        qemuDriverUnlock(driver);

        nanosleep(&ts, NULL);

        qemuDriverLock(driver);
        virDomainObjLock(vm);
    }

cleanup:
940 941 942 943 944
    while (priv->jobSignals) {
        qemuMigrationProcessJobSignals(driver, vm, job, true);
    }
    virCondBroadcast(&priv->signalCond);

945 946 947 948
    if (priv->jobInfo.type == VIR_DOMAIN_JOB_COMPLETED)
        return 0;
    else
        return -1;
949 950 951
}


952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974
static int
qemuDomainMigrateGraphicsRelocate(struct qemud_driver *driver,
                                  virDomainObjPtr vm,
                                  qemuMigrationCookiePtr cookie)
{
    qemuDomainObjPrivatePtr priv = vm->privateData;
    int ret;

    if (!cookie)
        return 0;

    if (!cookie->graphics)
        return 0;

    /* QEMU doesn't support VNC relocation yet, so
     * skip it to avoid generating an error
     */
    if (cookie->graphics->type != VIR_DOMAIN_GRAPHICS_TYPE_SPICE)
        return 0;

    qemuDomainObjEnterMonitorWithDriver(driver, vm);
    ret = qemuMonitorGraphicsRelocate(priv->mon,
                                      cookie->graphics->type,
975
                                      cookie->remoteHostname,
976 977 978 979 980 981 982 983 984
                                      cookie->graphics->port,
                                      cookie->graphics->tlsPort,
                                      cookie->graphics->tlsSubject);
    qemuDomainObjExitMonitorWithDriver(driver, vm);

    return ret;
}


985 986
char *qemuMigrationBegin(struct qemud_driver *driver,
                         virDomainObjPtr vm,
987
                         const char *xmlin,
988 989 990 991 992
                         char **cookieout,
                         int *cookieoutlen)
{
    char *rv = NULL;
    qemuMigrationCookiePtr mig = NULL;
993
    virDomainDefPtr def = NULL;
994 995
    VIR_DEBUG("driver=%p, vm=%p, xmlin=%s, cookieout=%p, cookieoutlen=%p",
              driver, vm, NULLSTR(xmlin), cookieout, cookieoutlen);
996 997 998 999 1000 1001 1002

    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_OPERATION_INVALID,
                        "%s", _("domain is not running"));
        goto cleanup;
    }

1003 1004 1005 1006 1007 1008
    if (qemuProcessAutoDestroyActive(driver, vm)) {
        qemuReportError(VIR_ERR_OPERATION_INVALID,
                        "%s", _("domain is marked for auto destroy"));
        goto cleanup;
    }

1009 1010 1011
    if (!qemuMigrationIsAllowed(vm->def))
        goto cleanup;

1012
    if (!(mig = qemuMigrationEatCookie(driver, vm, NULL, 0, 0)))
1013 1014 1015 1016
        goto cleanup;

    if (qemuMigrationBakeCookie(mig, driver, vm,
                                cookieout, cookieoutlen,
1017
                                QEMU_MIGRATION_COOKIE_LOCKSTATE) < 0)
1018 1019
        goto cleanup;

1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035
    if (xmlin) {
        if (!(def = virDomainDefParseString(driver->caps, xmlin,
                                            VIR_DOMAIN_XML_INACTIVE)))
            goto cleanup;

        if (!virDomainDefCheckABIStability(def, vm->def))
            goto cleanup;

        rv = qemuDomainDefFormatXML(driver, def,
                                    VIR_DOMAIN_XML_SECURE |
                                    VIR_DOMAIN_XML_UPDATE_CPU);
    } else {
        rv = qemuDomainFormatXML(driver, vm,
                                 VIR_DOMAIN_XML_SECURE |
                                 VIR_DOMAIN_XML_UPDATE_CPU);
    }
1036 1037 1038 1039

cleanup:
    virDomainObjUnlock(vm);
    qemuMigrationCookieFree(mig);
1040
    virDomainDefFree(def);
1041 1042 1043 1044
    return rv;
}


1045 1046 1047 1048 1049 1050 1051 1052
/* Prepare is the first step, and it runs on the destination host.
 *
 * This version starts an empty VM listening on a localhost TCP port, and
 * sets up the corresponding virStream to handle the incoming data.
 */
int
qemuMigrationPrepareTunnel(struct qemud_driver *driver,
                           virConnectPtr dconn,
1053 1054 1055 1056
                           const char *cookiein,
                           int cookieinlen,
                           char **cookieout,
                           int *cookieoutlen,
1057 1058 1059 1060 1061 1062 1063 1064 1065
                           virStreamPtr st,
                           const char *dname,
                           const char *dom_xml)
{
    virDomainDefPtr def = NULL;
    virDomainObjPtr vm = NULL;
    virDomainEventPtr event = NULL;
    int ret = -1;
    int internalret;
1066
    int dataFD[2] = { -1, -1 };
1067
    qemuDomainObjPrivatePtr priv = NULL;
J
Jiri Denemark 已提交
1068
    unsigned long long now;
1069
    qemuMigrationCookiePtr mig = NULL;
J
Jiri Denemark 已提交
1070

1071 1072 1073 1074
    VIR_DEBUG("driver=%p, dconn=%p, cookiein=%s, cookieinlen=%d, "
              "cookieout=%p, cookieoutlen=%p, st=%p, dname=%s, dom_xml=%s",
              driver, dconn, NULLSTR(cookiein), cookieinlen,
              cookieout, cookieoutlen, st, NULLSTR(dname), dom_xml);
1075

J
Jiri Denemark 已提交
1076
    if (virTimeMs(&now) < 0)
1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106
        return -1;

    /* Parse the domain XML. */
    if (!(def = virDomainDefParseString(driver->caps, dom_xml,
                                        VIR_DOMAIN_XML_INACTIVE)))
        goto cleanup;

    if (!qemuMigrationIsAllowed(def))
        goto cleanup;

    /* Target domain name, maybe renamed. */
    if (dname) {
        VIR_FREE(def->name);
        def->name = strdup(dname);
        if (def->name == NULL)
            goto cleanup;
    }

    if (virDomainObjIsDuplicate(&driver->domains, def, 1) < 0)
        goto cleanup;

    if (!(vm = virDomainAssignDef(driver->caps,
                                  &driver->domains,
                                  def, true))) {
        /* virDomainAssignDef already set the error */
        goto cleanup;
    }
    def = NULL;
    priv = vm->privateData;

1107 1108
    if (!(mig = qemuMigrationEatCookie(driver, vm, cookiein, cookieinlen,
                                       QEMU_MIGRATION_COOKIE_LOCKSTATE)))
1109 1110
        goto cleanup;

1111 1112
    if (qemuDomainObjBeginJobWithDriver(driver, vm) < 0)
        goto cleanup;
1113
    priv->jobActive = QEMU_JOB_MIGRATION_IN;
1114 1115 1116 1117

    /* Domain starts inactive, even if the domain XML had an id field. */
    vm->def->id = -1;

1118
    if (pipe(dataFD) < 0 ||
1119
        virSetCloseExec(dataFD[1]) < 0) {
1120 1121
        virReportSystemError(errno, "%s",
                             _("cannot create pipe for tunnelled migration"));
1122 1123 1124 1125
        goto endjob;
    }

    /* Start the QEMU daemon, with the same command-line arguments plus
1126
     * -incoming stdio (which qemu_command might convert to exec:cat or fd:n)
1127
     */
1128
    internalret = qemuProcessStart(dconn, driver, vm, "stdio", true,
1129
                                   true, dataFD[0], NULL,
1130
                                   VIR_VM_OP_MIGRATE_IN_START);
1131
    if (internalret < 0) {
1132
        qemuAuditDomainStart(vm, "migrated", false);
1133 1134 1135 1136 1137 1138 1139 1140 1141 1142
        /* Note that we don't set an error here because qemuProcessStart
         * should have already done that.
         */
        if (!vm->persistent) {
            virDomainRemoveInactive(&driver->domains, vm);
            vm = NULL;
        }
        goto endjob;
    }

1143
    if (virFDStreamOpen(st, dataFD[1]) < 0) {
1144
        qemuAuditDomainStart(vm, "migrated", false);
J
Jiri Denemark 已提交
1145
        qemuProcessStop(driver, vm, 0, VIR_DOMAIN_SHUTOFF_FAILED);
1146 1147 1148 1149 1150
        if (!vm->persistent) {
            if (qemuDomainObjEndJob(vm) > 0)
                virDomainRemoveInactive(&driver->domains, vm);
            vm = NULL;
        }
1151 1152
        virReportSystemError(errno, "%s",
                             _("cannot pass pipe for tunnelled migration"));
1153 1154
        goto endjob;
    }
1155
    dataFD[1] = -1; /* 'st' owns the FD now & will close it */
1156

1157
    qemuAuditDomainStart(vm, "migrated", true);
1158 1159 1160 1161

    event = virDomainEventNewFromObj(vm,
                                     VIR_DOMAIN_EVENT_STARTED,
                                     VIR_DOMAIN_EVENT_STARTED_MIGRATED);
1162

1163 1164
    if (qemuMigrationBakeCookie(mig, driver, vm, cookieout, cookieoutlen,
                                QEMU_MIGRATION_COOKIE_GRAPHICS) < 0) {
1165 1166 1167 1168 1169 1170 1171
        /* We could tear down the whole guest here, but
         * cookie data is (so far) non-critical, so that
         * seems a little harsh. We'll just warn for now.
         */
        VIR_WARN("Unable to encode migration cookie");
    }

1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187
    ret = 0;

endjob:
    if (vm &&
        qemuDomainObjEndJob(vm) == 0)
        vm = NULL;

    /* We set a fake job active which is held across
     * API calls until the finish() call. This prevents
     * any other APIs being invoked while incoming
     * migration is taking place
     */
    if (vm &&
        virDomainObjIsActive(vm)) {
        priv->jobActive = QEMU_JOB_MIGRATION_IN;
        priv->jobInfo.type = VIR_DOMAIN_JOB_UNBOUNDED;
J
Jiri Denemark 已提交
1188
        priv->jobStart = now;
1189 1190 1191 1192
    }

cleanup:
    virDomainDefFree(def);
1193 1194
    VIR_FORCE_CLOSE(dataFD[0]);
    VIR_FORCE_CLOSE(dataFD[1]);
1195 1196 1197 1198
    if (vm)
        virDomainObjUnlock(vm);
    if (event)
        qemuDomainEventQueue(driver, event);
1199
    qemuMigrationCookieFree(mig);
1200 1201 1202 1203 1204 1205 1206
    return ret;
}


int
qemuMigrationPrepareDirect(struct qemud_driver *driver,
                           virConnectPtr dconn,
1207 1208 1209 1210
                           const char *cookiein,
                           int cookieinlen,
                           char **cookieout,
                           int *cookieoutlen,
1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
                           const char *uri_in,
                           char **uri_out,
                           const char *dname,
                           const char *dom_xml)
{
    static int port = 0;
    virDomainDefPtr def = NULL;
    virDomainObjPtr vm = NULL;
    int this_port;
    char *hostname = NULL;
    char migrateFrom [64];
    const char *p;
    virDomainEventPtr event = NULL;
    int ret = -1;
    int internalret;
    qemuDomainObjPrivatePtr priv = NULL;
J
Jiri Denemark 已提交
1227
    unsigned long long now;
1228
    qemuMigrationCookiePtr mig = NULL;
J
Jiri Denemark 已提交
1229

1230 1231 1232 1233 1234 1235
    VIR_DEBUG("driver=%p, dconn=%p, cookiein=%s, cookieinlen=%d, "
              "cookieout=%p, cookieoutlen=%p, uri_in=%s, uri_out=%p, "
              "dname=%s, dom_xml=%s",
              driver, dconn, NULLSTR(cookiein), cookieinlen,
              cookieout, cookieoutlen, NULLSTR(uri_in), uri_out,
              NULLSTR(dname), dom_xml);
1236

J
Jiri Denemark 已提交
1237
    if (virTimeMs(&now) < 0)
1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
        return -1;

    /* The URI passed in may be NULL or a string "tcp://somehostname:port".
     *
     * If the URI passed in is NULL then we allocate a port number
     * from our pool of port numbers and return a URI of
     * "tcp://ourhostname:port".
     *
     * If the URI passed in is not NULL then we try to parse out the
     * port number and use that (note that the hostname is assumed
     * to be a correct hostname which refers to the target machine).
     */
    if (uri_in == NULL) {
        this_port = QEMUD_MIGRATION_FIRST_PORT + port++;
        if (port == QEMUD_MIGRATION_NUM_PORTS) port = 0;

        /* Get hostname */
        if ((hostname = virGetHostname(NULL)) == NULL)
            goto cleanup;

        if (STRPREFIX(hostname, "localhost")) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                            _("hostname on destination resolved to localhost, but migration requires an FQDN"));
            goto cleanup;
        }

        /* XXX this really should have been a properly well-formed
         * URI, but we can't add in tcp:// now without breaking
         * compatability with old targets. We at least make the
         * new targets accept both syntaxes though.
         */
        /* Caller frees */
        internalret = virAsprintf(uri_out, "tcp:%s:%d", hostname, this_port);
        if (internalret < 0) {
            virReportOOMError();
            goto cleanup;
        }
    } else {
        /* Check the URI starts with "tcp:".  We will escape the
         * URI when passing it to the qemu monitor, so bad
         * characters in hostname part don't matter.
         */
        if (!STRPREFIX (uri_in, "tcp:")) {
            qemuReportError (VIR_ERR_INVALID_ARG,
                             "%s", _("only tcp URIs are supported for KVM/QEMU migrations"));
            goto cleanup;
        }

        /* Get the port number. */
        p = strrchr (uri_in, ':');
        if (p == strchr(uri_in, ':')) {
            /* Generate a port */
            this_port = QEMUD_MIGRATION_FIRST_PORT + port++;
            if (port == QEMUD_MIGRATION_NUM_PORTS)
                port = 0;

            /* Caller frees */
            if (virAsprintf(uri_out, "%s:%d", uri_in, this_port) < 0) {
                virReportOOMError();
                goto cleanup;
            }

        } else {
            p++; /* definitely has a ':' in it, see above */
            this_port = virParseNumber (&p);
            if (this_port == -1 || p-uri_in != strlen (uri_in)) {
                qemuReportError(VIR_ERR_INVALID_ARG,
                                "%s", _("URI ended with incorrect ':port'"));
                goto cleanup;
            }
        }
    }

    if (*uri_out)
        VIR_DEBUG("Generated uri_out=%s", *uri_out);

    /* Parse the domain XML. */
    if (!(def = virDomainDefParseString(driver->caps, dom_xml,
                                        VIR_DOMAIN_XML_INACTIVE)))
        goto cleanup;

    if (!qemuMigrationIsAllowed(def))
        goto cleanup;

    /* Target domain name, maybe renamed. */
    if (dname) {
        VIR_FREE(def->name);
        def->name = strdup(dname);
        if (def->name == NULL)
            goto cleanup;
    }

    if (virDomainObjIsDuplicate(&driver->domains, def, 1) < 0)
        goto cleanup;

    if (!(vm = virDomainAssignDef(driver->caps,
                                  &driver->domains,
                                  def, true))) {
        /* virDomainAssignDef already set the error */
        goto cleanup;
    }
    def = NULL;
    priv = vm->privateData;

1342 1343
    if (!(mig = qemuMigrationEatCookie(driver, vm, cookiein, cookieinlen,
                                       QEMU_MIGRATION_COOKIE_LOCKSTATE)))
1344 1345
        goto cleanup;

1346 1347
    if (qemuDomainObjBeginJobWithDriver(driver, vm) < 0)
        goto cleanup;
1348
    priv->jobActive = QEMU_JOB_MIGRATION_IN;
1349 1350 1351 1352 1353 1354 1355 1356

    /* Domain starts inactive, even if the domain XML had an id field. */
    vm->def->id = -1;

    /* Start the QEMU daemon, with the same command-line arguments plus
     * -incoming tcp:0.0.0.0:port
     */
    snprintf (migrateFrom, sizeof (migrateFrom), "tcp:0.0.0.0:%d", this_port);
1357
    if (qemuProcessStart(dconn, driver, vm, migrateFrom, true, true,
1358
                         -1, NULL, VIR_VM_OP_MIGRATE_IN_START) < 0) {
1359
        qemuAuditDomainStart(vm, "migrated", false);
1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370
        /* Note that we don't set an error here because qemuProcessStart
         * should have already done that.
         */
        if (!vm->persistent) {
            if (qemuDomainObjEndJob(vm) > 0)
                virDomainRemoveInactive(&driver->domains, vm);
            vm = NULL;
        }
        goto endjob;
    }

1371 1372 1373 1374 1375 1376 1377 1378 1379
    if (mig->lockState) {
        VIR_DEBUG("Received lockstate %s", mig->lockState);
        VIR_FREE(priv->lockState);
        priv->lockState = mig->lockState;
        mig->lockState = NULL;
    } else {
        VIR_DEBUG("Received no lockstate");
    }

1380 1381
    if (qemuMigrationBakeCookie(mig, driver, vm, cookieout, cookieoutlen,
                                QEMU_MIGRATION_COOKIE_GRAPHICS) < 0) {
1382 1383 1384 1385 1386 1387 1388
        /* We could tear down the whole guest here, but
         * cookie data is (so far) non-critical, so that
         * seems a little harsh. We'll just warn for now.
         */
        VIR_WARN("Unable to encode migration cookie");
    }

1389
    qemuAuditDomainStart(vm, "migrated", true);
1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408
    event = virDomainEventNewFromObj(vm,
                                     VIR_DOMAIN_EVENT_STARTED,
                                     VIR_DOMAIN_EVENT_STARTED_MIGRATED);
    ret = 0;

endjob:
    if (vm &&
        qemuDomainObjEndJob(vm) == 0)
        vm = NULL;

    /* We set a fake job active which is held across
     * API calls until the finish() call. This prevents
     * any other APIs being invoked while incoming
     * migration is taking place
     */
    if (vm &&
        virDomainObjIsActive(vm)) {
        priv->jobActive = QEMU_JOB_MIGRATION_IN;
        priv->jobInfo.type = VIR_DOMAIN_JOB_UNBOUNDED;
J
Jiri Denemark 已提交
1409
        priv->jobStart = now;
1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420
    }

cleanup:
    VIR_FREE(hostname);
    virDomainDefFree(def);
    if (ret != 0)
        VIR_FREE(*uri_out);
    if (vm)
        virDomainObjUnlock(vm);
    if (event)
        qemuDomainEventQueue(driver, event);
1421
    qemuMigrationCookieFree(mig);
1422 1423 1424 1425 1426 1427 1428 1429 1430 1431
    return ret;
}


/* Perform migration using QEMU's native TCP migrate support,
 * not encrypted obviously
 */
static int doNativeMigrate(struct qemud_driver *driver,
                           virDomainObjPtr vm,
                           const char *uri,
1432 1433 1434 1435
                           const char *cookiein,
                           int cookieinlen,
                           char **cookieout,
                           int *cookieoutlen,
1436 1437 1438 1439 1440 1441 1442 1443
                           unsigned int flags,
                           const char *dname ATTRIBUTE_UNUSED,
                           unsigned long resource)
{
    int ret = -1;
    xmlURIPtr uribits = NULL;
    qemuDomainObjPrivatePtr priv = vm->privateData;
    unsigned int background_flags = QEMU_MONITOR_MIGRATE_BACKGROUND;
1444
    qemuMigrationCookiePtr mig = NULL;
1445 1446 1447 1448
    VIR_DEBUG("driver=%p, vm=%p, uri=%s, cookiein=%s, cookieinlen=%d, "
              "cookieout=%p, cookieoutlen=%p, flags=%u, dname=%s, resource=%lu",
              driver, vm, uri, NULLSTR(cookiein), cookieinlen,
              cookieout, cookieoutlen, flags, NULLSTR(dname), resource);
1449

1450 1451 1452 1453 1454 1455 1456 1457 1458
    if (virLockManagerPluginUsesState(driver->lockManager) &&
        !cookieout) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Migration with lock driver %s requires cookie support"),
                        virLockManagerPluginGetName(driver->lockManager));
        return -1;
    }

    if (!(mig = qemuMigrationEatCookie(driver, vm, cookiein, cookieinlen,
1459
                                       QEMU_MIGRATION_COOKIE_GRAPHICS)))
1460
        goto cleanup;
1461

1462 1463 1464
    if (qemuDomainMigrateGraphicsRelocate(driver, vm, mig) < 0)
        VIR_WARN("unable to provide data for graphics client relocation");

1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483
    /* Issue the migrate command. */
    if (STRPREFIX(uri, "tcp:") && !STRPREFIX(uri, "tcp://")) {
        /* HACK: source host generates bogus URIs, so fix them up */
        char *tmpuri;
        if (virAsprintf(&tmpuri, "tcp://%s", uri + strlen("tcp:")) < 0) {
            virReportOOMError();
            goto cleanup;
        }
        uribits = xmlParseURI(tmpuri);
        VIR_FREE(tmpuri);
    } else {
        uribits = xmlParseURI(uri);
    }
    if (!uribits) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("cannot parse URI %s"), uri);
        goto cleanup;
    }

1484 1485 1486 1487 1488 1489 1490
    /* Before EnterMonitor, since qemuProcessStopCPUs already does that */
    if (!(flags & VIR_MIGRATE_LIVE) &&
        virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
        if (qemuMigrationSetOffline(driver, vm) < 0)
            goto cleanup;
    }

1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513
    qemuDomainObjEnterMonitorWithDriver(driver, vm);
    if (resource > 0 &&
        qemuMonitorSetMigrationSpeed(priv->mon, resource) < 0) {
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        goto cleanup;
    }

    if (flags & VIR_MIGRATE_NON_SHARED_DISK)
        background_flags |= QEMU_MONITOR_MIGRATE_NON_SHARED_DISK;

    if (flags & VIR_MIGRATE_NON_SHARED_INC)
        background_flags |= QEMU_MONITOR_MIGRATE_NON_SHARED_INC;

    if (qemuMonitorMigrateToHost(priv->mon, background_flags, uribits->server,
                                 uribits->port) < 0) {
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        goto cleanup;
    }
    qemuDomainObjExitMonitorWithDriver(driver, vm);

    if (qemuMigrationWaitForCompletion(driver, vm) < 0)
        goto cleanup;

1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
    /* When migration completed, QEMU will have paused the
     * CPUs for us, but unless we're using the JSON monitor
     * we won't have been notified of this, so might still
     * think we're running. For v2 protocol this doesn't
     * matter because we'll kill the VM soon, but for v3
     * this is important because we stay paused until the
     * confirm3 step, but need to release the lock state
     */
    if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
        if (qemuMigrationSetOffline(driver, vm) < 0)
            goto cleanup;
    }

1527 1528 1529
    if (qemuMigrationBakeCookie(mig, driver, vm, cookieout, cookieoutlen, 0) < 0)
        VIR_WARN("Unable to encode migration cookie");

1530 1531 1532
    ret = 0;

cleanup:
1533
    qemuMigrationCookieFree(mig);
1534 1535 1536 1537 1538 1539 1540
    xmlFreeURI(uribits);
    return ret;
}


#define TUNNEL_SEND_BUF_SIZE 65536

1541 1542 1543 1544 1545 1546 1547 1548 1549 1550
typedef struct _qemuMigrationIOThread qemuMigrationIOThread;
typedef qemuMigrationIOThread *qemuMigrationIOThreadPtr;
struct _qemuMigrationIOThread {
    virThread thread;
    virStreamPtr st;
    int sock;
    virError err;
};

static void qemuMigrationIOFunc(void *arg)
1551
{
1552
    qemuMigrationIOThreadPtr data = arg;
1553 1554 1555 1556 1557
    char *buffer;
    int nbytes = TUNNEL_SEND_BUF_SIZE;

    if (VIR_ALLOC_N(buffer, TUNNEL_SEND_BUF_SIZE) < 0) {
        virReportOOMError();
1558 1559
        virStreamAbort(data->st);
        goto error;
1560 1561 1562
    }

    for (;;) {
1563
        nbytes = saferead(data->sock, buffer, TUNNEL_SEND_BUF_SIZE);
1564 1565 1566
        if (nbytes < 0) {
            virReportSystemError(errno, "%s",
                                 _("tunnelled migration failed to read from qemu"));
1567
            virStreamAbort(data->st);
1568
            VIR_FREE(buffer);
1569
            goto error;
1570 1571 1572 1573 1574
        }
        else if (nbytes == 0)
            /* EOF; get out of here */
            break;

1575
        if (virStreamSend(data->st, buffer, nbytes) < 0) {
1576
            VIR_FREE(buffer);
1577
            goto error;
1578 1579 1580 1581 1582
        }
    }

    VIR_FREE(buffer);

1583 1584
    if (virStreamFinish(data->st) < 0)
        goto error;
1585

1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637
    return;

error:
    virCopyLastError(&data->err);
    virResetLastError();
}


static qemuMigrationIOThreadPtr
qemuMigrationStartTunnel(virStreamPtr st,
                         int sock)
{
    qemuMigrationIOThreadPtr io;

    if (VIR_ALLOC(io) < 0) {
        virReportOOMError();
        return NULL;
    }

    io->st = st;
    io->sock = sock;

    if (virThreadCreate(&io->thread, true,
                        qemuMigrationIOFunc,
                        io) < 0) {
        virReportSystemError(errno, "%s",
                             _("Unable to create migration thread"));
        VIR_FREE(io);
        return NULL;
    }

    return io;
}

static int
qemuMigrationStopTunnel(qemuMigrationIOThreadPtr io)
{
    int rv = -1;
    virThreadJoin(&io->thread);

    /* Forward error from the IO thread, to this thread */
    if (io->err.code != VIR_ERR_OK) {
        virSetError(&io->err);
        virResetError(&io->err);
        goto cleanup;
    }

    rv = 0;

cleanup:
    VIR_FREE(io);
    return rv;
1638 1639
}

1640

1641 1642
static int doTunnelMigrate(struct qemud_driver *driver,
                           virDomainObjPtr vm,
1643
                           virStreamPtr st,
1644 1645 1646 1647
                           const char *cookiein,
                           int cookieinlen,
                           char **cookieout,
                           int *cookieoutlen,
1648
                           unsigned long flags,
1649
                           unsigned long resource)
1650 1651 1652 1653 1654 1655 1656 1657
{
    qemuDomainObjPrivatePtr priv = vm->privateData;
    int client_sock = -1;
    int qemu_sock = -1;
    struct sockaddr_un sa_qemu, sa_client;
    socklen_t addrlen;
    int status;
    unsigned long long transferred, remaining, total;
1658
    char *unixfile = NULL;
1659
    unsigned int background_flags = QEMU_MONITOR_MIGRATE_BACKGROUND;
1660
    int ret = -1;
1661
    qemuMigrationCookiePtr mig = NULL;
1662
    qemuMigrationIOThreadPtr iothread = NULL;
1663 1664 1665 1666
    VIR_DEBUG("driver=%p, vm=%p, st=%p, cookiein=%s, cookieinlen=%d, "
              "cookieout=%p, cookieoutlen=%p, flags=%lu, resource=%lu",
              driver, vm, st, NULLSTR(cookiein), cookieinlen,
              cookieout, cookieoutlen, flags, resource);
1667

1668 1669 1670 1671 1672 1673 1674 1675
    if (virLockManagerPluginUsesState(driver->lockManager) &&
        !cookieout) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Migration with lock driver %s requires cookie support"),
                        virLockManagerPluginGetName(driver->lockManager));
        return -1;
    }

1676 1677 1678 1679 1680 1681
    if (!qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_UNIX) &&
        !qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_EXEC)) {
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        "%s", _("Source qemu is too old to support tunnelled migration"));
        goto cleanup;
    }
1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734


    if (virAsprintf(&unixfile, "%s/qemu.tunnelmigrate.src.%s",
                    driver->libDir, vm->def->name) < 0) {
        virReportOOMError();
        goto cleanup;
    }

    qemu_sock = socket(AF_UNIX, SOCK_STREAM, 0);
    if (qemu_sock < 0) {
        virReportSystemError(errno, "%s",
                             _("cannot open tunnelled migration socket"));
        goto cleanup;
    }
    memset(&sa_qemu, 0, sizeof(sa_qemu));
    sa_qemu.sun_family = AF_UNIX;
    if (virStrcpy(sa_qemu.sun_path, unixfile,
                  sizeof(sa_qemu.sun_path)) == NULL) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("Unix socket '%s' too big for destination"),
                        unixfile);
        goto cleanup;
    }
    unlink(unixfile);
    if (bind(qemu_sock, (struct sockaddr *)&sa_qemu, sizeof(sa_qemu)) < 0) {
        virReportSystemError(errno,
                             _("Cannot bind to unix socket '%s' for tunnelled migration"),
                             unixfile);
        goto cleanup;
    }
    if (listen(qemu_sock, 1) < 0) {
        virReportSystemError(errno,
                             _("Cannot listen on unix socket '%s' for tunnelled migration"),
                             unixfile);
        goto cleanup;
    }

    if (chown(unixfile, driver->user, driver->group) < 0) {
        virReportSystemError(errno,
                             _("Cannot change unix socket '%s' owner"),
                             unixfile);
        goto cleanup;
    }

    /* the domain may have shutdown or crashed while we had the locks dropped
     * in qemuDomainObjEnterRemoteWithDriver, so check again
     */
    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("guest unexpectedly quit"));
        goto cleanup;
    }

1735
    if (!(mig = qemuMigrationEatCookie(driver, vm, cookiein, cookieinlen,
1736 1737 1738 1739 1740 1741
                                       QEMU_MIGRATION_COOKIE_GRAPHICS)))
        goto cleanup;

    if (qemuDomainMigrateGraphicsRelocate(driver, vm, mig) < 0)
        VIR_WARN("unable to provide data for graphics client relocation");

1742
    /*   3. start migration on source */
1743 1744 1745 1746 1747 1748 1749
    /* Before EnterMonitor, since qemuProcessStopCPUs already does that */
    if (!(flags & VIR_MIGRATE_LIVE) &&
        virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
        if (qemuMigrationSetOffline(driver, vm) < 0)
            goto cleanup;
    }

1750
    qemuDomainObjEnterMonitorWithDriver(driver, vm);
1751 1752 1753 1754 1755
    if (resource > 0 &&
        qemuMonitorSetMigrationSpeed(priv->mon, resource) < 0) {
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        goto cleanup;
    }
1756

1757 1758 1759 1760
    if (flags & VIR_MIGRATE_NON_SHARED_DISK)
        background_flags |= QEMU_MONITOR_MIGRATE_NON_SHARED_DISK;
    if (flags & VIR_MIGRATE_NON_SHARED_INC)
        background_flags |= QEMU_MONITOR_MIGRATE_NON_SHARED_INC;
1761

1762
    if (qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_UNIX)) {
1763
        ret = qemuMonitorMigrateToUnix(priv->mon, background_flags,
1764
                                               unixfile);
1765
    } else if (qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_EXEC)) {
1766
        const char *args[] = { "nc", "-U", unixfile, NULL };
1767
        ret = qemuMonitorMigrateToCommand(priv->mon, QEMU_MONITOR_MIGRATE_BACKGROUND, args);
1768
    } else {
1769
        ret = -1;
1770 1771
    }
    qemuDomainObjExitMonitorWithDriver(driver, vm);
1772
    if (ret < 0) {
1773 1774
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        "%s", _("tunnelled migration monitor command failed"));
1775
        goto cleanup;
1776
    }
1777
    ret = -1;
1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816

    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("guest unexpectedly quit"));
        goto cleanup;
    }

    /* From this point onwards we *must* call cancel to abort the
     * migration on source if anything goes wrong */

    /* it is also possible that the migrate didn't fail initially, but
     * rather failed later on.  Check the output of "info migrate"
     */
    qemuDomainObjEnterMonitorWithDriver(driver, vm);
    if (qemuMonitorGetMigrationStatus(priv->mon,
                                      &status,
                                      &transferred,
                                      &remaining,
                                      &total) < 0) {
        qemuDomainObjExitMonitorWithDriver(driver, vm);
        goto cancel;
    }
    qemuDomainObjExitMonitorWithDriver(driver, vm);

    if (status == QEMU_MONITOR_MIGRATION_STATUS_ERROR) {
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        "%s",_("migrate failed"));
        goto cancel;
    }

    addrlen = sizeof(sa_client);
    while ((client_sock = accept(qemu_sock, (struct sockaddr *)&sa_client, &addrlen)) < 0) {
        if (errno == EAGAIN || errno == EINTR)
            continue;
        virReportSystemError(errno, "%s",
                             _("tunnelled migration failed to accept from qemu"));
        goto cancel;
    }

1817 1818 1819 1820 1821
    if (!(iothread = qemuMigrationStartTunnel(st, client_sock)))
        goto cancel;

    ret = qemuMigrationWaitForCompletion(driver, vm);

1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834
    /* When migration completed, QEMU will have paused the
     * CPUs for us, but unless we're using the JSON monitor
     * we won't have been notified of this, so might still
     * think we're running. For v2 protocol this doesn't
     * matter because we'll kill the VM soon, but for v3
     * this is important because we stay paused until the
     * confirm3 step, but need to release the lock state
     */
    if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
        if (qemuMigrationSetOffline(driver, vm) < 0)
            goto cleanup;
    }

1835 1836 1837 1838 1839
    /* Close now to ensure the IO thread quits & is joinable in next method */
    VIR_FORCE_CLOSE(client_sock);

    if (qemuMigrationStopTunnel(iothread) < 0)
        ret = -1;
1840

1841 1842 1843 1844
    if (ret == 0 &&
        qemuMigrationBakeCookie(mig, driver, vm, cookieout, cookieoutlen, 0) < 0)
        VIR_WARN("Unable to encode migration cookie");

1845
cancel:
1846
    if (ret != 0 && virDomainObjIsActive(vm)) {
1847 1848
        VIR_FORCE_CLOSE(client_sock);
        VIR_FORCE_CLOSE(qemu_sock);
1849 1850 1851 1852 1853
        qemuDomainObjEnterMonitorWithDriver(driver, vm);
        qemuMonitorMigrateCancel(priv->mon);
        qemuDomainObjExitMonitorWithDriver(driver, vm);
    }

1854
cleanup:
1855
    qemuMigrationCookieFree(mig);
1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866
    VIR_FORCE_CLOSE(client_sock);
    VIR_FORCE_CLOSE(qemu_sock);
    if (unixfile) {
        unlink(unixfile);
        VIR_FREE(unixfile);
    }

    return ret;
}


1867 1868 1869 1870 1871 1872
/* This is essentially a re-impl of virDomainMigrateVersion2
 * from libvirt.c, but running in source libvirtd context,
 * instead of client app context & also adding in tunnel
 * handling */
static int doPeer2PeerMigrate2(struct qemud_driver *driver,
                               virConnectPtr sconn,
1873 1874
                               virConnectPtr dconn,
                               virDomainObjPtr vm,
1875
                               const char *dconnuri,
1876 1877 1878
                               unsigned long flags,
                               const char *dname,
                               unsigned long resource)
1879 1880 1881
{
    virDomainPtr ddomain = NULL;
    char *uri_out = NULL;
1882
    char *cookie = NULL;
1883 1884 1885 1886 1887
    char *dom_xml = NULL;
    int cookielen = 0, ret;
    virErrorPtr orig_err = NULL;
    int cancelled;
    virStreamPtr st = NULL;
1888 1889 1890 1891
    VIR_DEBUG("driver=%p, sconn=%p, dconn=%p, vm=%p, dconnuri=%s, "
              "flags=%lu, dname=%s, resource=%lu",
              driver, sconn, dconn, vm, NULLSTR(dconnuri),
              flags, NULLSTR(dname), resource);
1892

1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927
    /* In version 2 of the protocol, the prepare step is slightly
     * different.  We fetch the domain XML of the source domain
     * and pass it to Prepare2.
     */
    if (!(dom_xml = qemuDomainFormatXML(driver, vm,
                                        VIR_DOMAIN_XML_SECURE |
                                        VIR_DOMAIN_XML_UPDATE_CPU)))
        return -1;

    if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_PAUSED)
        flags |= VIR_MIGRATE_PAUSED;

    VIR_DEBUG("Prepare2 %p", dconn);
    if (flags & VIR_MIGRATE_TUNNELLED) {
        /*
         * Tunnelled Migrate Version 2 does not support cookies
         * due to missing parameters in the prepareTunnel() API.
         */

        if (!(st = virStreamNew(dconn, 0)))
            goto cleanup;

        qemuDomainObjEnterRemoteWithDriver(driver, vm);
        ret = dconn->driver->domainMigratePrepareTunnel
            (dconn, st, flags, dname, resource, dom_xml);
        qemuDomainObjExitRemoteWithDriver(driver, vm);
    } else {
        qemuDomainObjEnterRemoteWithDriver(driver, vm);
        ret = dconn->driver->domainMigratePrepare2
            (dconn, &cookie, &cookielen, NULL, &uri_out,
             flags, dname, resource, dom_xml);
        qemuDomainObjExitRemoteWithDriver(driver, vm);
    }
    VIR_FREE(dom_xml);
    if (ret == -1)
1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938
        goto cleanup;

    /* the domain may have shutdown or crashed while we had the locks dropped
     * in qemuDomainObjEnterRemoteWithDriver, so check again
     */
    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("guest unexpectedly quit"));
        goto cleanup;
    }

1939 1940 1941
    if (!(flags & VIR_MIGRATE_TUNNELLED) &&
        (uri_out == NULL)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
1942
                        _("domainMigratePrepare2 did not set uri"));
1943 1944
        cancelled = 1;
        goto finish;
1945 1946
    }

1947 1948 1949 1950 1951
    /* Perform the migration.  The driver isn't supposed to return
     * until the migration is complete.
     */
    VIR_DEBUG("Perform %p", sconn);
    if (flags & VIR_MIGRATE_TUNNELLED)
1952 1953 1954
        ret = doTunnelMigrate(driver, vm, st,
                              NULL, 0, NULL, NULL,
                              flags, resource);
1955 1956 1957 1958 1959 1960 1961 1962 1963
    else
        ret = doNativeMigrate(driver, vm, uri_out,
                              cookie, cookielen,
                              NULL, NULL, /* No out cookie with v2 migration */
                              flags, dname, resource);

    /* Perform failed. Make sure Finish doesn't overwrite the error */
    if (ret < 0)
        orig_err = virSaveLastError();
1964

1965 1966 1967 1968
    /* If Perform returns < 0, then we need to cancel the VM
     * startup on the destination
     */
    cancelled = ret < 0 ? 1 : 0;
1969 1970

finish:
1971 1972 1973 1974
    /* In version 2 of the migration protocol, we pass the
     * status code from the sender to the destination host,
     * so it can do any cleanup if the migration failed.
     */
1975
    dname = dname ? dname : vm->def->name;
1976
    VIR_DEBUG("Finish2 %p ret=%d", dconn, ret);
1977 1978
    qemuDomainObjEnterRemoteWithDriver(driver, vm);
    ddomain = dconn->driver->domainMigrateFinish2
1979
        (dconn, dname, cookie, cookielen,
1980
         uri_out ? uri_out : dconnuri, flags, cancelled);
1981 1982
    qemuDomainObjExitRemoteWithDriver(driver, vm);

1983 1984
cleanup:
    if (ddomain) {
1985
        virUnrefDomain(ddomain);
1986 1987 1988 1989
        ret = 0;
    } else {
        ret = -1;
    }
1990

1991 1992 1993 1994 1995 1996 1997 1998
    if (st)
        virUnrefStream(st);

    if (orig_err) {
        virSetError(orig_err);
        virFreeError(orig_err);
    }
    VIR_FREE(uri_out);
1999
    VIR_FREE(cookie);
2000 2001

    return ret;
2002 2003 2004
}


2005 2006 2007 2008 2009 2010 2011 2012
/* This is essentially a re-impl of virDomainMigrateVersion3
 * from libvirt.c, but running in source libvirtd context,
 * instead of client app context & also adding in tunnel
 * handling */
static int doPeer2PeerMigrate3(struct qemud_driver *driver,
                               virConnectPtr sconn,
                               virConnectPtr dconn,
                               virDomainObjPtr vm,
2013
                               const char *xmlin,
2014
                               const char *dconnuri,
2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
                               const char *uri,
                               unsigned long flags,
                               const char *dname,
                               unsigned long resource)
{
    virDomainPtr ddomain = NULL;
    char *uri_out = NULL;
    char *cookiein = NULL;
    char *cookieout = NULL;
    char *dom_xml = NULL;
    int cookieinlen = 0;
    int cookieoutlen = 0;
    int ret = -1;
    virErrorPtr orig_err = NULL;
    int cancelled;
    virStreamPtr st = NULL;
2031 2032 2033 2034 2035
    VIR_DEBUG("driver=%p, sconn=%p, dconn=%p, vm=%p, xmlin=%s, "
              "dconnuri=%s, uri=%s, flags=%lu, dname=%s, resource=%lu",
              driver, sconn, dconn, vm, NULLSTR(xmlin),
              NULLSTR(dconnuri), NULLSTR(uri), flags,
              NULLSTR(dname), resource);
2036

2037
    dom_xml = qemuMigrationBegin(driver, vm, xmlin,
2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063
                                 &cookieout, &cookieoutlen);
    if (!dom_xml)
        goto cleanup;

    if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_PAUSED)
        flags |= VIR_MIGRATE_PAUSED;

    VIR_DEBUG("Prepare3 %p", dconn);
    cookiein = cookieout;
    cookieinlen = cookieoutlen;
    cookieout = NULL;
    cookieoutlen = 0;
    if (flags & VIR_MIGRATE_TUNNELLED) {
        if (!(st = virStreamNew(dconn, 0)))
            goto cleanup;

        qemuDomainObjEnterRemoteWithDriver(driver, vm);
        ret = dconn->driver->domainMigratePrepareTunnel3
            (dconn, st, cookiein, cookieinlen,
             &cookieout, &cookieoutlen,
             flags, dname, resource, dom_xml);
        qemuDomainObjExitRemoteWithDriver(driver, vm);
    } else {
        qemuDomainObjEnterRemoteWithDriver(driver, vm);
        ret = dconn->driver->domainMigratePrepare3
            (dconn, cookiein, cookieinlen, &cookieout, &cookieoutlen,
2064
             uri, &uri_out, flags, dname, resource, dom_xml);
2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083
        qemuDomainObjExitRemoteWithDriver(driver, vm);
    }
    VIR_FREE(dom_xml);
    if (ret == -1)
        goto cleanup;

    if (!(flags & VIR_MIGRATE_TUNNELLED) &&
        (uri_out == NULL)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR,
                        _("domainMigratePrepare3 did not set uri"));
        cancelled = 1;
        goto finish;
    }

    /* Perform the migration.  The driver isn't supposed to return
     * until the migration is complete. The src VM should remain
     * running, but in paused state until the destination can
     * confirm migration completion.
     */
2084
    VIR_DEBUG("Perform3 %p uri=%s uri_out=%s", sconn, uri, uri_out);
2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124
    VIR_FREE(cookiein);
    cookiein = cookieout;
    cookieinlen = cookieoutlen;
    cookieout = NULL;
    cookieoutlen = 0;
    if (flags & VIR_MIGRATE_TUNNELLED)
        ret = doTunnelMigrate(driver, vm, st,
                              cookiein, cookieinlen,
                              &cookieout, &cookieoutlen,
                              flags, resource);
    else
        ret = doNativeMigrate(driver, vm, uri_out,
                              cookiein, cookieinlen,
                              &cookieout, &cookieoutlen,
                              flags, dname, resource);

    /* Perform failed. Make sure Finish doesn't overwrite the error */
    if (ret < 0)
        orig_err = virSaveLastError();

    /* If Perform returns < 0, then we need to cancel the VM
     * startup on the destination
     */
    cancelled = ret < 0 ? 1 : 0;

finish:
    /*
     * The status code from the source is passed to the destination.
     * The dest can cleanup in the source indicated it failed to
     * send all migration data. Returns NULL for ddomain if
     * the dest was unable to complete migration.
     */
    VIR_DEBUG("Finish3 %p ret=%d", dconn, ret);
    VIR_FREE(cookiein);
    cookiein = cookieout;
    cookieinlen = cookieoutlen;
    cookieout = NULL;
    cookieoutlen = 0;
    dname = dname ? dname : vm->def->name;
    qemuDomainObjEnterRemoteWithDriver(driver, vm);
2125
    ddomain = dconn->driver->domainMigrateFinish3
2126
        (dconn, dname, cookiein, cookieinlen, &cookieout, &cookieoutlen,
2127
         dconnuri, uri_out ? uri_out : uri, flags, cancelled);
2128 2129
    qemuDomainObjExitRemoteWithDriver(driver, vm);

2130 2131 2132 2133 2134 2135 2136
    /* If ddomain is NULL, then we were unable to start
     * the guest on the target, and must restart on the
     * source. There is a small chance that the ddomain
     * is NULL due to an RPC failure, in which case
     * ddomain could in fact be running on the dest.
     * The lock manager plugins should take care of
     * safety in this scenario.
2137
     */
2138
    cancelled = ddomain == NULL ? 1 : 0;
2139

2140 2141 2142 2143 2144 2145
    /* If finish3 set an error, and we don't have an earlier
     * one we need to preserve it in case confirm3 overwrites
     */
    if (!orig_err)
        orig_err = virSaveLastError();

2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157
    /*
     * If cancelled, then src VM will be restarted, else
     * it will be killed
     */
    VIR_DEBUG("Confirm3 %p ret=%d vm=%p", sconn, ret, vm);
    VIR_FREE(cookiein);
    cookiein = cookieout;
    cookieinlen = cookieoutlen;
    cookieout = NULL;
    cookieoutlen = 0;
    ret = qemuMigrationConfirm(driver, sconn, vm,
                               cookiein, cookieinlen,
2158
                               flags, cancelled);
2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186
    /* If Confirm3 returns -1, there's nothing more we can
     * do, but fortunately worst case is that there is a
     * domain left in 'paused' state on source.
     */

 cleanup:
    if (ddomain) {
        virUnrefDomain(ddomain);
        ret = 0;
    } else {
        ret = -1;
    }

    if (st)
        virUnrefStream(st);

    if (orig_err) {
        virSetError(orig_err);
        virFreeError(orig_err);
    }
    VIR_FREE(uri_out);
    VIR_FREE(cookiein);
    VIR_FREE(cookieout);

    return ret;
}


2187
static int doPeer2PeerMigrate(struct qemud_driver *driver,
2188
                              virConnectPtr sconn,
2189
                              virDomainObjPtr vm,
2190
                              const char *xmlin,
2191
                              const char *dconnuri,
2192 2193 2194
                              const char *uri,
                              unsigned long flags,
                              const char *dname,
2195 2196
                              unsigned long resource,
                              bool *v3proto)
2197 2198 2199 2200
{
    int ret = -1;
    virConnectPtr dconn = NULL;
    bool p2p;
2201 2202 2203 2204
    VIR_DEBUG("driver=%p, sconn=%p, vm=%p, xmlin=%s, dconnuri=%s, "
              "uri=%s, flags=%lu, dname=%s, resource=%lu",
              driver, sconn, vm, NULLSTR(xmlin), NULLSTR(dconnuri),
              NULLSTR(uri), flags, NULLSTR(dname), resource);
2205 2206 2207 2208 2209 2210

    /* the order of operations is important here; we make sure the
     * destination side is completely setup before we touch the source
     */

    qemuDomainObjEnterRemoteWithDriver(driver, vm);
2211
    dconn = virConnectOpen(dconnuri);
2212 2213 2214 2215 2216 2217 2218 2219 2220 2221
    qemuDomainObjExitRemoteWithDriver(driver, vm);
    if (dconn == NULL) {
        qemuReportError(VIR_ERR_OPERATION_FAILED,
                        _("Failed to connect to remote libvirt URI %s"), uri);
        return -1;
    }

    qemuDomainObjEnterRemoteWithDriver(driver, vm);
    p2p = VIR_DRV_SUPPORTS_FEATURE(dconn->driver, dconn,
                                   VIR_DRV_FEATURE_MIGRATION_P2P);
2222 2223 2224 2225 2226 2227
        /* v3proto reflects whether the caller used Perform3, but with
         * p2p migrate, regardless of whether Perform3 or Perform3
         * were used, we decide protocol based on what target supports
         */
    *v3proto = VIR_DRV_SUPPORTS_FEATURE(dconn->driver, dconn,
                                        VIR_DRV_FEATURE_MIGRATION_V3);
2228
    qemuDomainObjExitRemoteWithDriver(driver, vm);
2229

2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242
    if (!p2p) {
        qemuReportError(VIR_ERR_OPERATION_FAILED, "%s",
                        _("Destination libvirt does not support peer-to-peer migration protocol"));
        goto cleanup;
    }

    /* domain may have been stopped while we were talking to remote daemon */
    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("guest unexpectedly quit"));
        goto cleanup;
    }

2243
    if (*v3proto)
2244
        ret = doPeer2PeerMigrate3(driver, sconn, dconn, vm, xmlin,
2245
                                  dconnuri, uri, flags, dname, resource);
2246 2247
    else
        ret = doPeer2PeerMigrate2(driver, sconn, dconn, vm,
2248
                                  dconnuri, flags, dname, resource);
2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262

cleanup:
    /* don't call virConnectClose(), because that resets any pending errors */
    qemuDomainObjEnterRemoteWithDriver(driver, vm);
    virUnrefConnect(dconn);
    qemuDomainObjExitRemoteWithDriver(driver, vm);

    return ret;
}


int qemuMigrationPerform(struct qemud_driver *driver,
                         virConnectPtr conn,
                         virDomainObjPtr vm,
2263
                         const char *xmlin,
2264
                         const char *dconnuri,
2265
                         const char *uri,
2266 2267 2268 2269
                         const char *cookiein,
                         int cookieinlen,
                         char **cookieout,
                         int *cookieoutlen,
2270 2271
                         unsigned long flags,
                         const char *dname,
2272
                         unsigned long resource,
2273
                         bool v3proto)
2274 2275 2276 2277 2278
{
    virDomainEventPtr event = NULL;
    int ret = -1;
    int resume = 0;
    qemuDomainObjPrivatePtr priv = vm->privateData;
2279 2280 2281 2282 2283 2284 2285
    VIR_DEBUG("driver=%p, conn=%p, vm=%p, xmlin=%s, dconnuri=%s, "
              "uri=%s, cookiein=%s, cookieinlen=%d, cookieout=%p, "
              "cookieoutlen=%p, flags=%lu, dname=%s, resource=%lu, v3proto=%d",
              driver, conn, vm, NULLSTR(xmlin), NULLSTR(dconnuri),
              NULLSTR(uri), NULLSTR(cookiein), cookieinlen,
              cookieout, cookieoutlen, flags, NULLSTR(dname),
              resource, v3proto);
2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296

    if (qemuDomainObjBeginJobWithDriver(driver, vm) < 0)
        goto cleanup;
    priv->jobActive = QEMU_JOB_MIGRATION_OUT;

    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_OPERATION_INVALID,
                        "%s", _("domain is not running"));
        goto endjob;
    }

2297 2298 2299 2300 2301 2302
    if (qemuProcessAutoDestroyActive(driver, vm)) {
        qemuReportError(VIR_ERR_OPERATION_INVALID,
                        "%s", _("domain is marked for auto destroy"));
        goto endjob;
    }

2303 2304 2305
    memset(&priv->jobInfo, 0, sizeof(priv->jobInfo));
    priv->jobInfo.type = VIR_DOMAIN_JOB_UNBOUNDED;

J
Jiri Denemark 已提交
2306
    resume = virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING;
2307 2308

    if ((flags & (VIR_MIGRATE_TUNNELLED | VIR_MIGRATE_PEER2PEER))) {
2309 2310 2311 2312 2313 2314
        if (cookieinlen) {
            qemuReportError(VIR_ERR_OPERATION_INVALID,
                            "%s", _("received unexpected cookie with P2P migration"));
            goto endjob;
        }

2315
        if (doPeer2PeerMigrate(driver, conn, vm, xmlin,
2316 2317
                               dconnuri, uri, flags, dname,
                               resource, &v3proto) < 0)
2318 2319 2320
            /* doPeer2PeerMigrate already set the error, so just get out */
            goto endjob;
    } else {
2321 2322 2323 2324 2325
        if (dconnuri) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR,
                            "%s", _("Unexpected dconnuri parameter with non-peer2peer migration"));
            goto endjob;
        }
2326 2327 2328
        if (doNativeMigrate(driver, vm, uri, cookiein, cookieinlen,
                            cookieout, cookieoutlen,
                            flags, dname, resource) < 0)
2329 2330 2331
            goto endjob;
    }

2332 2333 2334 2335 2336 2337 2338
    /*
     * In v3 protocol, the source VM is not killed off until the
     * confirm step.
     */
    if (v3proto) {
        resume = 0;
    } else {
2339 2340 2341
        qemuProcessStop(driver, vm, 1, VIR_DOMAIN_SHUTOFF_MIGRATED);
        qemuAuditDomainStop(vm, "migrated");
        resume = 0;
2342

2343 2344 2345
        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_STOPPED,
                                         VIR_DOMAIN_EVENT_STOPPED_MIGRATED);
2346
    }
2347

2348 2349 2350
    ret = 0;

endjob:
J
Jiri Denemark 已提交
2351
    if (resume && virDomainObjGetState(vm, NULL) == VIR_DOMAIN_PAUSED) {
2352
        /* we got here through some sort of failure; start the domain again */
J
Jiri Denemark 已提交
2353 2354
        if (qemuProcessStartCPUs(driver, vm, conn,
                                 VIR_DOMAIN_RUNNING_MIGRATION_CANCELED) < 0) {
2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366
            /* Hm, we already know we are in error here.  We don't want to
             * overwrite the previous error, though, so we just throw something
             * to the logs and hope for the best
             */
            VIR_ERROR(_("Failed to resume guest %s after failure"),
                      vm->def->name);
        }

        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_RESUMED,
                                         VIR_DOMAIN_EVENT_RESUMED_MIGRATED);
    }
2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377
    if (vm) {
        if (qemuDomainObjEndJob(vm) == 0) {
            vm = NULL;
        } else if (!virDomainObjIsActive(vm) &&
                   (!vm->persistent || (flags & VIR_MIGRATE_UNDEFINE_SOURCE))) {
            if (flags & VIR_MIGRATE_UNDEFINE_SOURCE)
                virDomainDeleteConfig(driver->configDir, driver->autostartDir, vm);
            virDomainRemoveInactive(&driver->domains, vm);
            vm = NULL;
        }
    }
2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432

cleanup:
    if (vm)
        virDomainObjUnlock(vm);
    if (event)
        qemuDomainEventQueue(driver, event);
    return ret;
}


#if WITH_MACVTAP
static void
qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def) {
    int i;
    int last_good_net = -1;
    virDomainNetDefPtr net;

    for (i = 0; i < def->nnets; i++) {
        net = def->nets[i];
        if (net->type == VIR_DOMAIN_NET_TYPE_DIRECT) {
            if (vpAssociatePortProfileId(net->ifname,
                                         net->mac,
                                         net->data.direct.linkdev,
                                         &net->data.direct.virtPortProfile,
                                         def->uuid,
                                         VIR_VM_OP_MIGRATE_IN_FINISH) != 0)
                goto err_exit;
        }
        last_good_net = i;
    }

    return;

err_exit:
    for (i = 0; i < last_good_net; i++) {
        net = def->nets[i];
        if (net->type == VIR_DOMAIN_NET_TYPE_DIRECT) {
            vpDisassociatePortProfileId(net->ifname,
                                        net->mac,
                                        net->data.direct.linkdev,
                                        &net->data.direct.virtPortProfile,
                                        VIR_VM_OP_MIGRATE_IN_FINISH);
        }
    }
}
#else /* !WITH_MACVTAP */
static void
qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def ATTRIBUTE_UNUSED) { }
#endif /* WITH_MACVTAP */


virDomainPtr
qemuMigrationFinish(struct qemud_driver *driver,
                    virConnectPtr dconn,
                    virDomainObjPtr vm,
2433 2434 2435 2436
                    const char *cookiein,
                    int cookieinlen,
                    char **cookieout,
                    int *cookieoutlen,
2437
                    unsigned long flags,
2438 2439
                    int retcode,
                    bool v3proto)
2440 2441 2442 2443 2444
{
    virDomainPtr dom = NULL;
    virDomainEventPtr event = NULL;
    int newVM = 1;
    qemuDomainObjPrivatePtr priv = NULL;
2445
    qemuMigrationCookiePtr mig = NULL;
2446 2447 2448 2449
    VIR_DEBUG("driver=%p, dconn=%p, vm=%p, cookiein=%s, cookieinlen=%d, "
              "cookieout=%p, cookieoutlen=%p, flags=%lu, retcode=%d",
              driver, dconn, vm, NULLSTR(cookiein), cookieinlen,
              cookieout, cookieoutlen, flags, retcode);
2450
    virErrorPtr orig_err = NULL;
2451 2452 2453 2454 2455 2456 2457 2458 2459 2460

    priv = vm->privateData;
    if (priv->jobActive != QEMU_JOB_MIGRATION_IN) {
        qemuReportError(VIR_ERR_NO_DOMAIN,
                        _("domain '%s' is not processing incoming migration"), vm->def->name);
        goto cleanup;
    }
    priv->jobActive = QEMU_JOB_NONE;
    memset(&priv->jobInfo, 0, sizeof(priv->jobInfo));

2461
    if (!(mig = qemuMigrationEatCookie(driver, vm, cookiein, cookieinlen, 0)))
2462 2463
        goto cleanup;

2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513
    if (qemuDomainObjBeginJobWithDriver(driver, vm) < 0)
        goto cleanup;

    /* Did the migration go as planned?  If yes, return the domain
     * object, but if no, clean up the empty qemu process.
     */
    if (retcode == 0) {
        if (!virDomainObjIsActive(vm)) {
            qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                            _("guest unexpectedly quit"));
            goto cleanup;
        }

        qemuMigrationVPAssociatePortProfiles(vm->def);

        if (flags & VIR_MIGRATE_PERSIST_DEST) {
            if (vm->persistent)
                newVM = 0;
            vm->persistent = 1;

            if (virDomainSaveConfig(driver->configDir, vm->def) < 0) {
                /* Hmpf.  Migration was successful, but making it persistent
                 * was not.  If we report successful, then when this domain
                 * shuts down, management tools are in for a surprise.  On the
                 * other hand, if we report failure, then the management tools
                 * might try to restart the domain on the source side, even
                 * though the domain is actually running on the destination.
                 * Return a NULL dom pointer, and hope that this is a rare
                 * situation and management tools are smart.
                 */
                vm = NULL;
                goto endjob;
            }

            event = virDomainEventNewFromObj(vm,
                                             VIR_DOMAIN_EVENT_DEFINED,
                                             newVM ?
                                             VIR_DOMAIN_EVENT_DEFINED_ADDED :
                                             VIR_DOMAIN_EVENT_DEFINED_UPDATED);
            if (event)
                qemuDomainEventQueue(driver, event);
            event = NULL;

        }

        if (!(flags & VIR_MIGRATE_PAUSED)) {
            /* run 'cont' on the destination, which allows migration on qemu
             * >= 0.10.6 to work properly.  This isn't strictly necessary on
             * older qemu's, but it also doesn't hurt anything there
             */
J
Jiri Denemark 已提交
2514 2515
            if (qemuProcessStartCPUs(driver, vm, dconn,
                                     VIR_DOMAIN_RUNNING_MIGRATED) < 0) {
2516 2517 2518 2519 2520 2521 2522 2523
                if (virGetLastError() == NULL)
                    qemuReportError(VIR_ERR_INTERNAL_ERROR,
                                    "%s", _("resume operation failed"));
                /* Need to save the current error, in case shutting
                 * down the process overwrites it
                 */
                orig_err = virSaveLastError();

2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543
                /*
                 * In v3 protocol, the source VM is still available to
                 * restart during confirm() step, so we kill it off
                 * now.
                 * In v2 protocol, the source is dead, so we leave
                 * target in paused state, in case admin can fix
                 * things up
                 */
                if (v3proto) {
                    qemuProcessStop(driver, vm, 1, VIR_DOMAIN_SHUTOFF_FAILED);
                    qemuAuditDomainStop(vm, "failed");
                    event = virDomainEventNewFromObj(vm,
                                                     VIR_DOMAIN_EVENT_STOPPED,
                                                     VIR_DOMAIN_EVENT_STOPPED_FAILED);
                    if (!vm->persistent) {
                        if (qemuDomainObjEndJob(vm) > 0)
                            virDomainRemoveInactive(&driver->domains, vm);
                        vm = NULL;
                    }
                }
2544 2545 2546 2547
                goto endjob;
            }
        }

2548 2549
        dom = virGetDomain (dconn, vm->def->name, vm->def->uuid);

2550 2551 2552
        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_RESUMED,
                                         VIR_DOMAIN_EVENT_RESUMED_MIGRATED);
J
Jiri Denemark 已提交
2553 2554
        if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_PAUSED) {
            virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_USER);
2555 2556
            if (event)
                qemuDomainEventQueue(driver, event);
2557 2558 2559 2560 2561 2562 2563 2564
            event = virDomainEventNewFromObj(vm,
                                             VIR_DOMAIN_EVENT_SUSPENDED,
                                             VIR_DOMAIN_EVENT_SUSPENDED_PAUSED);
        }
        if (virDomainSaveStatus(driver->caps, driver->stateDir, vm) < 0) {
            VIR_WARN("Failed to save status on vm %s", vm->def->name);
            goto endjob;
        }
2565 2566 2567

        /* Guest is successfully running, so cancel previous auto destroy */
        qemuProcessAutoDestroyRemove(driver, vm);
2568
    } else {
J
Jiri Denemark 已提交
2569
        qemuProcessStop(driver, vm, 1, VIR_DOMAIN_SHUTOFF_FAILED);
2570
        qemuAuditDomainStop(vm, "failed");
2571 2572 2573 2574 2575 2576 2577 2578 2579 2580
        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_STOPPED,
                                         VIR_DOMAIN_EVENT_STOPPED_FAILED);
        if (!vm->persistent) {
            if (qemuDomainObjEndJob(vm) > 0)
                virDomainRemoveInactive(&driver->domains, vm);
            vm = NULL;
        }
    }

2581 2582 2583
    if (qemuMigrationBakeCookie(mig, driver, vm, cookieout, cookieoutlen, 0) < 0)
        VIR_WARN("Unable to encode migration cookie");

2584 2585 2586 2587 2588 2589 2590 2591 2592 2593
endjob:
    if (vm &&
        qemuDomainObjEndJob(vm) == 0)
        vm = NULL;

cleanup:
    if (vm)
        virDomainObjUnlock(vm);
    if (event)
        qemuDomainEventQueue(driver, event);
2594
    qemuMigrationCookieFree(mig);
2595 2596 2597 2598
    if (orig_err) {
        virSetError(orig_err);
        virFreeError(orig_err);
    }
2599 2600
    return dom;
}
2601

2602 2603 2604 2605 2606 2607

int qemuMigrationConfirm(struct qemud_driver *driver,
                         virConnectPtr conn,
                         virDomainObjPtr vm,
                         const char *cookiein,
                         int cookieinlen,
2608 2609
                         unsigned int flags ATTRIBUTE_UNUSED,
                         int retcode)
2610 2611 2612 2613
{
    qemuMigrationCookiePtr mig;
    virDomainEventPtr event = NULL;
    int rv = -1;
2614 2615 2616 2617
    VIR_DEBUG("driver=%p, conn=%p, vm=%p, cookiein=%s, cookieinlen=%d, "
              "flags=%u, retcode=%d",
              driver, conn, vm, NULLSTR(cookiein), cookieinlen,
              flags, retcode);
2618

2619
    if (!(mig = qemuMigrationEatCookie(driver, vm, cookiein, cookieinlen, 0)))
2620 2621 2622 2623 2624
        return -1;

    if (!virDomainObjIsActive(vm)) {
        qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                        _("guest unexpectedly quit"));
2625
        goto cleanup;
2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648
    }

    /* Did the migration go as planned?  If yes, kill off the
     * domain object, but if no, resume CPUs
     */
    if (retcode == 0) {
        qemuProcessStop(driver, vm, 1, VIR_DOMAIN_SHUTOFF_MIGRATED);
        qemuAuditDomainStop(vm, "migrated");

        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_STOPPED,
                                         VIR_DOMAIN_EVENT_STOPPED_MIGRATED);
    } else {

        /* run 'cont' on the destination, which allows migration on qemu
         * >= 0.10.6 to work properly.  This isn't strictly necessary on
         * older qemu's, but it also doesn't hurt anything there
         */
        if (qemuProcessStartCPUs(driver, vm, conn,
                                 VIR_DOMAIN_RUNNING_MIGRATED) < 0) {
            if (virGetLastError() == NULL)
                qemuReportError(VIR_ERR_INTERNAL_ERROR,
                                "%s", _("resume operation failed"));
2649
            goto cleanup;
2650 2651 2652 2653 2654 2655 2656
        }

        event = virDomainEventNewFromObj(vm,
                                         VIR_DOMAIN_EVENT_RESUMED,
                                         VIR_DOMAIN_EVENT_RESUMED_MIGRATED);
        if (virDomainSaveStatus(driver->caps, driver->stateDir, vm) < 0) {
            VIR_WARN("Failed to save status on vm %s", vm->def->name);
2657
            goto cleanup;
2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670
        }
    }

    qemuMigrationCookieFree(mig);
    rv = 0;

cleanup:
    if (event)
        qemuDomainEventQueue(driver, event);
    return rv;
}


2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682
/* Helper function called while driver lock is held and vm is active.  */
int
qemuMigrationToFile(struct qemud_driver *driver, virDomainObjPtr vm,
                    int fd, off_t offset, const char *path,
                    const char *compressor,
                    bool is_reg, bool bypassSecurityDriver)
{
    qemuDomainObjPrivatePtr priv = vm->privateData;
    virCgroupPtr cgroup = NULL;
    int ret = -1;
    int rc;
    bool restoreLabel = false;
2683 2684
    virCommandPtr cmd = NULL;
    int pipeFD[2] = { -1, -1 };
2685

2686
    if (qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_FD) &&
2687
        (!compressor || pipe(pipeFD) == 0)) {
2688
        /* All right! We can use fd migration, which means that qemu
2689 2690 2691
         * doesn't have to open() the file, so while we still have to
         * grant SELinux access, we can do it on fd and avoid cleanup
         * later, as well as skip futzing with cgroup.  */
2692 2693
        if (virSecurityManagerSetImageFDLabel(driver->securityManager, vm,
                                              compressor ? pipeFD[1] : fd) < 0)
2694
            goto cleanup;
2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719
        bypassSecurityDriver = true;
    } else {
        /* Phooey - we have to fall back on exec migration, where qemu
         * has to popen() the file by name.  We might also stumble on
         * a race present in some qemu versions where it does a wait()
         * that botches pclose.  */
        if (!is_reg &&
            qemuCgroupControllerActive(driver,
                                       VIR_CGROUP_CONTROLLER_DEVICES)) {
            if (virCgroupForDomain(driver->cgroup, vm->def->name,
                                   &cgroup, 0) != 0) {
                qemuReportError(VIR_ERR_INTERNAL_ERROR,
                                _("Unable to find cgroup for %s"),
                                vm->def->name);
                goto cleanup;
            }
            rc = virCgroupAllowDevicePath(cgroup, path,
                                          VIR_CGROUP_DEVICE_RW);
            qemuAuditCgroupPath(vm, cgroup, "allow", path, "rw", rc);
            if (rc < 0) {
                virReportSystemError(-rc,
                                     _("Unable to allow device %s for %s"),
                                     path, vm->def->name);
                goto cleanup;
            }
2720
        }
2721 2722 2723
        if ((!bypassSecurityDriver) &&
            virSecurityManagerSetSavedStateLabel(driver->securityManager,
                                                 vm, path) < 0)
2724
            goto cleanup;
2725
        restoreLabel = true;
2726 2727
    }

2728
    qemuDomainObjEnterMonitorWithDriver(driver, vm);
2729 2730 2731
    if (!compressor) {
        const char *args[] = { "cat", NULL };

2732
        if (qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_FD) &&
2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748
            priv->monConfig->type == VIR_DOMAIN_CHR_TYPE_UNIX) {
            rc = qemuMonitorMigrateToFd(priv->mon,
                                        QEMU_MONITOR_MIGRATE_BACKGROUND,
                                        fd);
        } else {
            rc = qemuMonitorMigrateToFile(priv->mon,
                                          QEMU_MONITOR_MIGRATE_BACKGROUND,
                                          args, path, offset);
        }
    } else {
        const char *prog = compressor;
        const char *args[] = {
            prog,
            "-c",
            NULL
        };
2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767
        if (pipeFD[0] != -1) {
            cmd = virCommandNewArgs(args);
            virCommandSetInputFD(cmd, pipeFD[0]);
            virCommandSetOutputFD(cmd, &fd);
            if (virSetCloseExec(pipeFD[1]) < 0) {
                virReportSystemError(errno, "%s",
                                     _("Unable to set cloexec flag"));
                qemuDomainObjExitMonitorWithDriver(driver, vm);
                goto cleanup;
            }
            if (virCommandRunAsync(cmd, NULL) < 0) {
                qemuDomainObjExitMonitorWithDriver(driver, vm);
                goto cleanup;
            }
            rc = qemuMonitorMigrateToFd(priv->mon,
                                        QEMU_MONITOR_MIGRATE_BACKGROUND,
                                        pipeFD[1]);
            if (VIR_CLOSE(pipeFD[0]) < 0 ||
                VIR_CLOSE(pipeFD[1]) < 0)
2768
                VIR_WARN("failed to close intermediate pipe");
2769 2770 2771 2772 2773
        } else {
            rc = qemuMonitorMigrateToFile(priv->mon,
                                          QEMU_MONITOR_MIGRATE_BACKGROUND,
                                          args, path, offset);
        }
2774
    }
2775
    qemuDomainObjExitMonitorWithDriver(driver, vm);
2776 2777 2778 2779 2780 2781 2782 2783 2784

    if (rc < 0)
        goto cleanup;

    rc = qemuMigrationWaitForCompletion(driver, vm);

    if (rc < 0)
        goto cleanup;

2785 2786 2787
    if (cmd && virCommandWait(cmd, NULL) < 0)
        goto cleanup;

2788 2789 2790
    ret = 0;

cleanup:
2791 2792 2793
    VIR_FORCE_CLOSE(pipeFD[0]);
    VIR_FORCE_CLOSE(pipeFD[1]);
    virCommandFree(cmd);
2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809
    if (restoreLabel && (!bypassSecurityDriver) &&
        virSecurityManagerRestoreSavedStateLabel(driver->securityManager,
                                                 vm, path) < 0)
        VIR_WARN("failed to restore save state label on %s", path);

    if (cgroup != NULL) {
        rc = virCgroupDenyDevicePath(cgroup, path,
                                     VIR_CGROUP_DEVICE_RWM);
        qemuAuditCgroupPath(vm, cgroup, "deny", path, "rwm", rc);
        if (rc < 0)
            VIR_WARN("Unable to deny device %s for %s %d",
                     path, vm->def->name, rc);
        virCgroupFree(&cgroup);
    }
    return ret;
}