static int
qemuAgentOpenPty(const char *monitor)
{
int monfd;
if ((monfd = open(monitor, O_RDWR | O_NONBLOCK)) < 0) {
virReportSystemError(errno,
_("Unable to open monitor path %s"), monitor);
return -1;
}
if (virSetCloseExec(monfd) < 0) {
virReportSystemError(errno, "%s",
_("Unable to set monitor close-on-exec flag"));
goto error;
}
return monfd;
error:
VIR_FORCE_CLOSE(monfd);
return -1;
}
static int
qemuAgentOpenUnix(const char *monitor, pid_t cpid, bool *inProgress)
{
struct sockaddr_un addr;
int monfd;
int timeout = 3; /* In seconds */
int ret, i = 0;
*inProgress = false;
if ((monfd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) {
virReportSystemError(errno,
"%s", _("failed to create socket"));
return -1;
}
if (virSetNonBlock(monfd) < 0) {
virReportSystemError(errno, "%s",
_("Unable to put monitor "
"into non-blocking mode"));
goto error;
}
if (virSetCloseExec(monfd) < 0) {
virReportSystemError(errno, "%s",
_("Unable to set monitor "
"close-on-exec flag"));
goto error;
}
memset(&addr, 0, sizeof(addr));
addr.sun_family = AF_UNIX;
if (virStrcpyStatic(addr.sun_path, monitor) == NULL) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("Agent path %s too big for destination"), monitor);
goto error;
}
do {
ret = connect(monfd, (struct sockaddr *) &addr, sizeof(addr));
if (ret == 0)
break;
if ((errno == ENOENT || errno == ECONNREFUSED) &&
virProcessKill(cpid, 0) == 0) {
/* ENOENT : Socket may not have shown up yet
* ECONNREFUSED : Leftover socket hasn't been removed yet */
continue;
}
if ((errno == EINPROGRESS) ||
(errno == EAGAIN)) {
VIR_DEBUG("Connection attempt continuing in background");
*inProgress = true;
ret = 0;
break;
}
virReportSystemError(errno, "%s",
_("failed to connect to monitor socket"));
goto error;
} while ((++i <= timeout*5) && (usleep(.2 * 1000000) <= 0));
if (ret != 0) {
virReportSystemError(errno, "%s",
_("monitor socket did not show up."));
goto error;
}
return monfd;
error:
VIR_FORCE_CLOSE(monfd);
return -1;
}
static virNetSocketPtr virNetSocketNew(virSocketAddrPtr localAddr,
virSocketAddrPtr remoteAddr,
bool isClient,
int fd, int errfd, pid_t pid)
{
virNetSocketPtr sock;
int no_slow_start = 1;
VIR_DEBUG("localAddr=%p remoteAddr=%p fd=%d errfd=%d pid=%d",
localAddr, remoteAddr,
fd, errfd, pid);
if (virSetCloseExec(fd) < 0) {
virReportSystemError(errno, "%s",
_("Unable to set close-on-exec flag"));
return NULL;
}
if (virSetNonBlock(fd) < 0) {
virReportSystemError(errno, "%s",
_("Unable to enable non-blocking flag"));
return NULL;
}
if (VIR_ALLOC(sock) < 0) {
virReportOOMError();
return NULL;
}
if (virMutexInit(&sock->lock) < 0) {
virReportSystemError(errno, "%s",
_("Unable to initialize mutex"));
VIR_FREE(sock);
return NULL;
}
sock->refs = 1;
if (localAddr)
sock->localAddr = *localAddr;
if (remoteAddr)
sock->remoteAddr = *remoteAddr;
sock->fd = fd;
sock->errfd = errfd;
sock->pid = pid;
/* Disable nagle for TCP sockets */
if (sock->localAddr.data.sa.sa_family == AF_INET ||
sock->localAddr.data.sa.sa_family == AF_INET6) {
if (setsockopt(fd, IPPROTO_TCP, TCP_NODELAY,
&no_slow_start,
sizeof(no_slow_start)) < 0) {
virReportSystemError(errno, "%s",
_("Unable to disable nagle algorithm"));
goto error;
}
}
if (localAddr &&
!(sock->localAddrStr = virSocketFormatAddrFull(localAddr, true, ";")))
goto error;
if (remoteAddr &&
!(sock->remoteAddrStr = virSocketFormatAddrFull(remoteAddr, true, ";")))
goto error;
sock->client = isClient;
PROBE(RPC_SOCKET_NEW,
"sock=%p refs=%d fd=%d errfd=%d pid=%d localAddr=%s, remoteAddr=%s",
sock, sock->refs, fd, errfd,
pid, NULLSTR(sock->localAddrStr), NULLSTR(sock->remoteAddrStr));
return sock;
error:
sock->fd = sock->errfd = -1; /* Caller owns fd/errfd on failure */
virNetSocketFree(sock);
return NULL;
}
static virLockSpaceResourcePtr
virLockSpaceResourceNew(virLockSpacePtr lockspace,
const char *resname,
unsigned int flags,
pid_t owner)
{
virLockSpaceResourcePtr res;
bool shared = !!(flags & VIR_LOCK_SPACE_ACQUIRE_SHARED);
if (VIR_ALLOC(res) < 0)
return NULL;
res->fd = -1;
res->flags = flags;
if (VIR_STRDUP(res->name, resname) < 0)
goto error;
if (!(res->path = virLockSpaceGetResourcePath(lockspace, resname)))
goto error;
if (flags & VIR_LOCK_SPACE_ACQUIRE_AUTOCREATE) {
while (1) {
struct stat a, b;
if ((res->fd = open(res->path, O_RDWR|O_CREAT, 0600)) < 0) {
virReportSystemError(errno,
_("Unable to open/create resource %s"),
res->path);
goto error;
}
if (virSetCloseExec(res->fd) < 0) {
virReportSystemError(errno,
_("Failed to set close-on-exec flag '%s'"),
res->path);
goto error;
}
if (fstat(res->fd, &b) < 0) {
virReportSystemError(errno,
_("Unable to check status of pid file '%s'"),
res->path);
goto error;
}
if (virFileLock(res->fd, shared, 0, 1, false) < 0) {
if (errno == EACCES || errno == EAGAIN) {
virReportError(VIR_ERR_RESOURCE_BUSY,
_("Lockspace resource '%s' is locked"),
resname);
} else {
virReportSystemError(errno,
_("Unable to acquire lock on '%s'"),
res->path);
}
goto error;
}
/* Now make sure the pidfile we locked is the same
* one that now exists on the filesystem
*/
if (stat(res->path, &a) < 0) {
char ebuf[1024] ATTRIBUTE_UNUSED;
VIR_DEBUG("Resource '%s' disappeared: %s",
res->path, virStrerror(errno, ebuf, sizeof(ebuf)));
VIR_FORCE_CLOSE(res->fd);
/* Someone else must be racing with us, so try again */
continue;
}
if (a.st_ino == b.st_ino)
break;
VIR_DEBUG("Resource '%s' was recreated", res->path);
VIR_FORCE_CLOSE(res->fd);
/* Someone else must be racing with us, so try again */
}
} else {
int virPidFileAcquirePath(const char *path,
pid_t pid)
{
int fd = -1;
char pidstr[INT_BUFSIZE_BOUND(pid)];
if (path[0] == '\0')
return 0;
while (1) {
struct stat a, b;
if ((fd = open(path, O_WRONLY|O_CREAT, 0644)) < 0) {
virReportSystemError(errno,
_("Failed to open pid file '%s'"),
path);
return -1;
}
if (virSetCloseExec(fd) < 0) {
virReportSystemError(errno,
_("Failed to set close-on-exec flag '%s'"),
path);
VIR_FORCE_CLOSE(fd);
return -1;
}
if (fstat(fd, &b) < 0) {
virReportSystemError(errno,
_("Unable to check status of pid file '%s'"),
path);
VIR_FORCE_CLOSE(fd);
return -1;
}
if (virFileLock(fd, false, 0, 1) < 0) {
virReportSystemError(errno,
_("Failed to acquire pid file '%s'"),
path);
VIR_FORCE_CLOSE(fd);
return -1;
}
/* Now make sure the pidfile we locked is the same
* one that now exists on the filesystem
*/
if (stat(path, &a) < 0) {
char ebuf[1024] ATTRIBUTE_UNUSED;
VIR_DEBUG("Pid file '%s' disappeared: %s",
path, virStrerror(errno, ebuf, sizeof(ebuf)));
VIR_FORCE_CLOSE(fd);
/* Someone else must be racing with us, so try agin */
continue;
}
if (a.st_ino == b.st_ino)
break;
VIR_DEBUG("Pid file '%s' was recreated", path);
VIR_FORCE_CLOSE(fd);
/* Someone else must be racing with us, so try agin */
}
int
qemuMigrationWaitForCompletion(struct qemud_driver *driver, virDomainObjPtr vm)
{
int ret = -1;
int status;
unsigned long long memProcessed;
unsigned long long memRemaining;
unsigned long long memTotal;
qemuDomainObjPrivatePtr priv = vm->privateData;
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 };
struct timeval now;
int rc;
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");
}
if (!virDomainObjIsActive(vm)) {
qemuReportError(VIR_ERR_INTERNAL_ERROR, _("%s: %s"),
job, _("guest unexpectedly quit"));
goto cleanup;
}
if (priv->jobSignals & QEMU_JOB_SIGNAL_CANCEL) {
priv->jobSignals ^= QEMU_JOB_SIGNAL_CANCEL;
VIR_DEBUG0("Cancelling job at client request");
qemuDomainObjEnterMonitorWithDriver(driver, vm);
rc = qemuMonitorMigrateCancel(priv->mon);
qemuDomainObjExitMonitorWithDriver(driver, vm);
if (rc < 0) {
VIR_WARN0("Unable to cancel job");
}
} else if (priv->jobSignals & QEMU_JOB_SIGNAL_SUSPEND) {
priv->jobSignals ^= QEMU_JOB_SIGNAL_SUSPEND;
VIR_DEBUG0("Pausing domain for non-live migration");
if (qemuMigrationSetOffline(driver, vm) < 0)
VIR_WARN0("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);
rc = qemuMonitorSetMigrationDowntime(priv->mon, ms);
qemuDomainObjExitMonitorWithDriver(driver, vm);
if (rc < 0)
VIR_WARN0("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);
rc = qemuMonitorSetMigrationSpeed(priv->mon, bandwidth);
qemuDomainObjExitMonitorWithDriver(driver, vm);
if (rc < 0)
VIR_WARN0("Unable to set migration speed");
}
/* Repeat check because the job signals might have caused
* guest to die
*/
if (!virDomainObjIsActive(vm)) {
qemuReportError(VIR_ERR_INTERNAL_ERROR, _("%s: %s"),
job, _("guest unexpectedly quit"));
goto cleanup;
}
qemuDomainObjEnterMonitorWithDriver(driver, vm);
rc = qemuMonitorGetMigrationStatus(priv->mon,
&status,
&memProcessed,
&memRemaining,
&memTotal);
qemuDomainObjExitMonitorWithDriver(driver, vm);
if (rc < 0) {
priv->jobInfo.type = VIR_DOMAIN_JOB_FAILED;
goto cleanup;
}
if (gettimeofday(&now, NULL) < 0) {
priv->jobInfo.type = VIR_DOMAIN_JOB_FAILED;
virReportSystemError(errno, "%s",
_("cannot get time of day"));
goto cleanup;
}
priv->jobInfo.timeElapsed = timeval_to_ms(now) - priv->jobStart;
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;
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;
}
virDomainObjUnlock(vm);
qemuDriverUnlock(driver);
nanosleep(&ts, NULL);
qemuDriverLock(driver);
virDomainObjLock(vm);
}
cleanup:
return ret;
}
/* 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,
virStreamPtr st,
const char *dname,
const char *dom_xml)
{
virDomainDefPtr def = NULL;
virDomainObjPtr vm = NULL;
virDomainEventPtr event = NULL;
int ret = -1;
int internalret;
int dataFD[2] = { -1, -1 };
virBitmapPtr qemuCaps = NULL;
qemuDomainObjPrivatePtr priv = NULL;
struct timeval now;
if (gettimeofday(&now, NULL) < 0) {
virReportSystemError(errno, "%s",
_("cannot get time of day"));
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;
if (qemuDomainObjBeginJobWithDriver(driver, vm) < 0)
goto cleanup;
priv->jobActive = QEMU_JOB_MIGRATION_OUT;
/* Domain starts inactive, even if the domain XML had an id field. */
vm->def->id = -1;
if (pipe(dataFD) < 0 ||
virSetCloseExec(dataFD[0]) < 0) {
virReportSystemError(errno, "%s",
_("cannot create pipe for tunnelled migration"));
goto endjob;
}
/* check that this qemu version supports the interactive exec */
if (qemuCapsExtractVersionInfo(vm->def->emulator, vm->def->os.arch,
NULL, &qemuCaps) < 0) {
qemuReportError(VIR_ERR_INTERNAL_ERROR,
_("Cannot determine QEMU argv syntax %s"),
vm->def->emulator);
goto endjob;
}
/* Start the QEMU daemon, with the same command-line arguments plus
* -incoming stdin (which qemu_command might convert to exec:cat or fd:n)
*/
internalret = qemuProcessStart(dconn, driver, vm, "stdin", true, dataFD[1],
NULL, VIR_VM_OP_MIGRATE_IN_START);
if (internalret < 0) {
qemuAuditDomainStart(vm, "migrated", false);
/* 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;
}
if (virFDStreamOpen(st, dataFD[0]) < 0) {
qemuAuditDomainStart(vm, "migrated", false);
qemuProcessStop(driver, vm, 0);
if (!vm->persistent) {
if (qemuDomainObjEndJob(vm) > 0)
virDomainRemoveInactive(&driver->domains, vm);
vm = NULL;
}
virReportSystemError(errno, "%s",
_("cannot pass pipe for tunnelled migration"));
goto endjob;
}
qemuAuditDomainStart(vm, "migrated", true);
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;
priv->jobStart = timeval_to_ms(now);
}
cleanup:
qemuCapsFree(qemuCaps);
virDomainDefFree(def);
VIR_FORCE_CLOSE(dataFD[0]);
VIR_FORCE_CLOSE(dataFD[1]);
if (vm)
virDomainObjUnlock(vm);
if (event)
qemuDomainEventQueue(driver, event);
qemuDriverUnlock(driver);
return ret;
}
