- Devel - libvirt List Archives

[libvirt] [PATCH 00/20] Add support for (qcow*) volume encryption.
by Miloslav Trmač 11 Aug '09

11 Aug '09

Hello, the following patches add full support for qcow/qcow2 volume encryption, assuming a client that supports it. (Main changes since the previous version: * Use a separate API for "secret" management * Auto-generate the encryption format and passphrase if not supplied See the specific patch change logs for more details; patches without change logs are entirely new.) New XML tags are defined to represent encryption parameters (currently format and passphrase, more can be added in the future), e.g. <encryption format='qcow'> <secret type='passphrase' secret_id='724d95f2-0ed2-6ff9-84d0-0f3d1618428d'/> </encryption> The <encryption> tag can be added to a <volume> node passed to virStorageVolCreateXML() to create an encrypted volume, or to a <disk> node inside a <domain> to specify what encryption parameters to use for a domain. secret_id above refers to a separately-managed secret, which was set using virSecretSetValue(). Other properties of the secret can be managed using an XML representation. Detailed documentation of the formats and features is inside the patches.

3 25

[libvirt] [PATCH] add file descriptor migration
by Paolo Bonzini 11 Aug '09

11 Aug '09

This is the patch from Chris, with the accept removed and rebased above my previous cleanup. Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com> Cc: Chris Lalancette <clalance(a)redhat.com> Cc: Avi Kivity <avi(a)redhat.com> --- Makefile | 2 +- hw/hw.h | 1 + migration-fd.c | 137 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ migration.c | 4 ++ migration.h | 7 +++ savevm.c | 28 +++++++++++ 6 files changed, 178 insertions(+), 1 deletions(-) create mode 100644 migration-fd.c diff --git a/Makefile b/Makefile index 5279504..c7ff0be 100644 --- a/Makefile +++ b/Makefile @@ -93,7 +93,7 @@ obj-y += qdev.o qdev-properties.o ssi.o obj-$(CONFIG_BRLAPI) += baum.o obj-$(CONFIG_WIN32) += tap-win32.o -obj-$(CONFIG_POSIX) += migration-exec.o +obj-$(CONFIG_POSIX) += migration-exec.o migration-fd.o audio/audio.o audio/fmodaudio.o: QEMU_CFLAGS += $(FMOD_CFLAGS) diff --git a/hw/hw.h b/hw/hw.h index 322f077..91bf800 100644 --- a/hw/hw.h +++ b/hw/hw.h @@ -49,6 +49,7 @@ QEMUFile *qemu_fopen_ops(void *opaque, QEMUFilePutBufferFunc *put_buffer, QEMUFileRateLimit *rate_limit, QEMUFileSetRateLimit *set_rate_limit); QEMUFile *qemu_fopen(const char *filename, const char *mode); +QEMUFile *qemu_fdopen(int fd, const char *mode); QEMUFile *qemu_fopen_socket(int fd); QEMUFile *qemu_popen(FILE *popen_file, const char *mode); QEMUFile *qemu_popen_cmd(const char *command, const char *mode); diff --git a/migration-fd.c b/migration-fd.c new file mode 100644 index 0000000..794c2ac --- /dev/null +++ b/migration-fd.c @@ -0,0 +1,137 @@ +/* + * QEMU live migration via generic fd + * + * Copyright Red Hat, Inc. 2009 + * + * Authors: + * Chris Lalancette <clalance(a)redhat.com> + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include "qemu-common.h" +#include "qemu_socket.h" +#include "migration.h" +#include "monitor.h" +#include "qemu-char.h" +#include "sysemu.h" +#include "buffered_file.h" +#include "block.h" +#include "qemu_socket.h" + +//#define DEBUG_MIGRATION_FD + +#ifdef DEBUG_MIGRATION_FD +#define dprintf(fmt, ...) \ + do { printf("migration-fd: " fmt, ## __VA_ARGS__); } while (0) +#else +#define dprintf(fmt, ...) \ + do { } while (0) +#endif + +static int fd_errno(FdMigrationState *s) +{ + return errno; +} + +static int fd_write(FdMigrationState *s, const void * buf, size_t size) +{ + return write(s->fd, buf, size); +} + +static int fd_close(FdMigrationState *s) +{ + dprintf("fd_close\n"); + if (s->fd != -1) { + close(s->fd); + s->fd = -1; + } + return 0; +} + +MigrationState *fd_start_outgoing_migration(Monitor *mon, + const char *fdname, + int64_t bandwidth_limit, + int detach) +{ + FdMigrationState *s; + + s = qemu_mallocz(sizeof(*s)); + + s->fd = monitor_get_fd(mon, fdname); + if (s->fd == -1) { + dprintf("fd_migration: invalid file descriptor identifier\n"); + goto err_after_alloc; + } + + if (fcntl(s->fd, F_SETFD, O_NONBLOCK) == -1) { + dprintf("Unable to set nonblocking mode on file descriptor\n"); + goto err_after_open; + } + + s->get_error = fd_errno; + s->write = fd_write; + s->close = fd_close; + s->mig_state.cancel = migrate_fd_cancel; + s->mig_state.get_status = migrate_fd_get_status; + s->mig_state.release = migrate_fd_release; + + s->state = MIG_STATE_ACTIVE; + s->mon_resume = NULL; + s->bandwidth_limit = bandwidth_limit; + + if (!detach) + migrate_fd_monitor_suspend(s); + + migrate_fd_connect(s); + return &s->mig_state; + +err_after_open: + close(s->fd); +err_after_alloc: + qemu_free(s); + return NULL; +} + +static void fd_accept_incoming_migration(void *opaque) +{ + QEMUFile *f = opaque; + int ret; + + ret = qemu_loadvm_state(f); + if (ret < 0) { + fprintf(stderr, "load of migration failed\n"); + goto err; + } + qemu_announce_self(); + dprintf("successfully loaded vm state\n"); + /* we've successfully migrated, close the fd */ + qemu_set_fd_handler2(qemu_stdio_fd(f), NULL, NULL, NULL, NULL); + if (autostart) + vm_start(); + +err: + qemu_fclose(f); +} + +int fd_start_incoming_migration(const char *infd) +{ + int fd; + QEMUFile *f; + + dprintf("Attempting to start an incoming migration via fd\n"); + + fd = strtol(infd, NULL, 0); + f = qemu_fdopen(fd, "rb"); + if(f == NULL) { + dprintf("Unable to apply qemu wrapper to file descriptor\n"); + return -errno; + } + + qemu_set_fd_handler2(fd, NULL, fd_accept_incoming_migration, NULL, + (void *)(unsigned long)f); + + return 0; +} diff --git a/migration.c b/migration.c index ee64d41..0ca4399 100644 --- a/migration.c +++ b/migration.c @@ -43,6 +43,8 @@ void qemu_start_incoming_migration(const char *uri) #if !defined(WIN32) else if (strstart(uri, "exec:", &p)) exec_start_incoming_migration(p); + else if (strstart(uri, "fd:", &p)) + fd_start_incoming_migration(p); #endif else fprintf(stderr, "unknown migration protocol: %s\n", uri); @@ -58,6 +60,8 @@ void do_migrate(Monitor *mon, int detach, const char *uri) #if !defined(WIN32) else if (strstart(uri, "exec:", &p)) s = exec_start_outgoing_migration(p, max_throttle, detach); + else if (strstart(uri, "fd:", &p)) + s = fd_start_outgoing_migration(mon, p, max_throttle, detach); #endif else monitor_printf(mon, "unknown migration protocol: %s\n", uri); diff --git a/migration.h b/migration.h index 37c7f8e..5033d02 100644 --- a/migration.h +++ b/migration.h @@ -73,6 +73,13 @@ MigrationState *tcp_start_outgoing_migration(const char *host_port, int64_t bandwidth_limit, int detach); +int fd_start_incoming_migration(const char *path); + +MigrationState *fd_start_outgoing_migration(Monitor *mon, + const char *fdname, + int64_t bandwidth_limit, + int detach); + void migrate_fd_monitor_suspend(FdMigrationState *s); void migrate_fd_error(FdMigrationState *s); diff --git a/savevm.c b/savevm.c index 975e7ab..4575653 100644 --- a/savevm.c +++ b/savevm.c @@ -285,6 +285,34 @@ int qemu_stdio_fd(QEMUFile *f) return fd; } +QEMUFile *qemu_fdopen(int fd, const char *mode) +{ + QEMUFileStdio *s; + + if (mode == NULL || + (mode[0] != 'r' && mode[0] != 'w') || + mode[1] != 'b' || mode[2] != 0) { + fprintf(stderr, "qemu_fdopen: Argument validity check failed\n"); + return NULL; + } + + s = qemu_mallocz(sizeof(QEMUFileStdio)); + s->stdio_file = fdopen(fd, mode); + if (!s->stdio_file) + goto fail; + + if(mode[0] == 'r') { + s->file = qemu_fopen_ops(s, NULL, stdio_get_buffer, stdio_fclose, NULL, NULL); + } else { + s->file = qemu_fopen_ops(s, stdio_put_buffer, NULL, stdio_fclose, NULL, NULL); + } + return s->file; + +fail: + qemu_free(s); + return NULL; +} + QEMUFile *qemu_fopen_socket(int fd) { QEMUFileSocket *s = qemu_mallocz(sizeof(QEMUFileSocket)); -- 1.6.2.5

1 0

[libvirt] [PATCH] Split virDomainMigrate into functions.
by Chris Lalancette 11 Aug '09

11 Aug '09

Re-factor virDomainMigrate to split out the version 1 and version 2 protocols into their own functions. In reality, the two versions share very little in common, so forcing them together in the same function was just confusing. This will also make adding tunnelled migration easier. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/libvirt.c | 258 ++++++++++++++++++++++++++++++++++----------------------- 1 files changed, 155 insertions(+), 103 deletions(-) diff --git a/src/libvirt.c b/src/libvirt.c index 9fd864d..6c1cc3d 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -2874,6 +2874,146 @@ error: } +static virDomainPtr +virDomainMigrateVersion1 (virDomainPtr domain, + virConnectPtr dconn, + unsigned long flags, + const char *dname, + const char *uri, + unsigned long bandwidth) +{ + virDomainPtr ddomain = NULL; + char *uri_out = NULL; + char *cookie = NULL; + int cookielen = 0; + + /* Prepare the migration. + * + * The destination host may return a cookie, or leave cookie as + * NULL. + * + * The destination host MUST set uri_out if uri_in is NULL. + * + * If uri_in is non-NULL, then the destination host may modify + * the URI by setting uri_out. If it does not wish to modify + * the URI, it should leave uri_out as NULL. + */ + if (dconn->driver->domainMigratePrepare + (dconn, &cookie, &cookielen, uri, &uri_out, flags, dname, + bandwidth) == -1) + goto done; + + if (uri == NULL && uri_out == NULL) { + virLibConnError (domain->conn, VIR_ERR_INTERNAL_ERROR, + _("domainMigratePrepare did not set uri")); + goto done; + } + if (uri_out) + uri = uri_out; /* Did domainMigratePrepare change URI? */ + assert (uri != NULL); + + /* Perform the migration. The driver isn't supposed to return + * until the migration is complete. + */ + if (domain->conn->driver->domainMigratePerform + (domain, cookie, cookielen, uri, flags, dname, bandwidth) == -1) + goto done; + + /* Get the destination domain and return it or error. + * 'domain' no longer actually exists at this point + * (or so we hope), but we still use the object in memory + * in order to get the name. + */ + dname = dname ? dname : domain->name; + if (dconn->driver->domainMigrateFinish) + ddomain = dconn->driver->domainMigrateFinish + (dconn, dname, cookie, cookielen, uri, flags); + else + ddomain = virDomainLookupByName (dconn, dname); + + done: + VIR_FREE (uri_out); + VIR_FREE (cookie); + return ddomain; +} + +static virDomainPtr +virDomainMigrateVersion2 (virDomainPtr domain, + virConnectPtr dconn, + unsigned long flags, + const char *dname, + const char *uri, + unsigned long bandwidth) +{ + virDomainPtr ddomain = NULL; + char *uri_out = NULL; + char *cookie = NULL; + char *dom_xml = NULL; + int cookielen = 0, ret; + + /* Prepare the migration. + * + * The destination host may return a cookie, or leave cookie as + * NULL. + * + * The destination host MUST set uri_out if uri_in is NULL. + * + * If uri_in is non-NULL, then the destination host may modify + * the URI by setting uri_out. If it does not wish to modify + * the URI, it should leave uri_out as NULL. + */ + + /* 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 (!domain->conn->driver->domainDumpXML) { + virLibConnError (domain->conn, VIR_ERR_INTERNAL_ERROR, __FUNCTION__); + return NULL; + } + dom_xml = domain->conn->driver->domainDumpXML (domain, + VIR_DOMAIN_XML_SECURE); + if (!dom_xml) + return NULL; + + ret = dconn->driver->domainMigratePrepare2 + (dconn, &cookie, &cookielen, uri, &uri_out, flags, dname, + bandwidth, dom_xml); + VIR_FREE (dom_xml); + if (ret == -1) + goto done; + + if (uri == NULL && uri_out == NULL) { + virLibConnError (domain->conn, VIR_ERR_INTERNAL_ERROR, + _("domainMigratePrepare2 did not set uri")); + goto done; + } + if (uri_out) + uri = uri_out; /* Did domainMigratePrepare2 change URI? */ + assert (uri != NULL); + + /* Perform the migration. The driver isn't supposed to return + * until the migration is complete. + */ + ret = domain->conn->driver->domainMigratePerform + (domain, cookie, cookielen, uri, flags, dname, bandwidth); + if (ret == -1) + goto done; + + /* 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. + */ + dname = dname ? dname : domain->name; + ddomain = dconn->driver->domainMigrateFinish2 + (dconn, dname, cookie, cookielen, uri, flags, ret); + + done: + VIR_FREE (uri_out); + VIR_FREE (cookie); + return ddomain; +} + /** * virDomainMigrate: * @domain: a domain object @@ -2930,140 +3070,52 @@ virDomainMigrate (virDomainPtr domain, const char *uri, unsigned long bandwidth) { - virConnectPtr conn; virDomainPtr ddomain = NULL; - char *uri_out = NULL; - char *cookie = NULL; - char *dom_xml = NULL; - int cookielen = 0, ret, version = 0; DEBUG("domain=%p, dconn=%p, flags=%lu, dname=%s, uri=%s, bandwidth=%lu", domain, dconn, flags, NULLSTR(dname), NULLSTR(uri), bandwidth); virResetLastError(); + /* First checkout the source */ if (!VIR_IS_CONNECTED_DOMAIN (domain)) { virLibDomainError(NULL, VIR_ERR_INVALID_DOMAIN, __FUNCTION__); return NULL; } - conn = domain->conn; /* Source connection. */ - if (!VIR_IS_CONNECT (dconn)) { - virLibConnError (conn, VIR_ERR_INVALID_CONN, __FUNCTION__); + if (domain->conn->flags & VIR_CONNECT_RO) { + virLibDomainError(domain, VIR_ERR_OPERATION_DENIED, __FUNCTION__); goto error; } - if (domain->conn->flags & VIR_CONNECT_RO) { - virLibDomainError(domain, VIR_ERR_OPERATION_DENIED, __FUNCTION__); + /* Now checkout the destination */ + if (!VIR_IS_CONNECT (dconn)) { + virLibConnError (domain->conn, VIR_ERR_INVALID_CONN, __FUNCTION__); goto error; } if (dconn->flags & VIR_CONNECT_RO) { - /* NB, delibrately report error against source object, not dest here */ - virLibDomainError(domain, VIR_ERR_OPERATION_DENIED, __FUNCTION__); + /* NB, deliberately report error against source object, not dest */ + virLibDomainError (domain, VIR_ERR_OPERATION_DENIED, __FUNCTION__); goto error; } /* Check that migration is supported by both drivers. */ - if (VIR_DRV_SUPPORTS_FEATURE (conn->driver, conn, + if (VIR_DRV_SUPPORTS_FEATURE (domain->conn->driver, domain->conn, VIR_DRV_FEATURE_MIGRATION_V1) && VIR_DRV_SUPPORTS_FEATURE (dconn->driver, dconn, VIR_DRV_FEATURE_MIGRATION_V1)) - version = 1; - else if (VIR_DRV_SUPPORTS_FEATURE (conn->driver, conn, + ddomain = virDomainMigrateVersion1 (domain, dconn, flags, dname, uri, bandwidth); + else if (VIR_DRV_SUPPORTS_FEATURE (domain->conn->driver, domain->conn, VIR_DRV_FEATURE_MIGRATION_V2) && VIR_DRV_SUPPORTS_FEATURE (dconn->driver, dconn, VIR_DRV_FEATURE_MIGRATION_V2)) - version = 2; + ddomain = virDomainMigrateVersion2 (domain, dconn, flags, dname, uri, bandwidth); else { - virLibConnError (conn, VIR_ERR_NO_SUPPORT, __FUNCTION__); + virLibConnError (domain->conn, VIR_ERR_NO_SUPPORT, __FUNCTION__); goto error; } - /* Prepare the migration. - * - * The destination host may return a cookie, or leave cookie as - * NULL. - * - * The destination host MUST set uri_out if uri_in is NULL. - * - * If uri_in is non-NULL, then the destination host may modify - * the URI by setting uri_out. If it does not wish to modify - * the URI, it should leave uri_out as NULL. - */ - if (version == 1) { - ret = dconn->driver->domainMigratePrepare - (dconn, &cookie, &cookielen, uri, &uri_out, flags, dname, - bandwidth); - if (ret == -1) goto done; - if (uri == NULL && uri_out == NULL) { - virLibConnError (conn, VIR_ERR_INTERNAL_ERROR, - _("domainMigratePrepare did not set uri")); - goto done; - } - if (uri_out) uri = uri_out; /* Did domainMigratePrepare change URI? */ - - assert (uri != NULL); - } - else /* if (version == 2) */ { - /* 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 (!conn->driver->domainDumpXML) { - virLibConnError (conn, VIR_ERR_INTERNAL_ERROR, __FUNCTION__); - goto error; - } - dom_xml = conn->driver->domainDumpXML (domain, - VIR_DOMAIN_XML_SECURE); - - if (!dom_xml) - goto error; - - ret = dconn->driver->domainMigratePrepare2 - (dconn, &cookie, &cookielen, uri, &uri_out, flags, dname, - bandwidth, dom_xml); - VIR_FREE (dom_xml); - if (ret == -1) goto done; - if (uri == NULL && uri_out == NULL) { - virLibConnError (conn, VIR_ERR_INTERNAL_ERROR, - _("domainMigratePrepare2 did not set uri")); - goto done; - } - if (uri_out) uri = uri_out; /* Did domainMigratePrepare2 change URI? */ + if (ddomain == NULL) + goto error; - assert (uri != NULL); - } - - /* Perform the migration. The driver isn't supposed to return - * until the migration is complete. - */ - ret = conn->driver->domainMigratePerform - (domain, cookie, cookielen, uri, flags, dname, bandwidth); - - if (version == 1) { - if (ret == -1) goto done; - /* Get the destination domain and return it or error. - * 'domain' no longer actually exists at this point - * (or so we hope), but we still use the object in memory - * in order to get the name. - */ - dname = dname ? dname : domain->name; - if (dconn->driver->domainMigrateFinish) - ddomain = dconn->driver->domainMigrateFinish - (dconn, dname, cookie, cookielen, uri, flags); - else - ddomain = virDomainLookupByName (dconn, dname); - } else /* if (version == 2) */ { - /* 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. - */ - dname = dname ? dname : domain->name; - ddomain = dconn->driver->domainMigrateFinish2 - (dconn, dname, cookie, cookielen, uri, flags, ret); - } - - done: - VIR_FREE (uri_out); - VIR_FREE (cookie); return ddomain; error: -- 1.6.0.6

2 2

[libvirt] [PATCH] Fix QEMU domain status after restore.
by Chris Lalancette 11 Aug '09

11 Aug '09

When doing a restore, we were forgetting to update the state file for the VM. That means that if you do a save/restore, then shut down libvirtd, then start it back up, you'll see the state of the guest as "paused", even though it is really running. We were just forgetting a "virDomainSaveStatus" call in the restor path. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/qemu_driver.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/qemu_driver.c b/src/qemu_driver.c index 1fb2417..d789c4d 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -4099,6 +4099,7 @@ static int qemudDomainRestore(virConnectPtr conn, } VIR_FREE(info); vm->state = VIR_DOMAIN_RUNNING; + virDomainSaveStatus(conn, driver->stateDir, vm); } ret = 0; -- 1.6.0.6

2 2

[libvirt] [PATCH] Run 'cont' on successful migration finish.
by Chris Lalancette 11 Aug '09

11 Aug '09

As of qemu 0.10.6, qemu now honors the -S flag on incoming migration. That means that when the migration completes, we have to issue a 'cont' command to get the VM running again. We do it unconditionally since it won't hurt on older qemu. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/qemu_driver.c | 11 +++++++++++ 1 files changed, 11 insertions(+), 0 deletions(-) diff --git a/src/qemu_driver.c b/src/qemu_driver.c index bcacd41..1fb2417 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -6600,7 +6600,18 @@ qemudDomainMigrateFinish2 (virConnectPtr dconn, */ if (retcode == 0) { dom = virGetDomain (dconn, vm->def->name, vm->def->uuid); + + /* 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 (qemudMonitorCommand(vm, "cont", &info) < 0) { + qemudReportError(dconn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + "%s", _("resume operation failed")); + goto cleanup; + } VIR_FREE(info); + vm->state = VIR_DOMAIN_RUNNING; event = virDomainEventNewFromObj(vm, VIR_DOMAIN_EVENT_RESUMED, -- 1.6.0.6

2 2

[libvirt] [PATCH] Fix up a stray whitespace in virHashGrow.
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/hash.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/hash.c b/src/hash.c index bde3a0b..9308c0c 100644 --- a/src/hash.c +++ b/src/hash.c @@ -141,7 +141,7 @@ virHashGrow(virHashTablePtr table, int size) } table->size = size; - /* If the two loops are merged, there would be situations where + /* If the two loops are merged, there would be situations where * a new entry needs to allocated and data copied into it from * the main table. So instead, we run through the array twice, first * copying all the elements in the main array (where we can't get -- 1.6.0.6

2 2

[libvirt] [PATCH] Fix up a whitespace in comments in src/console.c
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/console.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/console.c b/src/console.c index badb62c..4a1a80b 100644 --- a/src/console.c +++ b/src/console.c @@ -78,7 +78,7 @@ int vshRunConsole(const char *tty) { return -1; } - /* Put STDIN into raw mode so that stuff typed + /* Put STDIN into raw mode so that stuff typed does not echo to the screen (the TTY reads will result in it being echoed back already), and also ensure Ctrl-C, etc is blocked, and misc -- 1.6.0.6

2 2

[libvirt] [PATCH] Minor cleanup of error path for c_oneVmInfo.
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/opennebula/one_client.c | 16 ++++++---------- 1 files changed, 6 insertions(+), 10 deletions(-) diff --git a/src/opennebula/one_client.c b/src/opennebula/one_client.c index d1641bf..21f303a 100644 --- a/src/opennebula/one_client.c +++ b/src/opennebula/one_client.c @@ -176,6 +176,7 @@ int c_oneVmInfo(int vmid, char* ret_info,int length) xmlrpc_value *resultP; int return_code; char *return_string; + int retval = -1; resultP = xmlrpc_client_call(&one_client.env, one_client.url, "one.vmget_info", "(si)", one_client.session, vmid); @@ -188,18 +189,13 @@ int c_oneVmInfo(int vmid, char* ret_info,int length) strncpy(ret_info, return_string, length-1); ret_info[length-1] = '\0'; - xmlrpc_DECREF(resultP); - free(return_string); - - return 0; + retval = 0; } - else - { - xmlrpc_DECREF(resultP); - free(return_string); - return -1; - } + xmlrpc_DECREF(resultP); + free(return_string); + + return retval; } void c_oneFree() -- 1.6.0.6

2 2

[libvirt] [PATCH] Make openvzGetVPSUUID take a len.
by Chris Lalancette 11 Aug '09

11 Aug '09

Minor fix to openvzGetVPSUUID to make it take a length parameter. This ensures that it doesn't make assumptions about the length of the UUID buffer, and paves the way for removal of strncpy in the future. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/openvz_conf.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/openvz_conf.c b/src/openvz_conf.c index 6e9af67..a172fe3 100644 --- a/src/openvz_conf.c +++ b/src/openvz_conf.c @@ -54,7 +54,7 @@ #define VIR_FROM_THIS VIR_FROM_OPENVZ static char *openvzLocateConfDir(void); -static int openvzGetVPSUUID(int vpsid, char *uuidstr); +static int openvzGetVPSUUID(int vpsid, char *uuidstr, size_t len); static int openvzLocateConfFile(int vpsid, char *conffile, int maxlen, const char *ext); static int openvzAssignUUIDs(void); @@ -469,7 +469,7 @@ int openvzLoadDomains(struct openvz_driver *driver) { if (virAsprintf(&dom->def->name, "%i", veid) < 0) goto no_memory; - openvzGetVPSUUID(veid, uuidstr); + openvzGetVPSUUID(veid, uuidstr, sizeof(uuidstr)); ret = virUUIDParse(uuidstr, dom->def->uuid); if (ret == -1) { @@ -805,7 +805,7 @@ openvz_readline(int fd, char *ptr, int maxlen) } static int -openvzGetVPSUUID(int vpsid, char *uuidstr) +openvzGetVPSUUID(int vpsid, char *uuidstr, size_t len) { char conf_file[PATH_MAX]; char line[1024]; @@ -832,7 +832,7 @@ openvzGetVPSUUID(int vpsid, char *uuidstr) sscanf(line, "%s %s\n", iden, uuidbuf); if(STREQ(iden, "#UUID:")) { - strncpy(uuidstr, uuidbuf, VIR_UUID_STRING_BUFLEN); + strncpy(uuidstr, uuidbuf, len); break; } } @@ -856,7 +856,7 @@ openvzSetDefinedUUID(int vpsid, unsigned char *uuid) if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX, "conf")<0) return -1; - if (openvzGetVPSUUID(vpsid, uuidstr)) + if (openvzGetVPSUUID(vpsid, uuidstr, sizeof(uuidstr))) return -1; if (uuidstr[0] == 0) { -- 1.6.0.6

2 2

[libvirt] [PATCH] Fix phyp escape_specialcharacters.
by Chris Lalancette 11 Aug '09

11 Aug '09

A couple of minor fixes to phyp escape_specialcharacters. Make it a static function (since it's only used in phyp/phyp_driver.c), and make it take a dstlen parameter. This paves the way for removing strncpy in the future. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/phyp/phyp_driver.c | 10 ++++++---- src/phyp/phyp_driver.h | 2 -- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/src/phyp/phyp_driver.c b/src/phyp/phyp_driver.c index cbfd31b..f457cf4 100644 --- a/src/phyp/phyp_driver.c +++ b/src/phyp/phyp_driver.c @@ -53,6 +53,8 @@ #define VIR_FROM_THIS VIR_FROM_PHYP +static int escape_specialcharacters(char *src, char *dst, size_t dstlen); + /* * URI: phyp://user@[hmc|ivm]/managed_system * */ @@ -94,7 +96,7 @@ phypOpen(virConnectPtr conn, return VIR_DRV_OPEN_ERROR; } - if (escape_specialcharacters(conn->uri->path, string) == -1) { + if (escape_specialcharacters(conn->uri->path, string, sizeof(string)) == -1) { virRaiseError(conn, NULL, NULL, 0, VIR_FROM_PHYP, VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, "%s", _("Error parsing 'path'. Invalid characters.")); @@ -1341,8 +1343,8 @@ init_uuid_db(virConnectPtr conn) return; } -int -escape_specialcharacters(char *src, char *dst) +static int +escape_specialcharacters(char *src, char *dst, size_t dstlen) { size_t len = strlen(src); char temp_buffer[len]; @@ -1367,7 +1369,7 @@ escape_specialcharacters(char *src, char *dst) } temp_buffer[j] = '\0'; - if (strncpy(dst, temp_buffer, j) == NULL) + if (strncpy(dst, temp_buffer, dstlen) == NULL) return -1; return 0; diff --git a/src/phyp/phyp_driver.h b/src/phyp/phyp_driver.h index fd824b3..f16b6fe 100644 --- a/src/phyp/phyp_driver.h +++ b/src/phyp/phyp_driver.h @@ -62,5 +62,3 @@ char *phypGetBackingDevice(virConnectPtr conn, const char *managed_system, int phypDiskType(virConnectPtr conn, char *backing_device); SSH_SESSION *openSSHSession(virConnectPtr conn, virConnectAuthPtr auth); - -int escape_specialcharacters(char *src, char *dst); -- 1.6.0.6

2 2

[libvirt] [PATCH] Fix up a few minor indentation issues.
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/xen_internal.c | 8 ++++---- src/xend_internal.c | 2 +- src/xm_internal.c | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/src/xen_internal.c b/src/xen_internal.c index b6bf10d..ae78f84 100644 --- a/src/xen_internal.c +++ b/src/xen_internal.c @@ -1147,7 +1147,7 @@ static const char *str_cap = "cap"; */ int xenHypervisorGetSchedulerParameters(virDomainPtr domain, - virSchedParameterPtr params, int *nparams) + virSchedParameterPtr params, int *nparams) { xenUnifiedPrivatePtr priv; @@ -1209,19 +1209,19 @@ xenHypervisorGetSchedulerParameters(virDomainPtr domain, return(-1); strncpy (params[0].field, str_weight, VIR_DOMAIN_SCHED_FIELD_LENGTH); - params[0].field[VIR_DOMAIN_SCHED_FIELD_LENGTH-1] = '\0'; + params[0].field[VIR_DOMAIN_SCHED_FIELD_LENGTH-1] = '\0'; params[0].type = VIR_DOMAIN_SCHED_FIELD_UINT; params[0].value.ui = op_dom.u.getschedinfo.u.credit.weight; strncpy (params[1].field, str_cap, VIR_DOMAIN_SCHED_FIELD_LENGTH); - params[1].field[VIR_DOMAIN_SCHED_FIELD_LENGTH-1] = '\0'; + params[1].field[VIR_DOMAIN_SCHED_FIELD_LENGTH-1] = '\0'; params[1].type = VIR_DOMAIN_SCHED_FIELD_UINT; params[1].value.ui = op_dom.u.getschedinfo.u.credit.cap; *nparams = 2; break; default: - virXenErrorFunc(domain->conn, VIR_ERR_INVALID_ARG, __FUNCTION__, + virXenErrorFunc(domain->conn, VIR_ERR_INVALID_ARG, __FUNCTION__, "Unknown scheduler", op_sys.u.getschedulerid.sched_id); return -1; } diff --git a/src/xend_internal.c b/src/xend_internal.c index 2585703..7bcee7d 100644 --- a/src/xend_internal.c +++ b/src/xend_internal.c @@ -4732,7 +4732,7 @@ static const char *str_cap = "cap"; */ static int xenDaemonGetSchedulerParameters(virDomainPtr domain, - virSchedParameterPtr params, int *nparams) + virSchedParameterPtr params, int *nparams) { xenUnifiedPrivatePtr priv; struct sexpr *root; diff --git a/src/xm_internal.c b/src/xm_internal.c index c869120..b7f9e67 100644 --- a/src/xm_internal.c +++ b/src/xm_internal.c @@ -1418,7 +1418,7 @@ xenXMDomainConfigParse(virConnectPtr conn, virConfPtr conf) { no_memory: virReportOOMError(conn); /* fallthrough */ - cleanup: +cleanup: virDomainGraphicsDefFree(graphics); virDomainNetDefFree(net); virDomainDiskDefFree(disk); -- 1.6.0.6

2 2

[libvirt] [PATCH] Remove a duplicated assignment in Xen PCI parsing.
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/xm_internal.c | 1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/src/xm_internal.c b/src/xm_internal.c index b7f9e67..71b852e 100644 --- a/src/xm_internal.c +++ b/src/xm_internal.c @@ -1160,7 +1160,6 @@ xenXMDomainConfigParse(virConnectPtr conn, virConfPtr conf) { goto skippci; /* pci=['0000:00:1b.0','0000:00:13.0'] */ - key = list->str; if (!(key = list->str)) goto skippci; if (!(nextkey = strchr(key, ':'))) -- 1.6.0.6

2 2

[libvirt] vbox save/restore
by Lucian Adrian Grijincu 11 Aug '09

11 Aug '09

Hello, I see that the vbox driver in virDomainSave doesn't use the destination file path and calls vbox's IConsole::SaveState. As I see it IConsole::SaveState is more like a "pause" command: you can save the state you're in and wait for the machine to stop running and you can resume it later. This limits the uses of the vbox support of libvirt. There is another way you can do virDomainSave/virDomainRestore: use snapshots. This way, you can specify the file in which you want to save the snapshot, you can continue to use the machine and do some crazy experiments and then restore the machine to the known good state. As of the current implementation, I don't think there's a way to do this in libvirt. I'd like to know if there's a real reason why virDomainSave is implemented this way (and virDomainRestore is unimplemented), or it just got implemented like this and no one cared much about it. -- . ..: Lucian

3 4

[libvirt] [PATCH] Compressed save image format for Qemu.
by Chris Lalancette 11 Aug '09

11 Aug '09

Implement a compressed save image format for qemu. While ideally we would have the choice between compressed/non-compressed available to the libvirt API, unfortunately there is no "flags" parameter to the virDomainSave() API. Therefore, implement this as a qemu.conf option. Both gzip and bzip2 are implemented, and it should be very easy to implement additional compression methods. One open question is if/how we should detect the gzip and bzip2 binaries. One way to do it is to do compile-time setting of the paths (via configure.in), but that doesn't seem like a great thing to do. Another solution (my preferred solution) is not to detect at all; when we go to run the commands that need them, if they aren't available, or aren't available in one of the standard paths, then we'll fail. Maybe somebody else has another option or opinion, though. In the future, we'll have a more robust (managed) save/restore API, at which time we can expose this functionality properly in the API. V2: get rid of redundant dd command and just use >> to append data. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/qemu.conf | 10 +++++++ src/qemu_conf.c | 11 ++++++++ src/qemu_conf.h | 2 + src/qemu_driver.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++++--- 4 files changed, 90 insertions(+), 4 deletions(-) diff --git a/src/qemu.conf b/src/qemu.conf index 653f487..86dcc9d 100644 --- a/src/qemu.conf +++ b/src/qemu.conf @@ -129,3 +129,13 @@ # "/dev/ptmx", "/dev/kvm", "/dev/kqemu", # "/dev/rtc", "/dev/hpet", "/dev/net/tun", #] + +# The default format for Qemu/KVM guest save images is raw; that is, the +# memory from the domain is dumped out directly to a file. If you have +# guests with a large amount of memory, however, this can take up quite +# a bit of space. If you would like to compress the images while they +# are being saved to disk, you can also set "gzip" or "bzip2" for the +# save_image_format. Note that this means you slow down the process +# of saving a domain in order to save disk space. +# +# save_image_format = "raw" diff --git a/src/qemu_conf.c b/src/qemu_conf.c index 7ca5a15..ed87e13 100644 --- a/src/qemu_conf.c +++ b/src/qemu_conf.c @@ -280,6 +280,17 @@ int qemudLoadDriverConfig(struct qemud_driver *driver, driver->cgroupDeviceACL[i] = NULL; } + p = virConfGetValue (conf, "save_image_format"); + CHECK_TYPE ("save_image_format", VIR_CONF_STRING); + if (p && p->str) { + VIR_FREE(driver->saveImageFormat); + if (!(driver->saveImageFormat = strdup(p->str))) { + virReportOOMError(NULL); + virConfFree(conf); + return -1; + } + } + virConfFree (conf); return 0; } diff --git a/src/qemu_conf.h b/src/qemu_conf.h index 8f4ef6a..e34baab 100644 --- a/src/qemu_conf.h +++ b/src/qemu_conf.h @@ -111,6 +111,8 @@ struct qemud_driver { char *securityDriverName; virSecurityDriverPtr securityDriver; + + char *saveImageFormat; }; diff --git a/src/qemu_driver.c b/src/qemu_driver.c index 3c92635..b146330 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -3437,11 +3437,26 @@ static int qemudDomainSave(virDomainPtr dom, struct qemud_save_header header; int ret = -1; virDomainEventPtr event = NULL; + int internalret; memset(&header, 0, sizeof(header)); memcpy(header.magic, QEMUD_SAVE_MAGIC, sizeof(header.magic)); header.version = QEMUD_SAVE_VERSION; + if (driver->saveImageFormat == NULL) + header.compressed = QEMUD_SAVE_FORMAT_RAW; + else if (STREQ(driver->saveImageFormat, "raw")) + header.compressed = QEMUD_SAVE_FORMAT_RAW; + else if (STREQ(driver->saveImageFormat, "gzip")) + header.compressed = QEMUD_SAVE_FORMAT_GZIP; + else if (STREQ(driver->saveImageFormat, "bzip2")) + header.compressed = QEMUD_SAVE_FORMAT_BZIP2; + else { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + "%s", _("Invalid save image format specified in configuration file")); + return -1; + } + qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); @@ -3514,11 +3529,25 @@ static int qemudDomainSave(virDomainPtr dom, virReportOOMError(dom->conn); goto cleanup; } - if (virAsprintf(&command, "migrate \"exec:" - "dd of='%s' oflag=append conv=notrunc 2>/dev/null" - "\"", safe_path) == -1) { + + if (header.compressed == QEMUD_SAVE_FORMAT_RAW) + internalret = virAsprintf(&command, "migrate \"exec:" + "dd of='%s' oflag=append conv=notrunc 2>/dev/null" + "\"", safe_path); + else if (header.compressed == QEMUD_SAVE_FORMAT_GZIP) + internalret = virAsprintf(&command, "migrate \"exec:" + "gzip -c >> '%s' 2>/dev/null\"", safe_path); + else if (header.compressed == QEMUD_SAVE_FORMAT_BZIP2) + internalret = virAsprintf(&command, "migrate \"exec:" + "bzip2 -c >> '%s' 2>/dev/null\"", safe_path); + else { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, + _("Invalid compress format %d"), + header.compressed); + goto cleanup; + } + if (internalret < 0) { virReportOOMError(dom->conn); - command = NULL; goto cleanup; } @@ -4039,6 +4068,9 @@ static int qemudDomainRestore(virConnectPtr conn, char *xml = NULL; struct qemud_save_header header; virDomainEventPtr event = NULL; + int intermediatefd = -1; + pid_t intermediate_pid = -1; + int childstat; qemuDriverLock(driver); /* Verify the header and read the XML */ @@ -4128,8 +4160,39 @@ static int qemudDomainRestore(virConnectPtr conn, } def = NULL; + if (header.version == 2) { + const char *intermediate_argv[3] = { NULL, "-dc", NULL }; + if (header.compressed == QEMUD_SAVE_FORMAT_GZIP) + intermediate_argv[0] = "gzip"; + else if (header.compressed == QEMUD_SAVE_FORMAT_BZIP2) + intermediate_argv[0] = "bzip2"; + else if (header.compressed != QEMUD_SAVE_FORMAT_RAW) { + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("Unknown compressed save format %d"), + header.compressed); + goto cleanup; + } + if (intermediate_argv[0] != NULL) { + intermediatefd = fd; + fd = -1; + if (virExec(conn, intermediate_argv, NULL, NULL, + &intermediate_pid, intermediatefd, &fd, NULL, 0) < 0) { + qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + _("Failed to start decompression binary %s"), + intermediate_argv[0]); + goto cleanup; + } + } + } /* Set the migration source and start it up. */ ret = qemudStartVMDaemon(conn, driver, vm, "stdio", fd); + if (intermediate_pid != -1) { + /* Wait for intermediate process to exit */ + while (waitpid(intermediate_pid, &childstat, 0) == -1 && + errno == EINTR); + } + if (intermediatefd != -1) + close(intermediatefd); close(fd); fd = -1; if (ret < 0) { -- 1.6.0.6

2 2

[libvirt] [PATCH] Remove use of strncpy in qemudExtractMonitorPath.
by Chris Lalancette 11 Aug '09

11 Aug '09

qemudExtractMonitorPath() was doing a VIR_ALLOC_N followed by a strncpy. However, this isn't necessary; we can do the same thing using virAsprintf(), which is much safer. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/qemu_driver.c | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/src/qemu_driver.c b/src/qemu_driver.c index 55a09f5..37fdec2 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -1017,12 +1017,13 @@ qemudExtractMonitorPath(virConnectPtr conn, */ while (*tmp) { if (c_isspace(*tmp)) { - if (VIR_ALLOC_N(*path, (tmp-dev)+1) < 0) { + if (virAsprintf(path, "%s", dev) < 0) { virReportOOMError(conn); return -1; } - strncpy(*path, dev, (tmp-dev)); - (*path)[(tmp-dev)] = '\0'; + /* the last character is a \n, so back up one to overwrite */ + (*path)[tmp-dev] = '\0'; + /* ... now further update offset till we get EOL */ *offset = tmp - haystack; return 0; -- 1.6.0.6

2 1

[libvirt] [PATCH] Convert one use of strncpy + strcat to snprintf in test driver.
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/test.c | 5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-) diff --git a/src/test.c b/src/test.c index 470816b..ab6cfdf 100644 --- a/src/test.c +++ b/src/test.c @@ -417,9 +417,8 @@ static char *testBuildFilename(const char *relativeTo, char *absFile; if (VIR_ALLOC_N(absFile, baseLen + strlen(filename) + 1) < 0) return NULL; - strncpy(absFile, relativeTo, baseLen); - absFile[baseLen] = '\0'; - strcat(absFile, filename); + snprintf(absFile, baseLen + strlen(filename) + 1, "%s%s", + relativeTo, filename); return absFile; } else { return strdup(filename); -- 1.6.0.6

2 1

[libvirt] [PATCH] Error path mutex unlock in virGetStorageVol.
by Chris Lalancette 11 Aug '09

11 Aug '09

Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/datatypes.c | 5 +---- 1 files changed, 1 insertions(+), 4 deletions(-) diff --git a/src/datatypes.c b/src/datatypes.c index 1889dc3..ac61682 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -863,19 +863,16 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c ret = (virStorageVolPtr) virHashLookup(conn->storageVols, key); if (ret == NULL) { if (VIR_ALLOC(ret) < 0) { - virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->pool = strdup(pool); if (ret->pool == NULL) { - virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->name = strdup(name); if (ret->name == NULL) { - virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } @@ -885,7 +882,6 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c ret->conn = conn; if (virHashAddEntry(conn->storageVols, key, ret) < 0) { - virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, "%s", _("failed to add storage vol to connection hash table")); goto error; @@ -897,6 +893,7 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c return(ret); error: + virMutexUnlock(&conn->lock); if (ret != NULL) { VIR_FREE(ret->name); VIR_FREE(ret->pool); -- 1.6.0.6

2 1

[libvirt] Crash while using virConnectRef
by Chris Lalancette 11 Aug '09

11 Aug '09

All, I've been doing some testing of the virConnectRef() function, and as far as I can tell, it's busted. Here's the simple example program I'm using: #include <stdio.h> #include <stdlib.h> #include <libvirt/libvirt.h> #include <unistd.h> int main(int argc, char *argv[]) { virConnectPtr conn; conn = virConnectOpen(NULL); printf("type is %s\n",virConnectGetType(conn)); virConnectRef(conn); virConnectClose(conn); virConnectClose(conn); return 0; } When compiling and running this program, I get a segfault during the *second* virConnectClose(). The crash comes in the remote_internal driver, because the first virConnectClose() actually freed up the remote_internal private memory, and now the second one comes along and accesses a NULL pointer. I think this actually points to a bug in virConnectClose(); we shouldn't be running the ->close() callbacks on the drivers until we are going to be really closing up the driver, i.e. we should move all of the ->close() callbacks into virUnrefConnect(). Is that the right way to fix this? -- Chris Lalancette

2 1

[libvirt] [PATCH] Handle kernels with no ipv6 support
by Mark McLoughlin 11 Aug '09

11 Aug '09

If the ipv6 kernel module is not loaded, then we get this when starting a virtual network: libvir: Network Config error : cannot enable /proc/sys/net/ipv6/conf/virbr0/disable_ipv6: No such file or directory If disable_ipv6 is not present, we should just merrily continue on our way. * src/network_driver.c: make networkDisableIPV6() not fail if the kernel has no ipv6 support --- src/network_driver.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/src/network_driver.c b/src/network_driver.c index eaea454..84910ab 100644 --- a/src/network_driver.c +++ b/src/network_driver.c @@ -801,6 +801,12 @@ static int networkDisableIPV6(virConnectPtr conn, goto cleanup; } + if (access(field, W_OK) < 0 && errno == ENOENT) { + VIR_DEBUG("ipv6 appears to already be disabled on %s", network->def->bridge); + ret = 0; + goto cleanup; + } + if (virFileWriteStr(field, "1") < 0) { virReportSystemError(conn, errno, _("cannot enable %s"), field); -- 1.6.2.5

3 2

[libvirt] does libvirt support qemu-kvm vmchannel option?
by cyliu7 11 Aug '09

11 Aug '09

Hi all, I noticed that qemu-kvm support an option vmchannel (-net channel ...) to create a new device to communicate between host and guest. does libvirt support this option when creating a KVM based VM. if so, how to config the VM? is there any sample to show this feature? any suggestions are greatly appreciated. Regards, ------------------------ Chunyang Liu

2 1

[libvirt] PATCH: Support PolicyKit 1.0
by Daniel P. Berrange 10 Aug '09

10 Aug '09

In the seriously annoying way of things, the newest PolicyKit in Fedora 12 has been completely re-written from scratch with a totally incompatible application facing API. Conceptually it is still pretty similar though, with the exception that client applications no longer need to explicitly launch an auth dialog - that's done out-of-band by policykit itself. This patch adjusts libvirtd to the new API, and removes the libvirt client side code that spawned the auth helper. On the libvirtd side avoid their APIs and instead spawn an external auth checking program 'pkcheck', which returns 0 on success, non-0 for denial. In the final annoying bit, the XML format for the policy has remained with exactly the same DTD version, except that it has quietly changed the allowed values for some attributes in an incompatible manner. So I have to add a new policy file too. NB, it may not look like we've changed the client side, but we have, since the #ifdef for the external auth agent is no longer set. Tested with old policy kit, and new policykit, and with none at all. b/configure.in | 73 +++++++++++++++++++++++++++-------------- b/qemud/Makefile.am | 11 +++++- b/qemud/libvirtd.policy-0 | 42 +++++++++++++++++++++++ b/qemud/libvirtd.policy-1 | 42 +++++++++++++++++++++++ b/qemud/qemud.c | 4 +- b/qemud/qemud.h | 4 +- b/qemud/remote.c | 81 +++++++++++++++++++++++++++++++++++++++++++--- qemud/libvirtd.policy | 42 ----------------------- 8 files changed, 223 insertions(+), 76 deletions(-) Regards, Daniel diff --git a/configure.in b/configure.in index b905b23..77d6c9e 100644 --- a/configure.in +++ b/configure.in @@ -607,40 +607,61 @@ AC_SUBST([SASL_LIBS]) dnl PolicyKit library POLKIT_CFLAGS= POLKIT_LIBS= +PKCHECK_PATH= AC_ARG_WITH([polkit], [ --with-polkit use PolicyKit for UNIX socket access checks], [], [with_polkit=check]) +with_polkit0=no +with_polkit1=no if test "x$with_polkit" = "xyes" -o "x$with_polkit" = "xcheck"; then - PKG_CHECK_MODULES(POLKIT, polkit-dbus >= $POLKIT_REQUIRED, - [with_polkit=yes], [ - if test "x$with_polkit" = "xcheck" ; then - with_polkit=no - else - AC_MSG_ERROR( - [You must install PolicyKit >= $POLKIT_REQUIRED to compile libvirt]) - fi - ]) - if test "x$with_polkit" = "xyes" ; then + dnl Check for new polkit first - just a binary + AC_PATH_PROG([PKCHECK_PATH],[pkcheck], [], [/usr/sbin:$PATH]) + if test "x$PKCHECK_PATH" != "x" ; then + AC_DEFINE_UNQUOTED([PKCHECK_PATH],["$PKCHECK_PATH"],[Location of pkcheck program]) AC_DEFINE_UNQUOTED([HAVE_POLKIT], 1, - [use PolicyKit for UNIX socket access checks]) - - old_CFLAGS=$CFLAGS - old_LDFLAGS=$LDFLAGS - CFLAGS="$CFLAGS $POLKIT_CFLAGS" - LDFLAGS="$LDFLAGS $POLKIT_LIBS" - AC_CHECK_FUNCS([polkit_context_is_caller_authorized]) - CFLAGS="$old_CFLAGS" - LDFLAGS="$old_LDFLAGS" - - AC_PATH_PROG([POLKIT_AUTH], [polkit-auth]) - if test "x$POLKIT_AUTH" != "x"; then - AC_DEFINE_UNQUOTED([POLKIT_AUTH],["$POLKIT_AUTH"],[Location of polkit-auth program]) + [use PolicyKit for UNIX socket access checks]) + AC_DEFINE_UNQUOTED([HAVE_POLKIT1], 1, + [use PolicyKit for UNIX socket access checks]) + with_polkit="yes" + with_polkit1="yes" + else + dnl Check for old polkit second - library + binary + PKG_CHECK_MODULES(POLKIT, polkit-dbus >= $POLKIT_REQUIRED, + [with_polkit=yes], [ + if test "x$with_polkit" = "xcheck" ; then + with_polkit=no + else + AC_MSG_ERROR( + [You must install PolicyKit >= $POLKIT_REQUIRED to compile libvirt]) + fi + ]) + if test "x$with_polkit" = "xyes" ; then + AC_DEFINE_UNQUOTED([HAVE_POLKIT], 1, + [use PolicyKit for UNIX socket access checks]) + AC_DEFINE_UNQUOTED([HAVE_POLKIT0], 1, + [use PolicyKit for UNIX socket access checks]) + + old_CFLAGS=$CFLAGS + old_LDFLAGS=$LDFLAGS + CFLAGS="$CFLAGS $POLKIT_CFLAGS" + LDFLAGS="$LDFLAGS $POLKIT_LIBS" + AC_CHECK_FUNCS([polkit_context_is_caller_authorized]) + CFLAGS="$old_CFLAGS" + LDFLAGS="$old_LDFLAGS" + + AC_PATH_PROG([POLKIT_AUTH], [polkit-auth]) + if test "x$POLKIT_AUTH" != "x"; then + AC_DEFINE_UNQUOTED([POLKIT_AUTH],["$POLKIT_AUTH"],[Location of polkit-auth program]) + fi + with_polkit0="yes" fi fi fi AM_CONDITIONAL([HAVE_POLKIT], [test "x$with_polkit" = "xyes"]) +AM_CONDITIONAL([HAVE_POLKIT0], [test "x$with_polkit0" = "xyes"]) +AM_CONDITIONAL([HAVE_POLKIT1], [test "x$with_polkit1" = "xyes"]) AC_SUBST([POLKIT_CFLAGS]) AC_SUBST([POLKIT_LIBS]) @@ -1621,7 +1642,11 @@ else AC_MSG_NOTICE([ avahi: no]) fi if test "$with_polkit" = "yes" ; then -AC_MSG_NOTICE([ polkit: $POLKIT_CFLAGS $POLKIT_LIBS]) +if test "$with_polkit0" = "yes" ; then +AC_MSG_NOTICE([ polkit: $POLKIT_CFLAGS $POLKIT_LIBS (version 0)]) +else +AC_MSG_NOTICE([ polkit: $PKCHECK_PATH (version 1)]) +fi else AC_MSG_NOTICE([ polkit: no]) fi diff --git a/qemud/Makefile.am b/qemud/Makefile.am index 959ff88..3d143da 100644 --- a/qemud/Makefile.am +++ b/qemud/Makefile.am @@ -21,7 +21,8 @@ EXTRA_DIST = \ remote_protocol.x \ libvirtd.conf \ libvirtd.init.in \ - libvirtd.policy \ + libvirtd.policy-0 \ + libvirtd.policy-1 \ libvirtd.sasl \ libvirtd.sysconf \ libvirtd.aug \ @@ -147,7 +148,13 @@ endif libvirtd_LDADD += ../src/libvirt.la if HAVE_POLKIT +if HAVE_POLKIT0 policydir = $(datadir)/PolicyKit/policy +policyfile = libvirtd.policy-0 +else +policydir = $(datadir)/polkit-1/actions +policyfile = libvirtd.policy-1 +endif endif if HAVE_AVAHI @@ -197,7 +204,7 @@ endif if HAVE_POLKIT install-data-polkit:: install-init mkdir -p $(DESTDIR)$(policydir) - $(INSTALL_DATA) $(srcdir)/libvirtd.policy $(DESTDIR)$(policydir)/org.libvirt.unix.policy + $(INSTALL_DATA) $(srcdir)/$(policyfile) $(DESTDIR)$(policydir)/org.libvirt.unix.policy uninstall-data-polkit:: install-init rm -f $(DESTDIR)$(policydir)/org.libvirt.unix.policy else diff --git a/qemud/libvirtd.policy b/qemud/libvirtd.policy deleted file mode 100644 index b6da946..0000000 --- a/qemud/libvirtd.policy +++ /dev/null @@ -1,42 +0,0 @@ -<!DOCTYPE policyconfig PUBLIC - "-//freedesktop//DTD PolicyKit Policy Configuration 1.0//EN" - "http://www.freedesktop.org/standards/PolicyKit/1.0/policyconfig.dtd"> - - - -<policyconfig> - <action id="org.libvirt.unix.monitor"> - <description>Monitor local virtualized systems</description> - <message>System policy prevents monitoring of local virtualized systems</message> - <defaults> -  - <allow_any>yes</allow_any> - <allow_inactive>yes</allow_inactive> - <allow_active>yes</allow_active> - </defaults> - </action> - - <action id="org.libvirt.unix.manage"> - <description>Manage local virtualized systems</description> - <message>System policy prevents management of local virtualized systems</message> - <defaults> -  - <allow_any>no</allow_any> - <allow_inactive>no</allow_inactive> - <allow_active>auth_admin_keep_session</allow_active> - </defaults> - </action> -</policyconfig> diff --git a/qemud/libvirtd.policy-0 b/qemud/libvirtd.policy-0 new file mode 100644 index 0000000..b6da946 --- /dev/null +++ b/qemud/libvirtd.policy-0 @@ -0,0 +1,42 @@ +<!DOCTYPE policyconfig PUBLIC + "-//freedesktop//DTD PolicyKit Policy Configuration 1.0//EN" + "http://www.freedesktop.org/standards/PolicyKit/1.0/policyconfig.dtd"> + + + +<policyconfig> + <action id="org.libvirt.unix.monitor"> + <description>Monitor local virtualized systems</description> + <message>System policy prevents monitoring of local virtualized systems</message> + <defaults> +  + <allow_any>yes</allow_any> + <allow_inactive>yes</allow_inactive> + <allow_active>yes</allow_active> + </defaults> + </action> + + <action id="org.libvirt.unix.manage"> + <description>Manage local virtualized systems</description> + <message>System policy prevents management of local virtualized systems</message> + <defaults> +  + <allow_any>no</allow_any> + <allow_inactive>no</allow_inactive> + <allow_active>auth_admin_keep_session</allow_active> + </defaults> + </action> +</policyconfig> diff --git a/qemud/libvirtd.policy-1 b/qemud/libvirtd.policy-1 new file mode 100644 index 0000000..6fa3a5e --- /dev/null +++ b/qemud/libvirtd.policy-1 @@ -0,0 +1,42 @@ +<!DOCTYPE policyconfig PUBLIC + "-//freedesktop//DTD PolicyKit Policy Configuration 1.0//EN" + "http://www.freedesktop.org/standards/PolicyKit/1.0/policyconfig.dtd"> + + + +<policyconfig> + <action id="org.libvirt.unix.monitor"> + <description>Monitor local virtualized systems</description> + <message>System policy prevents monitoring of local virtualized systems</message> + <defaults> +  + <allow_any>yes</allow_any> + <allow_inactive>yes</allow_inactive> + <allow_active>yes</allow_active> + </defaults> + </action> + + <action id="org.libvirt.unix.manage"> + <description>Manage local virtualized systems</description> + <message>System policy prevents management of local virtualized systems</message> + <defaults> +  + <allow_any>no</allow_any> + <allow_inactive>no</allow_inactive> + <allow_active>auth_admin_keep</allow_active> + </defaults> + </action> +</policyconfig> diff --git a/qemud/qemud.c b/qemud/qemud.c index 3e551ca..50b0cdd 100644 --- a/qemud/qemud.c +++ b/qemud/qemud.c @@ -895,7 +895,7 @@ static struct qemud_server *qemudNetworkInit(struct qemud_server *server) { } #endif -#ifdef HAVE_POLKIT +#if HAVE_POLKIT0 if (auth_unix_rw == REMOTE_AUTH_POLKIT || auth_unix_ro == REMOTE_AUTH_POLKIT) { DBusError derr; @@ -982,7 +982,7 @@ static struct qemud_server *qemudNetworkInit(struct qemud_server *server) { sock = sock->next; } -#ifdef HAVE_POLKIT +#if HAVE_POLKIT0 if (server->sysbus) dbus_connection_unref(server->sysbus); #endif diff --git a/qemud/qemud.h b/qemud/qemud.h index 254db44..e8ce209 100644 --- a/qemud/qemud.h +++ b/qemud/qemud.h @@ -34,7 +34,7 @@ #include <sasl/sasl.h> #endif -#ifdef HAVE_POLKIT +#if HAVE_POLKIT0 #include <dbus/dbus.h> #endif @@ -253,7 +253,7 @@ struct qemud_server { #if HAVE_SASL char **saslUsernameWhitelist; #endif -#if HAVE_POLKIT +#if HAVE_POLKIT0 DBusConnection *sysbus; #endif }; diff --git a/qemud/remote.c b/qemud/remote.c index d32d513..490a807 100644 --- a/qemud/remote.c +++ b/qemud/remote.c @@ -43,7 +43,7 @@ #include <fnmatch.h> #include "virterror_internal.h" -#ifdef HAVE_POLKIT +#if HAVE_POLKIT0 #include <polkit/polkit.h> #include <polkit-dbus/polkit-dbus.h> #endif @@ -3106,7 +3106,80 @@ remoteDispatchAuthSaslStep (struct qemud_server *server ATTRIBUTE_UNUSED, #endif /* HAVE_SASL */ -#if HAVE_POLKIT +#if HAVE_POLKIT1 +static int +remoteDispatchAuthPolkit (struct qemud_server *server, + struct qemud_client *client, + virConnectPtr conn ATTRIBUTE_UNUSED, + remote_error *rerr, + void *args ATTRIBUTE_UNUSED, + remote_auth_polkit_ret *ret) +{ + pid_t callerPid; + uid_t callerUid; + const char *action; + int status = -1; + char pidbuf[50]; + int rv; + + virMutexLock(&server->lock); + virMutexLock(&client->lock); + virMutexUnlock(&server->lock); + + action = client->readonly ? + "org.libvirt.unix.monitor" : + "org.libvirt.unix.manage"; + + const char * const pkcheck [] = { + PKCHECK_PATH, + "--action-id", action, + "--process", pidbuf, + "--allow-user-interaction", + NULL + }; + + REMOTE_DEBUG("Start PolicyKit auth %d", client->fd); + if (client->auth != REMOTE_AUTH_POLKIT) { + VIR_ERROR0(_("client tried invalid PolicyKit init request")); + goto authfail; + } + + if (qemudGetSocketIdentity(client->fd, &callerUid, &callerPid) < 0) { + VIR_ERROR0(_("cannot get peer socket identity")); + goto authfail; + } + + VIR_INFO(_("Checking PID %d running as %d"), callerPid, callerUid); + + rv = snprintf(pidbuf, sizeof pidbuf, "%d", callerPid); + if (rv < 0 || rv >= sizeof pidbuf) { + VIR_ERROR(_("Caller PID was too large %d"), callerPid); + goto authfail; + } + + if (virRun(NULL, pkcheck, &status) < 0) { + VIR_ERROR(_("Cannot invoke %s"), PKCHECK_PATH); + goto authfail; + } + if (status != 0) { + VIR_ERROR(_("Policy kit denied action %s from pid %d, uid %d, result: %d\n"), + action, callerPid, callerUid, status); + goto authfail; + } + VIR_INFO(_("Policy allowed action %s from pid %d, uid %d"), + action, callerPid, callerUid); + ret->complete = 1; + client->auth = REMOTE_AUTH_NONE; + + virMutexUnlock(&client->lock); + return 0; + +authfail: + remoteDispatchAuthError(rerr); + virMutexUnlock(&client->lock); + return -1; +} +#elif HAVE_POLKIT0 static int remoteDispatchAuthPolkit (struct qemud_server *server, struct qemud_client *client, @@ -3217,7 +3290,7 @@ authfail: return -1; } -#else /* HAVE_POLKIT */ +#else /* !HAVE_POLKIT0 & !HAVE_POLKIT1*/ static int remoteDispatchAuthPolkit (struct qemud_server *server ATTRIBUTE_UNUSED, @@ -3231,7 +3304,7 @@ remoteDispatchAuthPolkit (struct qemud_server *server ATTRIBUTE_UNUSED, remoteDispatchAuthError(rerr); return -1; } -#endif /* HAVE_POLKIT */ +#endif /* HAVE_POLKIT1 */ /*************************************************************** -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] Project announcement: abiCloud
by Diego Mariño 10 Aug '09

10 Aug '09

Hi all, This is a one-time post to announce a project which supports libvirt for managing virtual machines: abiCloud. AbiCloud is an open source cloud platform manager which allows to easily deploy a private cloud in your datacenter. One of the key differences of AbiCloud is the the web rich interface for managing the infrastructure. You can deploy a new service just dragging and dropping a VM. Currently, we use libvirt for: - Managing (deployment, configuration, administration...) of Xen & KVM virtual machines trought the java binding - Events monitoring trough a C API for openwsman We truly believe that V2V will boost the adoption of new hypervisors from users, and libvirt will have a prominent role in this space. So, we will continue adding support for more hypervisors trough libvirt. homepage: http://community.abiquo.com/display/AbiCloud Features: - Rich interface - User management through ACL - Infrastructure management (from datacenters to virtual machines) - Network management - Appliance repository - Design virtual datacenters Requirements: - Linux/Windows/MacOSX - Any hypervisor supported installed in a node - Sun Java Runtime Environment (JRE) 1.6 Licenses - CPAL 1.0 for the UI - MPL 1.1 for the services Thanks, PS: We will really appreciate if you could add us as in your website a "Third Party Libvirt Application" - Diego Mariño Co-founder & Community Manager +34 636-016-062 +1 408-386-8559 www.abiquo.com

3 4

[libvirt][PATCH] running libvirt on the latest kvm snapshot
by Anton Protopopov 10 Aug '09

10 Aug '09

Hi. Libvirt tryes to extract kvm version from the string "QEMU PC emulator version <major>.<minor>.<micro> (kvm-<version>)..." For examle, # kvm -help | head -1 QEMU PC emulator version 0.10.0 (kvm-85), Copyright (c) 2003-2008 Fabrice Bellard The problem occures if you are trying to run it on the devel version of kvm: # kvm -help | head -1 QEMU PC emulator version 0.10.0 (kvm-devel), Copyright (c) 2003-2008 Fabrice Bellard The virParseNumber() obvioulsy fails on the "devel" string here. I proposeto setup `kvm_version' to 1000 (that's enough, I think) in that case, so all checks, such as if (kvm_version > 74) ... will pass.

5 7

[libvirt] Is it possible to use more then 2 CPUs in the guest?
by Mirko Raasch 09 Aug '09

09 Aug '09

Hi, in my server i use an Intel Quad Core CPU and i want to use in one WindowsXP guest all 4 cores. The device manager shows four CPU (Qemu Virtual CPU version 0.10.5). When i start prime95 or superpi, only 2 of them are in use. CPU-Z shows also 2 Cores, instead of 4 like the device manager. How can i use all four cores in the guest? Greetings

4 10

[libvirt] ABI compliance checker
by Андрей Пономаренко 08 Aug '09

08 Aug '09

Colleagues, I'm from Institute for System Programing of Russian Academy of Sciences and we are developing a free lightweight tool for checking backward/forward binary compatibility of shared C/C++ libraries in OS Linux. It checks interface signatures and data type definitions in two library versions (headers and shared objects) and searches ABI changes that may lead to incompatibility. We have released 1.2 version of this tool and we'd like you to consider its usefulness for your project. The wiki-page with the latest release of binary compatibility checker is http://ispras.linux-foundation.org/index.php/ABI_compliance_checker Andrey Ponomarenko

1 0