- Devel - libvirt List Archives

[libvirt] PATCH: 0/5: Refactoring LXC & switching to generic domain APIs
by Daniel P. Berrange 16 Jul '08

16 Jul '08

The following series of 5 patches perform some general re-factoring of the LXC driver, the relationship between libvirtd & the processes it spawns, and finally switches LXC over to the generic domain XML APIs. This is just basically 1 day's work, so its a little rough around the edges in places, but I can confim that the core functionality is still all operational - by which I mean, start, stop, list, console, & re-start of daemon. Daniel -- |: 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 :|

3 8

[libvirt] Updating xml (cmdline)
by Stefan de Konink 16 Jul '08

16 Jul '08

Hi, Is it possible to 'update' a defined domain before it is started. I have run into a chicken and egg problem related to the <cmdline> and the first ethernet interface, I cannot create the interface before I have a domain and visa versa. Stefan

2 1

[libvirt] PATCH: Bash auto-completion for virsh
by Daniel P. Berrange 16 Jul '08

16 Jul '08

Everyone[1] seems to be doing funky bash auto-completion for commands these days, so I thought I'd make a stab at doing something for virsh. First of all I needed the command line help in an easier format to deal with, so I've added virsh _complete-command commands|arguments|options eg, $ virsh _complete-command commands help attach-device attach-disk attach-interface autostart capabilities connect console ...snip rest... $ virsh _complete-command arguments vcpupin domain vcpu cpulist $ virsh _complete-command options -c --connect -r --readonly -d --debug -h --help -q --quiet -t --timing -l --log -v --version $ virsh _complete-command options net-list --inactive --all That's enough to get all the basic commands & options auto-completing. For the arguments though I wanted to be able to auto-complete domain names, ids and UUIDs. likewise for storage pools and networks. I also wanted to auto-complete connect URIs ,eg test://default, qemu:///system $ virsh _complete-domain ISCSIServer 35e5efcc-f6a3-b489-fc33-99f2e4792e6d untangle dc502cb8-4bdd-7a15-f35a-cf76287ea2bd VirtTest 82038f2a-1344-aaf7-1a85-2a7250be2076 xenner 18c23d31-4d74-0bff-986d-e4251d2a2ff0 And $ virsh _complete-uri test:///default qemu:///system lxc:/// For this latter command, I needed to add a new API /** * virConnectURIs: * uris: pointer to a list to be filled with URI strings * * This functions returns the list of available connection * URIs. The caller is responsible for freeing the returned * strings, and the array they're stored in. * * Returns the number of URIs in case of success or -1 in * case of error. */ int virConnectURIs(char ***uris); This simply calls the existing probe() method for each driver and stores the list of URIs returned. Finally to tie this all together simply needs a little helper script which by convention is added to the /etc/bash_completion.d/. On Fedora this isn't enable by default, so you have to turn it on with source /etc/bash_completion.d/virsh To your $HOME/.bashrc Daniel [1] For a definition of 'everyone' which includes git & mercurial b/src/virsh.bash | 79 +++++++ include/libvirt/libvirt.h | 1 include/libvirt/libvirt.h.in | 1 libvirt.spec.in | 1 src/Makefile.am | 10 src/libvirt.c | 54 +++++ src/libvirt_sym.version | 2 src/test.c | 6 src/virsh.c | 442 ++++++++++++++++++++++++++++++++++++++++++- 9 files changed, 581 insertions(+), 15 deletions(-) diff -r f3413463b3ff include/libvirt/libvirt.h --- a/include/libvirt/libvirt.h Sun Jul 13 13:14:50 2008 +0100 +++ b/include/libvirt/libvirt.h Mon Jul 14 11:54:10 2008 +0100 @@ -397,6 +397,7 @@ virConnectAuthPtr auth, int flags); int virConnectClose (virConnectPtr conn); +int virConnectURIs (char ***uris); const char * virConnectGetType (virConnectPtr conn); int virConnectGetVersion (virConnectPtr conn, unsigned long *hvVer); diff -r f3413463b3ff include/libvirt/libvirt.h.in --- a/include/libvirt/libvirt.h.in Sun Jul 13 13:14:50 2008 +0100 +++ b/include/libvirt/libvirt.h.in Mon Jul 14 11:54:10 2008 +0100 @@ -397,6 +397,7 @@ virConnectAuthPtr auth, int flags); int virConnectClose (virConnectPtr conn); +int virConnectURIs (char ***uris); const char * virConnectGetType (virConnectPtr conn); int virConnectGetVersion (virConnectPtr conn, unsigned long *hvVer); diff -r f3413463b3ff libvirt.spec.in --- a/libvirt.spec.in Sun Jul 13 13:14:50 2008 +0100 +++ b/libvirt.spec.in Mon Jul 14 11:54:10 2008 +0100 @@ -227,6 +227,7 @@ %dir %attr(0700, root, root) %{_sysconfdir}/libvirt/qemu/ %dir %attr(0700, root, root) %{_sysconfdir}/libvirt/qemu/networks/ %dir %attr(0700, root, root) %{_sysconfdir}/libvirt/qemu/networks/autostart +%{_sysconfdir}/bash_completion.d/virsh %{_sysconfdir}/rc.d/init.d/libvirtd %config(noreplace) %{_sysconfdir}/sysconfig/libvirtd %config(noreplace) %{_sysconfdir}/libvirt/libvirtd.conf diff -r f3413463b3ff src/Makefile.am --- a/src/Makefile.am Sun Jul 13 13:14:50 2008 +0100 +++ b/src/Makefile.am Mon Jul 14 11:54:10 2008 +0100 @@ -25,8 +25,9 @@ confdir = $(sysconfdir)/libvirt/ conf_DATA = qemu.conf +bashdir = $(sysconfdir)/bash_completion.d -EXTRA_DIST = libvirt_sym.version $(conf_DATA) +EXTRA_DIST = libvirt_sym.version $(conf_DATA) virsh.bash lib_LTLIBRARIES = libvirt.la @@ -153,7 +154,12 @@ endif # Create the /var/cache/libvirt directory when installing. -install-exec-local: +install-exec-local:: $(MKDIR_P) $(DESTDIR)$(localstatedir)/cache/libvirt + $(MKDIR_P) $(DESTDIR)$(bashdir) + $(INSTALL_SCRIPT) virsh.bash $(bashdir)/virsh + +uninstall-local:: + $(RM) -f $(bashdir)/virsh CLEANFILES = *.gcov .libs/*.gcda .libs/*.gcno *.gcno *.gcda diff -r f3413463b3ff src/libvirt.c --- a/src/libvirt.c Sun Jul 13 13:14:50 2008 +0100 +++ b/src/libvirt.c Mon Jul 14 11:54:10 2008 +0100 @@ -37,6 +37,7 @@ #include "uuid.h" #include "util.h" #include "test.h" +#include "memory.h" #include "xen_unified.h" #include "remote_internal.h" #include "qemu_driver.h" @@ -701,7 +702,8 @@ int probes = 0; for (i = 0; i < virDriverTabCount; i++) { if ((virDriverTab[i]->probe != NULL) && - ((latest = virDriverTab[i]->probe()) != NULL)) { + ((latest = virDriverTab[i]->probe()) != NULL) && + STRNEQ(latest, "test:///default")) { probes++; DEBUG("Probed %s", latest); @@ -946,6 +948,56 @@ if (virUnrefConnect(conn) < 0) return (-1); return (0); +} + + +/** + * virConnectURIs: + * uris: pointer to a list to be filled with URI strings + * + * This functions returns the list of available connection + * URIs. The caller is responsible for freeing the returned + * strings, and the array they're stored in. + * + * Returns the number of URIs in case of success or -1 in + * case of error. + */ +int +virConnectURIs(char ***uris) +{ + const char *uri; + char **ret = NULL; + int nuris = 0, i; + DEBUG("uris=%p", uris); + + if (!initialized) + if (virInitialize() < 0) + return -1; + + if (!uris) + return -1; + + for (i = 0; i < virDriverTabCount; i++) { + if ((virDriverTab[i]->probe != NULL) && + ((uri = virDriverTab[i]->probe()) != NULL)) { + if (VIR_REALLOC_N(ret, nuris + 1) < 0) + goto no_memory; + if ((ret[nuris++] = strdup(uri)) == NULL) + goto no_memory; + + DEBUG("Probed %s", uri); + } + } + + *uris = ret; + return (nuris); + +no_memory: + for (i = 0 ; i < nuris ; i++) + VIR_FREE(ret[i]); + VIR_FREE(ret); + virLibConnError(NULL, VIR_ERR_NO_MEMORY, NULL); + return -1; } /* Not for public use. This function is part of the internal diff -r f3413463b3ff src/libvirt_sym.version --- a/src/libvirt_sym.version Sun Jul 13 13:14:50 2008 +0100 +++ b/src/libvirt_sym.version Mon Jul 14 11:54:10 2008 +0100 @@ -5,7 +5,7 @@ virConnectOpenReadOnly; virConnectOpenAuth; virConnectAuthPtrDefault; - + virConnectURIs; virConnectClose; virConnectGetType; virConnectGetVersion; diff -r f3413463b3ff src/test.c --- a/src/test.c Sun Jul 13 13:14:50 2008 +0100 +++ b/src/test.c Mon Jul 14 11:54:10 2008 +0100 @@ -215,6 +215,10 @@ " </ip>" "</network>"; + +static const char *testProbe(void) { + return "test:///default"; +} static int testOpenDefault(virConnectPtr conn) { int u; @@ -1545,7 +1549,7 @@ VIR_DRV_TEST, "Test", LIBVIR_VERSION_NUMBER, - NULL, /* probe */ + testProbe, /* probe */ testOpen, /* open */ testClose, /* close */ NULL, /* supports_feature */ diff -r f3413463b3ff src/virsh.bash --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/src/virsh.bash Mon Jul 14 11:54:10 2008 +0100 @@ -0,0 +1,79 @@ + +_virsh() +{ + local cur prev cmd cmd_index arg arg_index i results + local virsh="$1" + local canonical=0 + + COMPREPLY=() + cur="$2" + prev="$3" + + if [[ $COMP_CWORD > 1 && ${COMP_WORDS[COMP_CWORD-1]} = '--connect' ]]; then + results="$("$virsh" _complete-uri)" || nets="" + fi + + if [ -z "$results" ]; then + # searching for the command name + for ((i=1; $i<=$COMP_CWORD; i++)); do + if [[ ${COMP_WORDS[i]} != -* ]]; then + if [[ ${COMP_WORDS[i-1]} != '--connect' ]]; then + if [[ $i < $COMP_CWORD ]]; then + cmd="${COMP_WORDS[i]}" + cmd_index=$i + arg_index=`expr $COMP_CWORD - $cmd_index` + break + fi + fi + fi + done + + if [[ "$cur" == -* ]]; then + # Generate args - global or command specific + results="$("$virsh" _complete-command options "$cmd")" + else + if [ -z "$cmd" ]; then + # No command set, so generate list of all commands + results="$("$virsh" _complete-command commands)" || commands="" + else + # Command set, to generate command specific args + n=0 + for i in "$("$virsh" _complete-command arguments "$cmd")" + do + n=`expr $n + 1` + if [ $n = $arg_index ]; then + arg=$i + break + fi + done + + case $arg in + file) + COMPREPLY=(${COMPREPLY[@]:-} $(compgen -o "default" -- "$cur")) + ;; + + domain) + results="$("$virsh" _complete-domain)" || doms="" + ;; + + network) + results="$("$virsh" _complete-network)" || nets="" + ;; + + pool) + results="$("$virsh" _complete-pool)" || nets="" + ;; + + uri) + results="$("$virsh" _complete-uri)" || nets="" + ;; + esac + fi + fi + fi + if [ ! -z "$results" ]; then + COMPREPLY=(${COMPREPLY[@]:-} $(compgen -W '$results' -- "$cur")) + fi +} + +complete -F _virsh virsh diff -r f3413463b3ff src/virsh.c --- a/src/virsh.c Sun Jul 13 13:14:50 2008 +0100 +++ b/src/virsh.c Mon Jul 14 11:54:10 2008 +0100 @@ -329,6 +329,409 @@ * --------------- */ + + +/* + * "_compcmds" command + */ +static vshCmdInfo info_compcommand[] = { + {"syntax", "_complete-command [commands|arguments|options] [<command>]"}, + {"help", gettext_noop("print shell completion data")}, + {"desc", gettext_noop("Prints data useful for shell autocompletion of commands.")}, + + {NULL, NULL} +}; + +static vshCmdOptDef opts_compcommand[] = { + {"type", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("type of data")}, + {"cmd", VSH_OT_DATA, 0, gettext_noop("command name")}, + {NULL, 0, 0, NULL} +}; + +static int +cmdCompCommand(vshControl * ctl, vshCmd * cmd) +{ + const char *type = vshCommandOptString(cmd, "type", NULL); + const char *name = vshCommandOptString(cmd, "cmd", NULL); + + if (STREQ(type, "commands")) { + vshCmdDef *def; + for (def = commands; def->name; def++) + if (def->name[0] != '_') + vshPrint(ctl, "%s\n", def->name); + } else if (STREQ(type, "options")) { + if (name) { + vshCmdDef *def = vshCmddefSearch(name); + if (!def) { + vshError(ctl, FALSE, _("command '%s' doesn't exist"), name); + return FALSE; + } else if (def->opts) { + vshCmdOptDef *opt; + for (opt = def->opts; opt->name; opt++) + if (opt->type == VSH_OT_BOOL || + opt->type == VSH_OT_INT || + opt->type == VSH_OT_STRING) + vshPrint(ctl, "--%s\n", opt->name); + } + } else { + vshPrint(ctl, + "-c\n--connect\n" + "-r\n--readonly\n" + "-d\n--debug\n" + "-h\n--help\n" + "-q\n--quiet\n" + "-t\n--timing\n" + "-l\n--log\n" + "-v\n--version\n"); + } + } else if (STREQ(type, "arguments")) { + if (!name) { + vshError(ctl, FALSE, "%s", _("no command specified")); + return FALSE; + } else { + vshCmdDef *def = vshCmddefSearch(name); + if (!def) { + vshError(ctl, FALSE, _("command '%s' doesn't exist"), name); + return FALSE; + } else if (def->opts) { + vshCmdOptDef *opt; + for (opt = def->opts; opt->name; opt++) + if (opt->type == VSH_OT_DATA) + vshPrint(ctl, "%s\n", opt->name); + } + } + } + return TRUE; +} + +/* + * "_compuris" command + */ +static vshCmdInfo info_compuri[] = { + {"syntax", "_complete-uri"}, + {"help", gettext_noop("print shell completion data for URIs")}, + {"desc", gettext_noop("Prints data useful for shell autocompletion of URIs.")}, + + {NULL, NULL} +}; + +static vshCmdOptDef opts_compuri[] = { + {NULL, 0, 0, NULL} +}; + +static int +cmdCompUri(vshControl * ctl, vshCmd * cmd ATTRIBUTE_UNUSED) +{ + char **uris = NULL; + int nuris, i; + + nuris = virConnectURIs(&uris); + if (nuris < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list connection URIs")); + return FALSE; + } + for (i = 0 ; i < nuris ; i++) { + vshPrint(ctl, "%s\n", uris[i]); + free(uris[i]); + } + + free(uris); + return TRUE; +} +/* + * "_compdomains" command + */ +static vshCmdInfo info_compdomain[] = { + {"syntax", "_complete-domain"}, + {"help", gettext_noop("print shell completion data for domains")}, + {"desc", gettext_noop("Prints data useful for shell autocompletion of domains.")}, + + {NULL, NULL} +}; + +static vshCmdOptDef opts_compdomain[] = { + {NULL, 0, 0, NULL} +}; + +static int +cmdCompDomain(vshControl * ctl, vshCmd * cmd ATTRIBUTE_UNUSED) +{ + int *ids = NULL; + char **names = NULL; + int maxname = 0, maxid = 0, i; + + if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) + return FALSE; + + maxid = virConnectNumOfDomains(ctl->conn); + if (maxid < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list active domains")); + return FALSE; + } + if (maxid) { + ids = vshMalloc(ctl, sizeof(int) * maxid); + + if ((maxid = virConnectListDomains(ctl->conn, &ids[0], maxid)) < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list active domains")); + free(ids); + return FALSE; + } + + qsort(&ids[0], maxid, sizeof(int), idsorter); + } + maxname = virConnectNumOfDefinedDomains(ctl->conn); + if (maxname < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list inactive domains")); + free(ids); + return FALSE; + } + if (maxname) { + names = vshMalloc(ctl, sizeof(char *) * maxname); + + if ((maxname = virConnectListDefinedDomains(ctl->conn, names, maxname)) < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list inactive domains")); + free(ids); + free(names); + return FALSE; + } + + qsort(&names[0], maxname, sizeof(char*), namesorter); + } + + for (i = 0; i < maxid; i++) { + char uuid[VIR_UUID_STRING_BUFLEN]; + virDomainPtr dom = virDomainLookupByID(ctl->conn, ids[i]); + /* this kind of work with domains is not atomic operation */ + if (!dom) + continue; + virDomainGetUUIDString(dom, uuid); + vshPrint(ctl, "%d\n%s\n%s\n", + virDomainGetID(dom), + virDomainGetName(dom), + uuid); + virDomainFree(dom); + } + for (i = 0; i < maxname; i++) { + char uuid[VIR_UUID_STRING_BUFLEN]; + virDomainPtr dom = virDomainLookupByName(ctl->conn, names[i]); + + /* this kind of work with domains is not atomic operation */ + if (!dom) { + free(names[i]); + continue; + } + virDomainGetUUIDString(dom, uuid); + vshPrint(ctl, "%s\n%s\n", + virDomainGetName(dom), + uuid); + virDomainFree(dom); + free(names[i]); + } + free(ids); + free(names); + return TRUE; +} + +/* + * "_compnetworks" command + */ +static vshCmdInfo info_compnetwork[] = { + {"syntax", "_complete-network"}, + {"help", gettext_noop("print shell completion data for networks")}, + {"desc", gettext_noop("Prints data useful for shell autocompletion of networks.")}, + + {NULL, NULL} +}; + +static vshCmdOptDef opts_compnetwork[] = { + {NULL, 0, 0, NULL} +}; + +static int +cmdCompNetwork(vshControl * ctl, vshCmd * cmd ATTRIBUTE_UNUSED) +{ + char **activeNames = NULL, **inactiveNames = NULL; + int maxactive = 0, maxinactive = 0, i; + + if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) + return FALSE; + + maxactive = virConnectNumOfNetworks(ctl->conn); + if (maxactive < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list active networks")); + return FALSE; + } + if (maxactive) { + activeNames = vshMalloc(ctl, sizeof(char *) * maxactive); + + if ((maxactive = virConnectListNetworks(ctl->conn, activeNames, + maxactive)) < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list active networks")); + free(activeNames); + return FALSE; + } + + qsort(&activeNames[0], maxactive, sizeof(char *), namesorter); + } + maxinactive = virConnectNumOfDefinedNetworks(ctl->conn); + if (maxinactive < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list inactive networks")); + free(activeNames); + return FALSE; + } + if (maxinactive) { + inactiveNames = vshMalloc(ctl, sizeof(char *) * maxinactive); + + if ((maxinactive = virConnectListDefinedNetworks(ctl->conn, inactiveNames, maxinactive)) < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list inactive networks")); + free(activeNames); + free(inactiveNames); + return FALSE; + } + + qsort(&inactiveNames[0], maxinactive, sizeof(char*), namesorter); + } + + for (i = 0; i < maxactive; i++) { + char uuid[VIR_UUID_STRING_BUFLEN]; + virNetworkPtr network = virNetworkLookupByName(ctl->conn, activeNames[i]); + + /* this kind of work with networks is not atomic operation */ + if (!network) { + free(activeNames[i]); + continue; + } + virNetworkGetUUIDString(network, uuid); + vshPrint(ctl, "%s\n%s\n", + virNetworkGetName(network), + uuid); + virNetworkFree(network); + free(activeNames[i]); + } + for (i = 0; i < maxinactive; i++) { + char uuid[VIR_UUID_STRING_BUFLEN]; + virNetworkPtr network = virNetworkLookupByName(ctl->conn, inactiveNames[i]); + + /* this kind of work with networks is not atomic operation */ + if (!network) { + free(inactiveNames[i]); + continue; + } + virNetworkGetUUIDString(network, uuid); + vshPrint(ctl, "%s\n%s\n", + virNetworkGetName(network), + uuid); + + virNetworkFree(network); + free(inactiveNames[i]); + } + free(activeNames); + free(inactiveNames); + + return TRUE; +} + + +/* + * "_comppools" command + */ +static vshCmdInfo info_comppool[] = { + {"syntax", "_complete-pool"}, + {"help", gettext_noop("print shell completion data for pools")}, + {"desc", gettext_noop("Prints data useful for shell autocompletion of pools.")}, + + {NULL, NULL} +}; + +static vshCmdOptDef opts_comppool[] = { + {NULL, 0, 0, NULL} +}; + +static int +cmdCompPool(vshControl * ctl, vshCmd * cmd ATTRIBUTE_UNUSED) +{ + char **activeNames = NULL, **inactiveNames = NULL; + int maxactive = 0, maxinactive = 0, i; + + if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) + return FALSE; + + maxactive = virConnectNumOfStoragePools(ctl->conn); + if (maxactive < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list active pools")); + return FALSE; + } + if (maxactive) { + activeNames = vshMalloc(ctl, sizeof(char *) * maxactive); + + if ((maxactive = virConnectListStoragePools(ctl->conn, activeNames, + maxactive)) < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list active pools")); + free(activeNames); + return FALSE; + } + + qsort(&activeNames[0], maxactive, sizeof(char *), namesorter); + } + maxinactive = virConnectNumOfDefinedStoragePools(ctl->conn); + if (maxinactive < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list inactive pools")); + free(activeNames); + return FALSE; + } + if (maxinactive) { + inactiveNames = vshMalloc(ctl, sizeof(char *) * maxinactive); + + if ((maxinactive = virConnectListDefinedStoragePools(ctl->conn, inactiveNames, maxinactive)) < 0) { + vshError(ctl, FALSE, "%s", _("Failed to list inactive pools")); + free(activeNames); + free(inactiveNames); + return FALSE; + } + + qsort(&inactiveNames[0], maxinactive, sizeof(char*), namesorter); + } + + for (i = 0; i < maxactive; i++) { + char uuid[VIR_UUID_STRING_BUFLEN]; + virStoragePoolPtr pool = virStoragePoolLookupByName(ctl->conn, activeNames[i]); + + /* this kind of work with pools is not atomic operation */ + if (!pool) { + free(activeNames[i]); + continue; + } + virStoragePoolGetUUIDString(pool, uuid); + vshPrint(ctl, "%s\n%s\n", + virStoragePoolGetName(pool), + uuid); + virStoragePoolFree(pool); + free(activeNames[i]); + } + for (i = 0; i < maxinactive; i++) { + char uuid[VIR_UUID_STRING_BUFLEN]; + virStoragePoolPtr pool = virStoragePoolLookupByName(ctl->conn, inactiveNames[i]); + + /* this kind of work with pools is not atomic operation */ + if (!pool) { + free(inactiveNames[i]); + continue; + } + virStoragePoolGetUUIDString(pool, uuid); + vshPrint(ctl, "%s\n%s\n", + virStoragePoolGetName(pool), + uuid); + + virStoragePoolFree(pool); + free(inactiveNames[i]); + } + free(activeNames); + free(inactiveNames); + + return TRUE; +} + + /* * "help" command */ @@ -355,8 +758,9 @@ vshPrint(ctl, "%s", _("Commands:\n\n")); for (def = commands; def->name; def++) - vshPrint(ctl, " %-15s %s\n", def->name, - N_(vshCmddefGetInfo(def, "help"))); + if (def->name[0] != '_') + vshPrint(ctl, " %-15s %s\n", def->name, + N_(vshCmddefGetInfo(def, "help"))); return TRUE; } return vshCmddefHelp(ctl, cmdname, FALSE); @@ -426,7 +830,7 @@ }; static vshCmdOptDef opts_connect[] = { - {"name", VSH_OT_DATA, 0, gettext_noop("hypervisor connection URI")}, + {"uri", VSH_OT_DATA, 0, gettext_noop("hypervisor connection URI")}, {"readonly", VSH_OT_BOOL, 0, gettext_noop("read-only connection")}, {NULL, 0, 0, NULL} }; @@ -446,7 +850,7 @@ } free(ctl->name); - ctl->name = vshStrdup(ctl, vshCommandOptString(cmd, "name", NULL)); + ctl->name = vshStrdup(ctl, vshCommandOptString(cmd, "uri", NULL)); if (!ro) { ctl->conn = virConnectOpen(ctl->name); @@ -1004,7 +1408,7 @@ }; static vshCmdOptDef opts_start[] = { - {"name", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive domain")}, + {"domain", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive domain")}, {NULL, 0, 0, NULL} }; @@ -1017,7 +1421,7 @@ if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; - if (!(dom = vshCommandOptDomainBy(ctl, cmd, "name", NULL, VSH_BYNAME))) + if (!(dom = vshCommandOptDomainBy(ctl, cmd, "domain", NULL, VSH_BYNAME))) return FALSE; if (virDomainGetID(dom) != (unsigned int)-1) { @@ -2650,7 +3054,7 @@ }; static vshCmdOptDef opts_network_start[] = { - {"name", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive network")}, + {"network", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive network")}, {NULL, 0, 0, NULL} }; @@ -2663,7 +3067,7 @@ if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; - if (!(network = vshCommandOptNetworkBy(ctl, cmd, "name", NULL, VSH_BYNAME))) + if (!(network = vshCommandOptNetworkBy(ctl, cmd, "network", NULL, VSH_BYNAME))) return FALSE; if (virNetworkCreate(network) == 0) { @@ -3554,7 +3958,7 @@ }; static vshCmdOptDef opts_pool_start[] = { - {"name", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive pool")}, + {"pool", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive pool")}, {NULL, 0, 0, NULL} }; @@ -3567,7 +3971,7 @@ if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; - if (!(pool = vshCommandOptPoolBy(ctl, cmd, "name", NULL, VSH_BYNAME))) + if (!(pool = vshCommandOptPoolBy(ctl, cmd, "pool", NULL, VSH_BYNAME))) return FALSE; if (virStoragePoolCreate(pool, 0) == 0) { @@ -5074,6 +5478,12 @@ * Commands */ static vshCmdDef commands[] = { + {"_complete-command", cmdCompCommand, opts_compcommand, info_compcommand}, + {"_complete-uri", cmdCompUri, opts_compuri, info_compuri}, + {"_complete-domain", cmdCompDomain, opts_compdomain, info_compdomain}, + {"_complete-network", cmdCompNetwork, opts_compnetwork, info_compnetwork}, + {"_complete-pool", cmdCompPool, opts_comppool, info_comppool}, + {"help", cmdHelp, opts_help, info_help}, {"attach-device", cmdAttachDevice, opts_attach_device, info_attach_device}, {"attach-disk", cmdAttachDisk, opts_attach_disk, info_attach_disk}, @@ -6514,6 +6924,18 @@ ret = vshCommandParse(ctl, cmdstr); free(cmdstr); + + /* Special case 'help' to avoid virConnectOpen */ + if (ctl->cmd && + ctl->cmd->def && + ctl->cmd->def->name && + (STREQ(ctl->cmd->def->name, "help") || + STREQ(ctl->cmd->def->name, "_complete-command") || + STREQ(ctl->cmd->def->name, "_complete-uri"))) { + ret = vshCommandRun(ctl, ctl->cmd); + exit(ret ? EXIT_SUCCESS : EXIT_FAILURE); + } + return ret; } return TRUE; -- |: 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 :|

3 4

[libvirt] [PATCH] cleanup functions which exec OpenVZ tools
by Evgeniy Sokolov 16 Jul '08

16 Jul '08

Patch improve code which execute external OpenVZ tools.

3 3

[libvirt] auto shutdown for qemu / kvm
by Matthias Pfafferodt 16 Jul '08

16 Jul '08

Hello, there is the possibility to autostart a (qemu/kvm) domain. If libvirtd is stopped this domain is killed and the filesystem is in a unclean state. I miss the possible of an 'autostop' option that will shutdown a (qemu/kvm) domain if libvirtd is stopped. I changed the init script of my system (opensuse 11.0) to implement the 'autostop' option. Is there a simpler solution? Do you have tips how to improve the script? Matthias ## Path: System/Virtualization ## Description: libvirtd stop configuration ## Type: yesno ## Default: yes # # On 'libvirtd stop' all running (kvm) domains are killed. If # LIBVIRTD_KVM_SHUTDOWN is set to 'yes' all running vm's are given time # for a gracefully shutdown. # LIBVIRTD_KVM_SHUTDOWN="yes" ## Path: System/Virtualization ## Description: libvirtd timeout ## Type: integer ## Default: 300 # # On 'libvirtd stop' the vm's are ask to shutdown. This optine gives the # maximum number of seconds the script will wait until libvirt is stopped. # LIBVIRTD_KVM_SHUTDOWN_MAXWAIT=300

2 2

[libvirt] static address assignment via dnsmasq
by Kurt Sussman 15 Jul '08

15 Jul '08

I want my VMs to be assigned static addresses by hostname. I would prefer to set up a generic way to configure dnsmasq, since I don't think this is the last thing I'll want to change about address assignment. I saw this patch from "Mads Chr. Olesen" https://www.redhat.com/archives/libvir-list/2008-April/msg00327.html But it seems like a lot of work to give access to only one dnsmasq option. I'm working on a patch that enables a dnsmasq.d directory where you can (optionally) put individual configuration files. I have not fully tested it yet, so please don't consider this an official submission. I also don't feel that I have spent enough time with the code to understand the 'philosphy' if you will, the guiding attitudes. Is the all-XML approach better according the libvirt developer community standards? If so, why? Thanks. Untested patch follows. Please let me know if I'm on the right path and if not, why. --Kurt diff -u /root/libvirt-0.4.4-virgin/src/qemu_driver.c ./qemu_driver.c --- /root/libvirt-0.4.4-virgin/src/qemu_driver.c 2008-06-12 07:52:53.000000000 -0700 +++ ./qemu_driver.c 2008-07-13 11:31:26.000000000 -0700 @@ -1048,6 +1048,7 @@ 1 + /* --bind-interfaces */ 2 + /* --pid-file "" */ 2 + /* --conf-file "" */ + 2 + /* --conf-dir DNSMASQCONF_DIR */ /*2 + *//* --interface virbr0 */ 2 + /* --except-interface lo */ 2 + /* --listen-address 10.0.0.1 */ @@ -1081,6 +1082,9 @@ APPEND_ARG(*argv, i++, "--conf-file"); APPEND_ARG(*argv, i++, ""); + APPEND_ARG(*argv, i++, "--conf-dir"); + APPEND_ARG(*argv, i++, DNSMASQCONF_DIR); + /* * XXX does not actually work, due to some kind of * race condition setting up ipv6 addresses on the diff -u /root/libvirt-0.4.4-virgin/src/qemu_driver.h ./qemu_driver.h --- /root/libvirt-0.4.4-virgin/src/qemu_driver.h 2008-04-23 07:19:02.000000000 -0700 +++ ./qemu_driver.h 2008-07-13 11:26:47.000000000 -0700 @@ -33,6 +33,8 @@ int qemudRegister(void); +#define DNSMASQCONF_DIR SYSCONF_DIR "/dnsmasq.d" + #endif /* WITH_QEMU */ #endif /* QEMUD_DRIVER_H */ -- ---------------------------------------------------------------------- Merlot Research Group, Inc kls[at]merlot.com "Quality is free, but only to those who are willing to pay heavily for it." (Lister, DeMarco: "Peopleware")

2 1

[libvirt] PATCH: Refactor LXC to remove some state from lxc_vm_t
by Daniel P. Berrange 14 Jul '08

14 Jul '08

I've been working on getting LXC ready to use the common XML routines. One of the first tasks is finding a way to remove some of the state stored in the lxc_vm_t struct, so it can be switched over to virDomainObj struct instead. This patch does some simple re-factoring of the way the TTYs and control socket are handled to prepare for the switchover. lxc_conf.c | 1 lxc_conf.h | 10 -- lxc_container.c | 61 +++++++--------- lxc_container.h | 14 +++ lxc_driver.c | 211 ++++++++++++++++---------------------------------------- 5 files changed, 103 insertions(+), 194 deletions(-) Daniel diff -r 6273f6456bd2 src/lxc_conf.c --- a/src/lxc_conf.c Sun Jul 13 10:33:25 2008 +0100 +++ b/src/lxc_conf.c Sun Jul 13 11:45:04 2008 +0100 @@ -1041,7 +1041,6 @@ void lxcFreeVM(lxc_vm_t *vm) { lxcFreeVMDef(vm->def); - VIR_FREE(vm->containerTty); VIR_FREE(vm); } diff -r 6273f6456bd2 src/lxc_conf.h --- a/src/lxc_conf.h Sun Jul 13 10:33:25 2008 +0100 +++ b/src/lxc_conf.h Sun Jul 13 11:45:04 2008 +0100 @@ -35,12 +35,6 @@ #define LXC_MAX_XML_LENGTH 16384 #define LXC_MAX_ERROR_LEN 1024 #define LXC_DOMAIN_TYPE "lxc" -#define LXC_PARENT_SOCKET 0 -#define LXC_CONTAINER_SOCKET 1 - -/* messages between parent and container */ -typedef char lxc_message_t; -#define LXC_CONTINUE_MSG 'c' /* types of networks for containers */ enum lxc_net_type { @@ -100,10 +94,6 @@ char ttyPidFile[PATH_MAX]; int parentTty; - int containerTtyFd; - char *containerTty; - - int sockpair[2]; lxc_vm_def_t *def; diff -r 6273f6456bd2 src/lxc_container.c --- a/src/lxc_container.c Sun Jul 13 10:33:25 2008 +0100 +++ b/src/lxc_container.c Sun Jul 13 11:45:04 2008 +0100 @@ -33,7 +33,6 @@ #include <unistd.h> #include "lxc_container.h" -#include "lxc_conf.h" #include "util.h" #include "memory.h" #include "veth.h" @@ -83,7 +82,7 @@ * * Returns 0 on success or -1 in case of error */ -static int lxcSetContainerStdio(const char *ttyName) +static int lxcSetContainerStdio(const char *ttyPath) { int rc = -1; int ttyfd; @@ -94,10 +93,10 @@ goto error_out; } - ttyfd = open(ttyName, O_RDWR|O_NOCTTY); + ttyfd = open(ttyPath, O_RDWR|O_NOCTTY); if (ttyfd < 0) { lxcError(NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("open(%s) failed: %s"), ttyName, strerror(errno)); + _("open(%s) failed: %s"), ttyPath, strerror(errno)); goto error_out; } @@ -106,8 +105,6 @@ _("ioctl(TIOCSTTY) failed: %s"), strerror(errno)); goto cleanup; } - - close(0); close(1); close(2); if (dup2(ttyfd, 0) < 0) { lxcError(NULL, NULL, VIR_ERR_INTERNAL_ERROR, @@ -144,12 +141,18 @@ * * Returns 0 on success or -1 in case of error */ -static int lxcExecWithTty(lxc_vm_t *vm) +static int lxcExecWithTty(lxc_vm_def_t *vmDef, char *ttyPath) { int rc = -1; - lxc_vm_def_t *vmDef = vm->def; + int open_max, i; - if(lxcSetContainerStdio(vm->containerTty) < 0) { + /* Just in case someone forget to set FD_CLOEXEC, explicitly + * close all FDs before executing the container */ + open_max = sysconf (_SC_OPEN_MAX); + for (i = 0; i < open_max; i++) + close(i); + + if(lxcSetContainerStdio(ttyPath) < 0) { goto exit_with_error; } @@ -161,7 +164,7 @@ /** * lxcWaitForContinue: - * @vm: Pointer to vm structure + * @monitor: monitor FD from parent * * This function will wait for the container continue message from the * parent process. It will send this message on the socket pair stored in @@ -169,31 +172,23 @@ * * Returns 0 on success or -1 in case of error */ -static int lxcWaitForContinue(lxc_vm_t *vm) +static int lxcWaitForContinue(int monitor) { - int rc = -1; lxc_message_t msg; int readLen; - readLen = saferead(vm->sockpair[LXC_CONTAINER_SOCKET], &msg, sizeof(msg)); - if (readLen != sizeof(msg)) { + readLen = saferead(monitor, &msg, sizeof(msg)); + if (readLen != sizeof(msg) || + msg != LXC_CONTINUE_MSG) { lxcError(NULL, NULL, VIR_ERR_INTERNAL_ERROR, _("Failed to read the container continue message: %s"), strerror(errno)); - goto error_out; + return -1; } DEBUG0("Received container continue message"); - close(vm->sockpair[LXC_PARENT_SOCKET]); - vm->sockpair[LXC_PARENT_SOCKET] = -1; - close(vm->sockpair[LXC_CONTAINER_SOCKET]); - vm->sockpair[LXC_CONTAINER_SOCKET] = -1; - - rc = 0; - -error_out: - return rc; + return 0; } /** @@ -204,12 +199,12 @@ * * Returns 0 on success or nonzero in case of error */ -static int lxcEnableInterfaces(const lxc_vm_t *vm) +static int lxcEnableInterfaces(const lxc_vm_def_t *def) { int rc = 0; const lxc_net_def_t *net; - for (net = vm->def->nets; net; net = net->next) { + for (net = def->nets; net; net = net->next) { DEBUG("Enabling %s", net->containerVeth); rc = vethInterfaceUpOrDown(net->containerVeth, 1); if (0 != rc) { @@ -218,7 +213,7 @@ } /* enable lo device only if there were other net devices */ - if (vm->def->nets) + if (def->nets) rc = vethInterfaceUpOrDown("lo", 1); error_out: @@ -237,11 +232,11 @@ * * Returns 0 on success or -1 in case of error */ -int lxcChild( void *argv ) +int lxcChild( void *data ) { int rc = -1; - lxc_vm_t *vm = (lxc_vm_t *)argv; - lxc_vm_def_t *vmDef = vm->def; + lxc_child_argv_t *argv = data; + lxc_vm_def_t *vmDef = argv->config; lxc_mount_t *curMount; int i; @@ -278,16 +273,16 @@ } /* Wait for interface devices to show up */ - if (0 != (rc = lxcWaitForContinue(vm))) { + if (0 != (rc = lxcWaitForContinue(argv->monitor))) { goto cleanup; } /* enable interfaces */ - if (0 != (rc = lxcEnableInterfaces(vm))) { + if (0 != (rc = lxcEnableInterfaces(vmDef))) { goto cleanup; } - rc = lxcExecWithTty(vm); + rc = lxcExecWithTty(vmDef, argv->ttyPath); /* this function will only return if an error occured */ cleanup: diff -r 6273f6456bd2 src/lxc_container.h --- a/src/lxc_container.h Sun Jul 13 10:33:25 2008 +0100 +++ b/src/lxc_container.h Sun Jul 13 11:45:04 2008 +0100 @@ -24,7 +24,21 @@ #ifndef LXC_CONTAINER_H #define LXC_CONTAINER_H +#include "lxc_conf.h" + #ifdef WITH_LXC + +typedef struct __lxc_child_argv lxc_child_argv_t; +struct __lxc_child_argv { + lxc_vm_def_t *config; + int monitor; + char *ttyPath; +}; + +/* messages between parent and container */ +typedef char lxc_message_t; +#define LXC_CONTINUE_MSG 'c' + /* Function declarations */ int lxcChild( void *argv ); diff -r 6273f6456bd2 src/lxc_driver.c --- a/src/lxc_driver.c Sun Jul 13 10:33:25 2008 +0100 +++ b/src/lxc_driver.c Sun Jul 13 11:45:04 2008 +0100 @@ -561,27 +561,23 @@ /** * lxcSendContainerContinue: - * @vm: pointer to virtual machine structure + * @monitor: FD for communicating with child * * Sends the continue message via the socket pair stored in the vm * structure. * * Returns 0 on success or -1 in case of error */ -static int lxcSendContainerContinue(const lxc_vm_t *vm) +static int lxcSendContainerContinue(virConnectPtr conn, + int monitor) { int rc = -1; lxc_message_t msg = LXC_CONTINUE_MSG; int writeCount = 0; - if (NULL == vm) { - goto error_out; - } - - writeCount = safewrite(vm->sockpair[LXC_PARENT_SOCKET], &msg, - sizeof(msg)); + writeCount = safewrite(monitor, &msg, sizeof(msg)); if (writeCount != sizeof(msg)) { - lxcError(NULL, NULL, VIR_ERR_INTERNAL_ERROR, + lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, _("unable to send container continue message: %s"), strerror(errno)); goto error_out; @@ -605,12 +601,15 @@ */ static int lxcStartContainer(virConnectPtr conn, lxc_driver_t* driver, - lxc_vm_t *vm) + lxc_vm_t *vm, + int monitor, + char *ttyPath) { int rc = -1; int flags; int stacksize = getpagesize() * 4; char *stack, *stacktop; + lxc_child_argv_t args = { vm->def, monitor, ttyPath }; /* allocate a stack for the container */ if (VIR_ALLOC_N(stack, stacksize) < 0) { @@ -625,7 +624,7 @@ if (vm->def->nets != NULL) flags |= CLONE_NEWNET; - vm->def->id = clone(lxcChild, stacktop, flags, (void *)vm); + vm->def->id = clone(lxcChild, stacktop, flags, &args); DEBUG("clone() returned, %d", vm->def->id); @@ -643,117 +642,9 @@ return rc; } -/** - * lxcPutTtyInRawMode: - * @conn: pointer to connection - * @ttyDev: file descriptor for tty - * - * Sets tty attributes via cfmakeraw() - * - * Returns 0 on success or -1 in case of error - */ -static int lxcPutTtyInRawMode(virConnectPtr conn, int ttyDev) -{ - int rc = -1; - - struct termios ttyAttr; - - if (tcgetattr(ttyDev, &ttyAttr) < 0) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - "tcgetattr() failed: %s", strerror(errno)); - goto cleanup; - } - - cfmakeraw(&ttyAttr); - - if (tcsetattr(ttyDev, TCSADRAIN, &ttyAttr) < 0) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - "tcsetattr failed: %s", strerror(errno)); - goto cleanup; - } - - rc = 0; - -cleanup: - return rc; -} /** - * lxcSetupTtyTunnel: - * @conn: pointer to connection - * @vmDef: pointer to virtual machine definition structure - * @ttyDev: pointer to int. On success will be set to fd for master - * end of tty - * - * Opens and configures the parent side tty - * - * Returns 0 on success or -1 in case of error - */ -static int lxcSetupTtyTunnel(virConnectPtr conn, - lxc_vm_def_t *vmDef, - int* ttyDev) -{ - int rc = -1; - char *ptsStr; - - if (0 < strlen(vmDef->tty)) { - *ttyDev = posix_openpt(O_RDWR|O_NOCTTY|O_NONBLOCK); - if (*ttyDev < 0) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - "open() tty failed: %s", strerror(errno)); - goto setup_complete; - } - - rc = grantpt(*ttyDev); - if (rc < 0) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - "grantpt() failed: %s", strerror(errno)); - goto setup_complete; - } - - rc = unlockpt(*ttyDev); - if (rc < 0) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - "unlockpt() failed: %s", strerror(errno)); - goto setup_complete; - } - - /* get the name and print it to stdout */ - ptsStr = ptsname(*ttyDev); - if (ptsStr == NULL) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - "ptsname() failed"); - goto setup_complete; - } - /* This value needs to be stored in the container configuration file */ - VIR_FREE(vmDef->tty); - if (!(vmDef->tty = strdup(ptsStr))) { - lxcError(conn, NULL, VIR_ERR_NO_MEMORY, - _("unable to get storage for vm tty name")); - goto setup_complete; - } - - /* Enter raw mode, so all characters are passed directly to child */ - if (lxcPutTtyInRawMode(conn, *ttyDev) < 0) { - goto setup_complete; - } - - } else { - *ttyDev = -1; - } - - rc = 0; - -setup_complete: - if((0 != rc) && (*ttyDev > 0)) { - close(*ttyDev); - } - - return rc; -} - -/** - * lxcSetupContainerTty: + * lxcOpenTty: * @conn: pointer to connection * @ttymaster: pointer to int. On success, set to fd for master end * @ttyName: On success, will point to string slave end of tty. Caller @@ -763,12 +654,12 @@ * * Returns 0 on success or -1 in case of error */ -static int lxcSetupContainerTty(virConnectPtr conn, - int *ttymaster, - char **ttyName) +static int lxcOpenTty(virConnectPtr conn, + int *ttymaster, + char **ttyName, + int rawmode) { int rc = -1; - char tempTtyName[PATH_MAX]; *ttymaster = posix_openpt(O_RDWR|O_NOCTTY|O_NONBLOCK); if (*ttymaster < 0) { @@ -783,27 +674,43 @@ goto cleanup; } - if (0 != ptsname_r(*ttymaster, tempTtyName, sizeof(tempTtyName))) { - lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - _("ptsname_r failed: %s"), strerror(errno)); - goto cleanup; + if (rawmode) { + struct termios ttyAttr; + if (tcgetattr(*ttymaster, &ttyAttr) < 0) { + lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, + "tcgetattr() failed: %s", strerror(errno)); + goto cleanup; + } + + cfmakeraw(&ttyAttr); + + if (tcsetattr(*ttymaster, TCSADRAIN, &ttyAttr) < 0) { + lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, + "tcsetattr failed: %s", strerror(errno)); + goto cleanup; + } } - if (VIR_ALLOC_N(*ttyName, strlen(tempTtyName) + 1) < 0) { - lxcError(conn, NULL, VIR_ERR_NO_MEMORY, - _("unable to allocate container name string")); - goto cleanup; + if (ttyName) { + char tempTtyName[PATH_MAX]; + if (0 != ptsname_r(*ttymaster, tempTtyName, sizeof(tempTtyName))) { + lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, + _("ptsname_r failed: %s"), strerror(errno)); + goto cleanup; + } + + if ((*ttyName = strdup(tempTtyName)) == NULL) { + lxcError(conn, NULL, VIR_ERR_NO_MEMORY, NULL); + goto cleanup; + } } - - strcpy(*ttyName, tempTtyName); rc = 0; cleanup: - if (0 != rc) { - if (-1 != *ttymaster) { - close(*ttymaster); - } + if (rc != 0 && + *ttymaster != -1) { + close(*ttymaster); } return rc; @@ -989,15 +896,18 @@ lxc_vm_t * vm) { int rc = -1; - lxc_vm_def_t *vmDef = vm->def; + int sockpair[2]; + int containerTty; + char *containerTtyPath = NULL; /* open parent tty */ - if (lxcSetupTtyTunnel(conn, vmDef, &vm->parentTty) < 0) { + VIR_FREE(vm->def->tty); + if (lxcOpenTty(conn, &vm->parentTty, &vm->def->tty, 1) < 0) { goto cleanup; } /* open container tty */ - if (lxcSetupContainerTty(conn, &(vm->containerTtyFd), &(vm->containerTty)) < 0) { + if (lxcOpenTty(conn, &containerTty, &containerTtyPath, 0) < 0) { goto cleanup; } @@ -1011,7 +921,7 @@ if (vm->pid == 0) { /* child process calls forward routine */ - lxcTtyForward(vm->parentTty, vm->containerTtyFd); + lxcTtyForward(vm->parentTty, containerTty); } if (lxcStoreTtyPid(driver, vm)) { @@ -1019,7 +929,7 @@ } close(vm->parentTty); - close(vm->containerTtyFd); + close(containerTty); if (0 != (rc = lxcSetupInterfaces(conn, vm))) { goto cleanup; @@ -1027,7 +937,7 @@ /* create a socket pair to send continue message to the container once */ /* we've completed the post clone configuration */ - if (0 != socketpair(PF_UNIX, SOCK_STREAM, 0, vm->sockpair)) { + if (0 != socketpair(PF_UNIX, SOCK_STREAM, 0, sockpair)) { lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, _("sockpair failed: %s"), strerror(errno)); goto cleanup; @@ -1035,7 +945,9 @@ /* check this rc */ - rc = lxcStartContainer(conn, driver, vm); + rc = lxcStartContainer(conn, driver, vm, + sockpair[1], + containerTtyPath); if (rc != 0) goto cleanup; @@ -1043,7 +955,7 @@ if (rc != 0) goto cleanup; - rc = lxcSendContainerContinue(vm); + rc = lxcSendContainerContinue(conn, sockpair[0]); if (rc != 0) goto cleanup; @@ -1052,10 +964,9 @@ driver->nactivevms++; cleanup: - close(vm->sockpair[LXC_PARENT_SOCKET]); - vm->sockpair[LXC_PARENT_SOCKET] = -1; - close(vm->sockpair[LXC_CONTAINER_SOCKET]); - vm->sockpair[LXC_CONTAINER_SOCKET] = -1; + close(sockpair[0]); + close(sockpair[1]); + VIR_FREE(containerTtyPath); return rc; } -- |: 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] save / restore for KVM
by Matthias Pfafferodt 12 Jul '08

12 Jul '08

Hello, I try to save a KVM domain which has its hard disk on a LVM volume. I can execute the command to save the domain but restore fails # virsh list Id Name State ---------------------------------- 3 kvm-linux03 running 4 kvm-linux01 running # virsh save kvm-linux01 ./kvm-linux01.ram Domain kvm-linux01 saved to ./kvm-linux01.ram # virsh list Id Name State ---------------------------------- 3 kvm-linux03 running # virsh restore ./kvm-linux01.ram libvir: QEMU error : operation failed: failed to start VM error: Failed to restore domain from ./kvm-linux01.ram I tried to save and restore this domain using kvm directly # /usr/bin/qemu-kvm -m 1024 -boot c -drive file=/dev/virtual/kvm_linux01,if=ide,index=0,boot=on -net nic,macaddr=00:16:3e:00:01:02,vlan=0 -net tap,fd=24,script=,vlan=0,ifname=vnet3 (qemu) savevm kvm-linux01.ram Device id0-hd0 does not support VM state snapshots (qemu) Here I get an error message which I also found on the qemu homepage: to save a qemu vm the hard disk must be a file using the qcow format. But in the libvirt documentation there is no hint to this limitation. Is it possible to save the state of a kvm vm using physical hard disks with libvirt? Kind regards Matthias Pfafferodt -- Matthias Pfafferodt - http://www.mapfa.de Matthias.Pfafferodt <at> mapfa.de

3 4

[libvirt] [PATCH] default URI in probe
by Evgeniy Sokolov 11 Jul '08

11 Jul '08

I specified default URI in probe(). To use libvirt with OpenVZ: # ./configure --with-openvz # make # ./src/virsh list Id Name State ---------------------------------- 33 33 running 105 105 running 108 108 running

2 1

[libvirt] [PATCH] using one function to locate OpenVZ config files
by Evgeniy Sokolov 11 Jul '08

11 Jul '08

There were 3 places with the same code. I moved it to separate function. other: use O_RDONLY to read config file

2 1

[libvirt] bug in libvirt.c?
by Evgeniy Sokolov 10 Jul '08

10 Jul '08

for (i = 0; i < virNetworkDriverTabCount; i++) { if ((virDriverTab[i]->probe != NULL) && ((latest = virDriverTab[i]->probe()) != NULL)) { probes++; ..... } } We use array virDriverTab, but cicle is to virNetworkDriverTabCount which count number of virNetworkDriverTab drivers. I think virDriverTabCount should be used. Index: libvirt.c =================================================================== RCS file: /data/cvs/libvirt/src/libvirt.c,v retrieving revision 1.147 diff -u -p -r1.147 libvirt.c --- libvirt.c 26 Jun 2008 09:37:51 -0000 1.147 +++ libvirt.c 10 Jul 2008 14:48:44 -0000 @@ -699,7 +699,7 @@ do_open (const char *name, const char *use = NULL; const char *latest; int probes = 0; - for (i = 0; i < virNetworkDriverTabCount; i++) { + for (i = 0; i < virDriverTabCount; i++) { if ((virDriverTab[i]->probe != NULL) && ((latest = virDriverTab[i]->probe()) != NULL)) { probes++;

3 2

[libvirt] [Patch] Libvirt & Autobuild
by Mohammed Morsi 10 Jul '08

10 Jul '08

Attached is a small fix to get libvirt working w/ autobuild. Autobuild attempts to create and test the RPM, which lists the PolicyKit as one of the files to be installed, and thus if it is not installed, autobuild reports an error and fails. The was to get this file installed is to specify --with-polkit to ./autogen which I added to the autobuild.sh build script. (also the syntax check errors broke autobuild, but these have been fixed according to the latest email on the list). Knock on wood, after this libvirt should work w/ autobuild and will be built nightly, and I can finish getting oVirt working with it. -Mo

2 3

[libvirt] PV loader in capabilities?
by John Levon 10 Jul '08

10 Jul '08

Why is it NULL instead of the path to pygrub? On Solaris, the path is different. regards john

3 4

[libvirt] [PATCH] fix diagnostic when execute openvz commands
by Evgeniy Sokolov 10 Jul '08

10 Jul '08

-- Evgeniy Sokolov

4 7

[libvirt] [PATCH] renaming error -> openvzError
by Evgeniy Sokolov 10 Jul '08

10 Jul '08

function error() was renamed to openvzError() to don't conflict with other functions with the same name. other - openvzError() is added to msg_gen_function - fixed to pass "make syntax-check" Thanks to Jim Meyering for review.

2 1

[libvirt] [PATCH] fix up dom-name validation
by John Levon 10 Jul '08

10 Jul '08

libvirt schema: improve domain name regex This regexp is based upon what xend actually allows, plus the existing &, which I presume is there for a reason. Signed-off-by: John Levon <john.levon> Index: docs/libvirt.rng =================================================================== RCS file: /data/cvs/libvirt/docs/libvirt.rng,v retrieving revision 1.8 diff -u -r1.8 libvirt.rng --- docs/libvirt.rng 8 Jul 2008 12:05:13 -0000 1.8 +++ docs/libvirt.rng 10 Jul 2008 04:04:58 -0000 @@ -840,7 +840,7 @@ </define> <define name='domainName'> <data type='string'> - <param name="pattern">[a-zA-Z0-9_\+\-&]+</param> + <param name="pattern">[A-Za-z0-9_.:/\+\-&-]+</param> </data> </define> <define name='genericName'>

2 1

[libvirt] [PATCH] SetAutostart and GetAutostart in openvz driver
by Evgeniy Sokolov 10 Jul '08

10 Jul '08

Implemented functions domainSetAutostart and domainGetAutostart. other: fixed closing file descriptor during reading uuid from config.

3 3

[libvirt] virDomainMemoryPeek & maximum remote message buffer size
by Richard W.M. Jones 09 Jul '08

09 Jul '08

The kernel images that I want to snoop in virt-mem are around 16 MB in size. In the qemu / KVM case, these images have to travel over the remote connection. Because of limits on the maximum message size, they have to travel currently in 64 KB chunks, and it turns out that this is slow. Apparently the dominating factors are how long it takes to issue the 'memsave' command in the qemu monitor (there is some big constant overhead), and extra network round trips. The current remote message size is intentionally limited to 256KB (fully serialized, including all XDR headers and overhead), so the most we could practically send in a single message at the moment is 128KB if we stick to powers of two, or ~255KB if we don't. The reason we limit it is to avoid denial of service attacks, where a rogue client or server sends excessively large messages and causes the peer to allocate lots of memory [eg. if we didn't have any limit, then you could send a message which was several GB in size and cause problems at the other end, because the message is slurped in before it is fully parsed]. There is a second problem with reading the kernel in small chunks, namely that this allows the virtual machine to make a lot of progress, so we don't get anything near an 'instantaneous' snapshot (getting the kernel in a single chunk doesn't necessarily guarantee this either, but it's better). As an experiment, I tried increasing the maximum message to 32 MB, so that I could send the whole kernel in one go. Unfortunately just increasing the limit doesn't work for two reasons, one prosaic and one very weird: (1) The current code likes to keep message buffers on the stack, and because Linux limits the stack to something artificially small, this fails. Increasing the stack ulimit is a short-term fix for this, while testing. In the long term we could rewrite any code which does this to use heap buffers instead. (2) There is some really odd problem with our use of recv(2) which causes messages > 64 KB to fail. I have no idea what is really happening, but the sequence of events seems to be this: server client write(sock,buf,len) = len-k recv(sock,buf,len) = len-k write(sock,buf+len-k,len-k) = k recv(sock,buf,k) = 0 [NOT k] At this point the client assumes that the server has unexpectedly closed the connection and fails. I have stared at this for a while, but I've got no idea at all what's going on. A test program is attached. You'll need a 32 bit KVM guest. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/

3 2

[libvirt] Web Interface -> LibVirt Communication?
by Henri Cook 09 Jul '08

09 Jul '08

Hi, I want to write a web tool that lets me administer my cluster from one location (including operations that things like virt-manager don't provide) - what's the proper way of achieving this? Can i communicate over TCP with libvirt on each server (is this documented?) or should I be trying to get my web app to auth and run virsh commands over SSH? Thanks. Henri

4 11

Re: [libvirt] libvirt-java
by Daniel Veillard 09 Jul '08

09 Jul '08

On Mon, Jul 07, 2008 at 01:14:40PM +0200, Tóth István wrote: > Daniel Veillard wrote: > > Ah, okay.... Well my perspective is that libvirt-java should include > > only parts which are directly releated to using libvirt on Java. I would > > really not try to reinvent or push any XML related APIs there, at least > > as part of libvirt bindings (I still have the scars from libxml2, believe > > me you don't want to push new XML APIs to programmers even specialized ones). > > > My thoughts exactly. Okay seems we agree on the fundamentals :-) > > But things like the XSD descriptiond might be useful generally, and > > IMHO are very tied to libvirt, so i think they are a good fit. > > > Absolutely. I've had major problems when I tried to use some Java XML > mapping libraries, because they can only process xsd, > and not the language you use. So having high-quality xsd definitions is > high on my personal wish-list. Okay > > Also fitting into this model are build makefiles for other environment, > > I understand that for most Java developpers, configure and make are really > > foreign tools, so adding ant/Eclipse/... build recipes for the package also > > makes a lot of sense. > > > Well, they certainly wouldn't hurt, but I don't see that as a priority. > Java-only programmers (end users) won't touch the JNI core parts. > They will just install the package, and get on with using it, by adding > the jar > to their devel enviroment, and perhaps the JNI lib file to the command line. > > These kind of users don't really care about the java-libvirt build > environment. okay, so just the Jar availability is sufficient for them, okidoc. > If you want to do something more interesting than renaming the java > classes and methods, > the you have to hit the JNI bindings, and that code is so ugly, that > no-one without some C experience will want to touch it. > > Having said that, I think that an ant build file would be the most > useful, as it can be used directly in all major > development enviroments. There is also this maven thing, that seems to > be widely used, but I still haven't figured out > exactly what that is :-) me either, though there are examples in the Fedora-java packaging pages. [...] > My current plans for java-libvirt are: > > 1. Add the storage API: It's really mostly just copy-paste-search > replace but it still takes some work okay > 2. There are some consistency problems with the naming of classes and > methods. I'd like to revisit the java api, and make some changes in > names, and maybe class structure Hum, better done early than late. Basically i would prefer to avoid pushing incompatible changes. what kind of inconsistencies problems ? > 3. There are many places where the C part leaks memory, this should also > be audited/ fixed. Ah, okay I will have to reread the bindings code then. Not sure how to track leaks, I doubt valgrind can work with java ... > Number 2 is what worries me, I don't know if it's a good idea to push > toFedora, when I know I want to make incompatible API changes soon. yes, which is why I would like to know a bit better :-) > (Or you can just say that you won't accept them, but I'm a big fan of > clean and consistent APIs, and the current one can be improved) > I believe that I will get around to doing 1. and 2. at least in late > july/early august, It's about a three day job, I just don't have that > three days right now :-( Maybe if you can expose what you think is wrong i can try to clean things up. Daniel -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

3 3

[libvirt] PATCH: Fix ue of VIR_TEST_DEBUG var
by Daniel P. Berrange 09 Jul '08

09 Jul '08

The previous patch I did to add OOM testing accidentally disabled use of the VIR_TEST_DEBUG env variable when OOM testing is #ifdef'd out. This patch I just comitted makes sure its always used. Daniel diff -r d37150dc5895 tests/testutils.c --- a/tests/testutils.c Tue Jul 08 16:17:52 2008 +0100 +++ b/tests/testutils.c Wed Jul 09 11:23:27 2008 +0100 @@ -321,20 +321,24 @@ char **argv, int (*func)(int, char **)) { + char *debugStr; #if TEST_OOM int ret; int approxAlloc = 0; int n; - char *oomStr = NULL, *debugStr; + char *oomStr = NULL; int oomCount; int mp = 0; pid_t *workers; int worker = 0; +#endif + if ((debugStr = getenv("VIR_TEST_DEBUG")) != NULL) { if (virStrToLong_ui(debugStr, NULL, 10, &testDebug) < 0) testDebug = 0; } +#if TEST_OOM if ((oomStr = getenv("VIR_TEST_OOM")) != NULL) { if (virStrToLong_i(oomStr, NULL, 10, &oomCount) < 0) oomCount = 0; -- |: 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 :|

1 0

[libvirt] PATCH: Misc changes to policykit policy file
by Daniel P. Berrange 09 Jul '08

09 Jul '08

After discussions with policykit maintainers I've come to the conclusion that it is better for security if we default to 'auth_admin_keep_sesion' instead of 'auth_self_keep_session'. ie prompt for the root password (ala 'su') instead of the user's password (ala 'sudo'). This is because having access to libvirtd gives you very significant power over the host machine. Secondly, newer versions of policykit have imposed a naming constraint on policy files, so when we install our policy it needs to be in a file called org.libvirt.unix.policy, instead of just libvirt.policy. So there's a change to the Makefile to support this. Daniel diff -r e17c1bd119fc libvirt.spec.in --- a/libvirt.spec.in Wed Jul 09 09:26:26 2008 +0100 +++ b/libvirt.spec.in Wed Jul 09 10:32:24 2008 +0100 @@ -240,7 +240,7 @@ %dir %attr(0700, root, root) %{_localstatedir}/lib/libvirt/images/ %dir %attr(0700, root, root) %{_localstatedir}/lib/libvirt/boot/ %if %{with_polkit} -%{_datadir}/PolicyKit/policy/libvirtd.policy +%{_datadir}/PolicyKit/policy/org.libvirt.unix.policy %endif %dir %attr(0700, root, root) %{_localstatedir}/log/libvirt/qemu/ %if %{with_xen_proxy} diff -r e17c1bd119fc qemud/Makefile.am --- a/qemud/Makefile.am Wed Jul 09 09:26:26 2008 +0100 +++ b/qemud/Makefile.am Wed Jul 09 10:32:24 2008 +0100 @@ -13,7 +13,8 @@ remote_dispatch_proc_switch.h \ mdns.c mdns.h \ libvirtd.sasl \ - libvirtd.conf + libvirtd.conf \ + libvirtd.policy if RPCGEN SUFFIXES = .x @@ -75,9 +76,7 @@ if HAVE_POLKIT policydir = $(datadir)/PolicyKit/policy -policy_DATA = libvirtd.policy endif -EXTRA_DIST += libvirtd.policy if HAVE_AVAHI libvirtd_SOURCES += mdns.c mdns.h @@ -86,7 +85,7 @@ endif default_xml_dest = libvirt/qemu/networks/default.xml -install-data-local: install-init install-data-sasl +install-data-local: install-init install-data-sasl install-data-polkit mkdir -p $(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart $(INSTALL_DATA) $(srcdir)/default-network.xml \ $(DESTDIR)$(sysconfdir)/$(default_xml_dest) @@ -99,7 +98,7 @@ mkdir -p $(DESTDIR)$(localstatedir)/run/libvirt mkdir -p $(DESTDIR)$(localstatedir)/lib/libvirt -uninstall-local:: uninstall-init uninstall-data-sasl +uninstall-local:: uninstall-init uninstall-data-sasl install-data-polkit rm -f $(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart/default.xml rm -f $(DESTDIR)$(sysconfdir)/$(default_xml_dest) rmdir $(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart || : @@ -119,6 +118,18 @@ install-data-sasl: uninstall-data-sasl: endif + +if HAVE_POLKIT +install-data-polkit:: install-init + mkdir -p $(DESTDIR)$(policydir) + $(INSTALL_DATA) libvirtd.policy $(DESTDIR)$(policydir)/org.libvirt.unix.policy +uninstall-data-polkit:: install-init + rm -f $(DESTDIR)$(policydir)/org.libvirt.unix.policy +else +install-data-polkit:: +uninstall-data-polkit:: +endif + remote.c: remote_dispatch_prototypes.h \ remote_dispatch_localvars.h \ diff -r e17c1bd119fc qemud/libvirtd.policy --- a/qemud/libvirtd.policy Wed Jul 09 09:26:26 2008 +0100 +++ b/qemud/libvirtd.policy Wed Jul 09 10:32:24 2008 +0100 @@ -36,7 +36,7 @@ read-write mode for management, and we require user password --> <allow_any>no</allow_any> <allow_inactive>no</allow_inactive> - <allow_active>auth_self_keep_session</allow_active> + <allow_active>auth_admin_keep_session</allow_active> </defaults> </action> -</policyconfig> \ No newline at end of file +</policyconfig> -- |: 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] monitor cpu & network load
by Zu 09 Jul '08

09 Jul '08

Hi, I want to write a program to monitor the CPU and network usage of the Xen domUs. But I don't know how to implement. Can the libvirt APIs help me to do that? Does libvirt have the APIs to get the CPU and network usages? If yes I'll study libvirt APIs. Thanks Zu

3 2

[libvirt] PATCH: DOn't use policykit if running non-root
by Daniel P. Berrange 09 Jul '08

09 Jul '08

If the libvirtd daemon is running as non-root (ie a regular user) then there is no point in defaulting policykit, since only clients of same user account are able to connect to the UNIX socket anyway. This removes the pointless login prompt when using qemu:///session. Daniel diff -r 9a457b847ffc qemud/qemud.c --- a/qemud/qemud.c Tue Jul 08 16:17:52 2008 +0100 +++ b/qemud/qemud.c Wed Jul 09 09:24:53 2008 +0100 @@ -1926,6 +1926,14 @@ GET_CONF_STR (conf, filename, tcp_port); GET_CONF_STR (conf, filename, listen_addr); +#if HAVE_POLKIT + /* Change the default back to no auth for non-root */ + if (getuid() != 0 && auth_unix_rw == REMOTE_AUTH_POLKIT) + auth_unix_rw = REMOTE_AUTH_NONE; + if (getuid() != 0 && auth_unix_ro == REMOTE_AUTH_POLKIT) + auth_unix_ro = REMOTE_AUTH_NONE; +#endif + if (remoteConfigGetAuth(conf, "auth_unix_rw", &auth_unix_rw, filename) < 0) goto free_and_fail; #if HAVE_POLKIT -- |: 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] [PATCH] give a more useful diagnostic for tap-add failure w/ENOTSUP
by Jim Meyering 09 Jul '08

09 Jul '08

I invoked virt-isntall with --network=bridge:eth1 rather than --network=bridge:br1 and got the latter (current) diagnostic below. This change makes it so in this relatively common case people get a slightly more understandable diagnostic. >From 3d6dc8442fd9f1b0915c232040200154832fdf51 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 8 Jul 2008 13:58:10 +0200 Subject: [PATCH] give a more useful diagnostic for tap-add failure w/ENOTSUP * src/qemu_conf.c (qemudNetworkIfaceConnect): Suggestion from Daniel P. Berrange. --- src/qemu_conf.c | 14 +++++++++++--- 1 files changed, 11 insertions(+), 3 deletions(-) diff --git a/src/qemu_conf.c b/src/qemu_conf.c index 17f0162..4b8c01e 100644 --- a/src/qemu_conf.c +++ b/src/qemu_conf.c @@ -2304,9 +2304,17 @@ qemudNetworkIfaceConnect(virConnectPtr conn, if ((err = brAddTap(driver->brctl, brname, ifname, BR_IFNAME_MAXLEN, &tapfd))) { - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("Failed to add tap interface '%s' to bridge '%s' : %s"), - ifname, brname, strerror(err)); + if (errno == ENOTSUP) { + /* In this particular case, give a better diagnostic. */ + qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + _("Failed to add tap interface to bridge. " + "%s is not a bridge device"), brname); + } else { + qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + _("Failed to add tap interface '%s' " + "to bridge '%s' : %s"), + ifname, brname, strerror(err)); + } goto error; } -- 1.5.6.2.221.gb589

3 3