- Devel - libvirt List Archives

[libvirt] [PATCH 0/3 v5] use qemu's dump-guest-meory when vm uses host device
by Wen Congyang 15 Jun '12

15 Jun '12

Currently, we use migrate to dump guest's memory. There is one restriction in migrate command: the device's status should be stored in qemu because the device's status should be passed to target machine. If we passthrough a host device to guest, the device's status is stored in the real device. So migrate command will fail. We usually use dump when guest is panicked. So there is no need to store device's status in the vmcore. qemu have a new monitor command dump-guest-memory to dump guest memory, but it doesn't support async now(it will support later when the common async API is implemented). Changes from v4 to v5: 1. address Martin Kletzander's comment Changes from v3 to v4: 1. allow the user to specify the core file's format Changes from v2 to v3: 1. qemu supports the fd that is associated with a pipe, socket, or FIFO. So pass a pipe fd to qemu and O_DIRECT can work now. Change from v1 to v2: 1. remove the implemention for text mode. Wen Congyang (3): qemu: implement qemu's dump-guest-memory qemu: allow the client to choose the vmcore's format virsh: allow the user to specify vmcore's format include/libvirt/libvirt.h.in | 1 + src/qemu/qemu_capabilities.c | 5 +++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_domain.c | 1 + src/qemu/qemu_domain.h | 1 + src/qemu/qemu_driver.c | 60 ++++++++++++++++++++++++++++++++--------- src/qemu/qemu_monitor.c | 37 ++++++++++++++++++++++++++ src/qemu/qemu_monitor.h | 12 ++++++++ src/qemu/qemu_monitor_json.c | 34 +++++++++++++++++++++++ src/qemu/qemu_monitor_json.h | 6 ++++ tools/virsh.c | 3 ++ tools/virsh.pod | 5 +++- 12 files changed, 152 insertions(+), 14 deletions(-)

3 8

[libvirt] [PATCH] qemu: fix potential dead lock
by Wen Congyang 15 Jun '12

15 Jun '12

If we lock the qemu_driver, we should call qemuDomainObjBeginJobWithDriver() not qemuDomainObjBeginJob(). --- src/qemu/qemu_driver.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 39b27b1..3d62aab 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -8905,7 +8905,7 @@ qemuDomainMigrateBegin3(virDomainPtr domain, goto cleanup; asyncJob = QEMU_ASYNC_JOB_MIGRATION_OUT; } else { - if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) + if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; asyncJob = QEMU_ASYNC_JOB_NONE; } -- 1.7.1

3 2

[libvirt] RPC ref count question
by Radu Caragea 15 Jun '12

15 Jun '12

Hi everyone, >From what I understand after defining a domain with remoteDomainCreateXML, the corresponding destroy call is remoteDomainDestroy, both from the automatically generated remote_client_bodies.h My question is why the domain isn't freed in the destroy procedure (with virUnrefDomain or other) ? Is it supposed to be called separately or is it a bug ? I'm asking because from what I saw the connection refcount increments in domain creation but doesn't decrement in destroy.

2 1

[libvirt] can not connect hypervisor using libvirt-0.9.11 on windows
by xuteng 15 Jun '12

15 Jun '12

Hi,ALL: I compiled the libvirt-0.9.11 source code with mingw on windows according to the website: https://github.com/photron/msys_setup Then I wrote a simple test case ,as below: #include <stdio.h> #include <stdlib.h> #include "libvirt/libvirt.h" int main(int argc,char **argv) { virConnectPtr conn; conn = virConnectOpen("qemu+tcp://192.168.1.6/system"); if (conn = NULL) { fprintf(stderr,"Failed to open connection to qemu+tcp://192.168.1.6/system"); return 1; } virConnectClose(conn); return 0; } When linked to the dll just compiled , it came out with error : NOTE: you should run 'diskperf -y' to enable the disk statistics libvir: RPC error : Unable to set close-on-exec flag: No such file or directory libvir: error : invalid connection pointer in virConnectClose When using gdb to debug the testcase , break the function virConnectOpen , then step ,the error info was: Program received signal ?,Unknown signal. 0x7c92e4ff in ntdll!LdrAlternateResourcesEnabled () from C:\WINDOWS.0\system32\ntdll.dll Any suggestion would be appreciated ! Best Regards! xuteng via foxmail

1 0

[libvirt] [PATCH] virsh: Null terminated the string memcpy from buffer explicitly
by Osier Yang 15 Jun '12

15 Jun '12

--- tools/virsh.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 98305c0..4509020 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -15069,11 +15069,13 @@ cleanup: VIR_FREE(device_type); VIR_FREE(disk_type); if (xml_buf) { - if (VIR_ALLOC_N(ret, xmlBufferLength(xml_buf)) < 0) { + int len = xmlBufferLength(xml_buf); + if (VIR_ALLOC_N(ret, len + 1) < 0) { virReportOOMError(); return NULL; } - memcpy(ret, (char *)xmlBufferContent(xml_buf), xmlBufferLength(xml_buf)); + memcpy(ret, (char *)xmlBufferContent(xml_buf), len); + ret[len] = '\0'; xmlBufferFree(xml_buf); } return ret; -- 1.7.7.3

2 2

[libvirt] [PATCH] Always pivot_root event if the new root source is '/'
by Daniel P. Berrange 14 Jun '12

14 Jun '12

From: "Daniel P. Berrange" <berrange(a)redhat.com> This reverts commit c16b4c43fcdd8ec02581f38377983b2e0925bfcd Author: Daniel P. Berrange <berrange(a)redhat.com> Date: Fri May 11 15:09:27 2012 +0100 Avoid LXC pivot root in the root source is still / This commit broke setup of /dev, because the code which deals with setting up a private /dev and /dev/pts only works if you do a pivotroot. The original intent of avoiding the pivot root was to try and ensure the new root has a minimumal mount tree. The better way todo this is to just unmount the bits we don't want (ie old /proc & /sys subtrees. So apply the logic from commit c529b47a756960d332fbe9903943dae855e7b949 Author: Daniel P. Berrange <berrange(a)redhat.com> Date: Fri May 11 11:35:28 2012 +0100 Trim /proc & /sys subtrees before mounting new instances to the pivot_root codepath as well --- src/lxc/lxc_container.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index bad7938..b8fb14e 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -1345,6 +1345,13 @@ static int lxcContainerSetupPivotRoot(virDomainDefPtr vmDef, if (lxcContainerPivotRoot(root) < 0) goto cleanup; + /* Gets rid of any existing stuff under /proc, since we need new + * namespace aware versions of those. We must do /proc second + * otherwise we won't find /proc/mounts :-) */ + if (lxcContainerUnmountSubtree("/sys", false) < 0 || + lxcContainerUnmountSubtree("/proc", false) < 0) + goto cleanup; + /* Mounts the core /proc, /sys, etc filesystems */ if (lxcContainerMountBasicFS(vmDef, true, securityDriver) < 0) goto cleanup; @@ -1472,11 +1479,7 @@ static int lxcContainerSetupMounts(virDomainDefPtr vmDef, if (lxcContainerResolveSymlinks(vmDef) < 0) return -1; - /* If the user has specified a dst '/' with a source of '/' - * then we don't really want to go down the pivot root - * path, as we're just tuning the existing root - */ - if (root && root->src && STRNEQ(root->src, "/")) + if (root && root->src) return lxcContainerSetupPivotRoot(vmDef, root, ttyPaths, nttyPaths, securityDriver); else return lxcContainerSetupExtraMounts(vmDef, root, securityDriver); -- 1.7.10.2

2 1

[libvirt] [PATCH] Improve error message diagnosing incorrect XML CPU mode
by Daniel P. Berrange 14 Jun '12

14 Jun '12

From: "Daniel P. Berrange" <berrange(a)redhat.com> Tell the user what CPU mode value is wrong --- src/conf/cpu_conf.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/src/conf/cpu_conf.c b/src/conf/cpu_conf.c index 018d571..b520f7c 100644 --- a/src/conf/cpu_conf.c +++ b/src/conf/cpu_conf.c @@ -234,13 +234,15 @@ virCPUDefParseXML(const xmlNodePtr node, goto error; } else { def->mode = virCPUModeTypeFromString(cpuMode); - VIR_FREE(cpuMode); if (def->mode < 0) { - virCPUReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("Invalid mode attribute")); + virCPUReportError(VIR_ERR_INTERNAL_ERROR, + _("Invalid mode attribute '%s'"), + cpuMode); + VIR_FREE(cpuMode); goto error; } + VIR_FREE(cpuMode); } } else { if (def->type == VIR_CPU_TYPE_HOST) -- 1.7.10.2

2 1

Re: [libvirt] [Qemu-devel] [PATCH v2 2/4] qapi: Add passfd QMP command
by Luiz Capitulino 14 Jun '12

14 Jun '12

On Fri, 8 Jun 2012 11:42:57 -0400 Corey Bryant <coreyb(a)linux.vnet.ibm.com> wrote: > This patch adds the passfd QMP command using the QAPI framework. > Like the getfd command, it is used to pass a file descriptor via > SCM_RIGHTS. However, the passfd command also returns the received > file descriptor, which is a difference in behavior from the getfd > command, which returns nothing. Let's call it pass-fd instead. Also, getfd automatically closes a fd if an existing fdname is passed again. I don't think this is a good behavior, I think pass-fd should fail instead (note that we can't fix getfd though). > > The closefd command can be used to close a file descriptor that was > passed with the passfd command. > > v2: > -Introduce new QMP command to pass/return fd (lcapitulino(a)redhat.com) > -Use passfd as command name (berrange(a)redhat.com) > > Signed-off-by: Corey Bryant <coreyb(a)linux.vnet.ibm.com> > --- > monitor.c | 14 ++++++++++---- > qapi-schema.json | 15 +++++++++++++++ > qmp-commands.hx | 27 +++++++++++++++++++++++++++ > 3 files changed, 52 insertions(+), 4 deletions(-) > > diff --git a/monitor.c b/monitor.c > index 4c53cb6..980a226 100644 > --- a/monitor.c > +++ b/monitor.c > @@ -2182,7 +2182,7 @@ static void do_inject_mce(Monitor *mon, const QDict *qdict) > } > #endif > > -void qmp_getfd(const char *fdname, Error **errp) > +int64_t qmp_passfd(const char *fdname, Error **errp) > { > mon_fd_t *monfd; > int fd; > @@ -2190,13 +2190,13 @@ void qmp_getfd(const char *fdname, Error **errp) > fd = qemu_chr_fe_get_msgfd(cur_mon->chr); > if (fd == -1) { > error_set(errp, QERR_FD_NOT_SUPPLIED); > - return; > + return -1; > } > > if (qemu_isdigit(fdname[0])) { > error_set(errp, QERR_INVALID_PARAMETER_VALUE, "fdname", > "a name not starting with a digit"); > - return; > + return -1; > } > > QLIST_FOREACH(monfd, &cur_mon->fds, next) { > @@ -2206,7 +2206,7 @@ void qmp_getfd(const char *fdname, Error **errp) > > close(monfd->fd); > monfd->fd = fd; > - return; > + return fd; > } > > monfd = g_malloc0(sizeof(mon_fd_t)); > @@ -2214,6 +2214,12 @@ void qmp_getfd(const char *fdname, Error **errp) > monfd->fd = fd; > > QLIST_INSERT_HEAD(&cur_mon->fds, monfd, next); > + return fd; > +} > + > +void qmp_getfd(const char *fdname, Error **errp) > +{ > + qmp_passfd(fdname, errp); > return; > } > > diff --git a/qapi-schema.json b/qapi-schema.json > index 6be1d90..15da1b8 100644 > --- a/qapi-schema.json > +++ b/qapi-schema.json > @@ -1864,6 +1864,21 @@ > { 'command': 'netdev_del', 'data': {'id': 'str'} } > > ## > +# @passfd: > +# > +# Pass a file descriptor via SCM rights and assign it a name > +# > +# @fdname: file descriptor name > +# > +# Returns: The QEMU file descriptor that was received > +# If file descriptor was not received, FdNotSupplied > +# If @fdname is not valid, InvalidParameterType > +# > +# Since: 1.2.0 > +## > +{ 'command': 'passfd', 'data': {'fdname': 'str'}, 'returns': 'int' } > + > +## > # @getfd: > # > # Receive a file descriptor via SCM rights and assign it a name > diff --git a/qmp-commands.hx b/qmp-commands.hx > index f8c0f68..338a0b3 100644 > --- a/qmp-commands.hx > +++ b/qmp-commands.hx > @@ -869,6 +869,33 @@ Example: > EQMP > > { > + .name = "passfd", > + .args_type = "fdname:s", > + .params = "passfd name", > + .help = "pass a file descriptor via SCM rights and assign it a name", > + .mhandler.cmd_new = qmp_marshal_input_passfd, > + }, > + > +SQMP > +passfd > +------ > + > +Pass a file descriptor via SCM rights and assign it a name. > + > +Arguments: > + > +- "fdname": file descriptor name (json-string) > + > +Return a json-int with the QEMU file descriptor that was received. > + > +Example: > + > +-> { "execute": "passfd", "arguments": { "fdname": "fd1" } } > +<- { "return": 42 } > + > +EQMP > + > + { > .name = "getfd", > .args_type = "fdname:s", > .params = "getfd name",

3 4

[libvirt] [PATCH] qemu: Enable disconnecting SPICE clients without changing password
by Peter Krempa 14 Jun '12

14 Jun '12

Libvirt updates the configuration of SPICE server only when something changes. This is unfortunate when the user wants to disconnect a existing spice session when the connected attribute is already "disconnect". This patch modifies the conditions for calling the password updater to be called when nothing changes, but the connected attribute is already "disconnect". --- src/qemu/qemu_hotplug.c | 11 ++++++++--- 1 files changed, 8 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 51b8915..c2fa75b 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1539,7 +1539,7 @@ qemuDomainChangeGraphics(struct qemud_driver *driver, _("cannot change listen address setting on spice graphics")); return -1; } - if (STRNEQ_NULLABLE(oldListenNetwork,newListenNetwork)) { + if (STRNEQ_NULLABLE(oldListenNetwork, newListenNetwork)) { qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("cannot change listen network setting on spice graphics")); return -1; @@ -1551,11 +1551,16 @@ qemuDomainChangeGraphics(struct qemud_driver *driver, return -1; } - /* If a password lifetime was, or is set, then we must always run, - * even if new password matches old password */ + /* We must reset the password if it has changed but also if: + * - password lifetime is or was set + * - the requested action has changed + * - the action is "disconnect" + */ if (olddev->data.spice.auth.expires || dev->data.spice.auth.expires || olddev->data.spice.auth.connected != dev->data.spice.auth.connected || + dev->data.spice.auth.connected == + VIR_DOMAIN_GRAPHICS_AUTH_CONNECTED_DISCONNECT || STRNEQ_NULLABLE(olddev->data.spice.auth.passwd, dev->data.spice.auth.passwd)) { VIR_DEBUG("Updating password on SPICE server %p %p", -- 1.7.8.6

2 2

[libvirt] [PATCH v3] virsh: Honor reedit opts printing to a function
by Michal Privoznik 14 Jun '12

14 Jun '12

When printing reedit options we make stdin raw. However, this results in stdout being raw as well. Therefore we need to return carriage when doing new line. Unfortunately, '\r' cannot be part of internationalized messages hence we must move them to formatting string which then in turn become huge and disarranged. To solve this, a new function is introduced which takes variable string arguments and prepend each with "\r\n" just before printing. --- tools/virsh.c | 24 ++++++++++++++++++++---- 1 files changed, 20 insertions(+), 4 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 98305c0..938f8c8 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -655,6 +655,19 @@ vshReconnect(vshControl *ctl) ctl->useSnapshotOld = false; } +static void +vshPrintRaw(vshControl *ctl, ...) +{ + va_list ap; + char *key; + + va_start(ap, ctl); + while ((key = va_arg(ap, char *)) != NULL) { + vshPrint(ctl, "%s\r\n", key); + } + va_end(ap); +} + /** * vshAskReedit: * @msg: Question to ask user @@ -690,10 +703,13 @@ vshAskReedit(vshControl *ctl, const char *msg) c = c_tolower(getchar()); if (c == '?') { - vshPrint(ctl, "\r\n%s", _("y - yes, start editor again\n" - "n - no, throw away my changes\n" - "f - force, try to redefine again\n" - "? - print this help\n")); + vshPrintRaw(ctl, + "", + _("y - yes, start editor again"), + _("n - no, throw away my changes"), + _("f - force, try to redefine again"), + _("? - print this help"), + NULL); continue; } else if (c == 'y' || c == 'n' || c == 'f') { break; -- 1.7.8.5

2 2

[libvirt] [PATCH] virsh: Report name from edited object
by Michal Privoznik 14 Jun '12

14 Jun '12

There is a little easter egg in virsh: one can easily clone an object (domain, network, ...). Just 'virsh edit' change the name and remove <uuid>. And then, in the end when reporting success the new name was printed out. However, with recent edit rewrite we lost the final part and are still printing the original name out. --- Or even better - if name changes, we can choose better message, something like "New domain %s defined". But I am not sure we want to expose this easter egg to users. tools/virsh.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 938f8c8..0b42f1a 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -8612,7 +8612,7 @@ cmdInterfaceEdit(vshControl *ctl, const vshCmd *cmd) #include "virsh-edit.c" vshPrint(ctl, _("Interface %s XML configuration edited.\n"), - virInterfaceGetName(iface)); + virInterfaceGetName(iface_edited)); ret = true; @@ -10006,7 +10006,7 @@ cmdNWFilterEdit(vshControl *ctl, const vshCmd *cmd) #include "virsh-edit.c" vshPrint(ctl, _("Network filter %s XML configuration edited.\n"), - virNWFilterGetName(nwfilter)); + virNWFilterGetName(nwfilter_edited)); ret = true; @@ -15811,7 +15811,7 @@ cmdEdit(vshControl *ctl, const vshCmd *cmd) #include "virsh-edit.c" vshPrint(ctl, _("Domain %s XML configuration edited.\n"), - virDomainGetName(dom)); + virDomainGetName(dom_edited)); ret = true; @@ -15883,7 +15883,7 @@ cmdNetworkEdit(vshControl *ctl, const vshCmd *cmd) #include "virsh-edit.c" vshPrint(ctl, _("Network %s XML configuration edited.\n"), - virNetworkGetName(network)); + virNetworkGetName(network_edited)); ret = true; @@ -15938,7 +15938,7 @@ cmdPoolEdit(vshControl *ctl, const vshCmd *cmd) #include "virsh-edit.c" vshPrint(ctl, _("Pool %s XML configuration edited.\n"), - virStoragePoolGetName(pool)); + virStoragePoolGetName(pool_edited)); ret = true; -- 1.7.8.5

2 2

[libvirt] [PATCH 0/2] fix two issues while doing virsh attach-disk
by Peter Krempa 14 Jun '12

14 Jun '12

First issue was caused by virsh generating invalid XML to attach the disk, and the second issue is a off-by-one error that crashes the daemon with 50-90% probability while doing virsh attach-disk dom /dev/zero vdx. Peter Krempa (2): virsh: Don't generate invalid XML in attach-disk command qemu: Fix off-by-one error while unescaping monitor strings src/qemu/qemu_monitor.c | 11 +++-------- tools/virsh.c | 16 ++++++++-------- 2 files changed, 11 insertions(+), 16 deletions(-) -- 1.7.8.6

2 4

[libvirt] [PATCH v2] virsh: Honor reedit opts printing to a function
by Michal Privoznik 14 Jun '12

14 Jun '12

When printing reedit options we make stdin raw. However, this results in stdout being raw as well. Therefore we need to return carriage when doing new line. Unfortunately, '\r' cannot be part of internationalized messages hence we must move them to formatting string which then in turn become huge and disarranged. To solve this, a new function is introduced which takes variable string arguments and prepend each with "\r\n" just before printing. --- tools/virsh.c | 24 ++++++++++++++++++++---- 1 files changed, 20 insertions(+), 4 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 98305c0..11dc0e7 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -655,6 +655,19 @@ vshReconnect(vshControl *ctl) ctl->useSnapshotOld = false; } +static void +vshPrintRaw(vshControl *ctl, ...) +{ + va_list ap; + char *key; + + va_start(ap, ctl); + while ((key = va_arg(ap, char *)) != NULL) { + vshPrint(ctl, "\r\n%s", key); + } + va_end(ap); +} + /** * vshAskReedit: * @msg: Question to ask user @@ -690,10 +703,13 @@ vshAskReedit(vshControl *ctl, const char *msg) c = c_tolower(getchar()); if (c == '?') { - vshPrint(ctl, "\r\n%s", _("y - yes, start editor again\n" - "n - no, throw away my changes\n" - "f - force, try to redefine again\n" - "? - print this help\n")); + vshPrintRaw(ctl, + _("y - yes, start editor again"), + _("n - no, throw away my changes"), + _("f - force, try to redefine again"), + _("? - print this help"), + "", + NULL); continue; } else if (c == 'y' || c == 'n' || c == 'f') { break; -- 1.7.8.5

1 0

[libvirt] [PATCH 0/8] API to list all snapshots
by Eric Blake 13 Jun '12

13 Jun '12

Requires these 11 patches to be reviewed first: https://www.redhat.com/archives/libvir-list/2012-June/msg00273.html https://www.redhat.com/archives/libvir-list/2012-June/msg00267.html https://www.redhat.com/archives/libvir-list/2012-May/msg01196.html I'm still working on at least two more patches (actually wiring up domain_conf.c to create the snapshot list, and then fixing qemu_driver.c to use the new domain_conf.c helpers), but wanted to get this out for review, especially since it has some interaction with Peter's series for listing all domains (translation: I've tried to make sure this series complies with the review comments I made against his series): https://www.redhat.com/archives/libvir-list/2012-June/msg00113.html I suppose that means that I technically haven't done any testing other than compilation so far, at least until I post the last two patches for qemu... And I suppose I also ought to try my hand at vbox and esx patches, even though I can't test those (it would be basically copy-and-paste from existing code in those drivers). And if I'm really up to it, I'll add snapshot support to test:///default. Eric Blake (6): snapshot: add additional filters when getting lists snapshot: expose new flags in virsh list: add virDomainListAllSnapshots API list: use the new snapshot API in virsh when possible list: provide python bindings for snapshots list: provide RPC call for snapshots daemon/remote.c | 126 ++++++++++++ include/libvirt/libvirt.h.in | 35 +++- python/Makefile.am | 2 + python/generator.py | 2 + python/libvirt-override-api.xml | 16 +- python/libvirt-override-virDomain.py | 11 + python/libvirt-override-virDomainSnapshot.py | 11 + python/libvirt-override.c | 90 +++++++++ src/conf/domain_conf.c | 53 ++++- src/driver.h | 12 ++ src/libvirt.c | 281 ++++++++++++++++++++++---- src/libvirt_public.syms | 2 + src/remote/remote_driver.c | 126 ++++++++++++ src/remote/remote_protocol.x | 26 ++- src/remote_protocol-structs | 26 +++ tools/virsh.c | 83 +++++++- tools/virsh.pod | 14 +- 17 files changed, 849 insertions(+), 67 deletions(-) create mode 100644 python/libvirt-override-virDomain.py create mode 100644 python/libvirt-override-virDomainSnapshot.py -- 1.7.10.2

2 14

[libvirt] [PATCH 0/6] refactor 'virsh snapshot-list'
by Eric Blake 13 Jun '12

13 Jun '12

I'm in the middle of writing a patch series for virDomainListAllSnapshots; this series is preliminary to that, and I'd like to get the review started now. The end result of this series is a net gain in lines of code, but a much nicer framework for changing just one point in the code to use new APIs and having everything else continue to work as before. Eric Blake (6): buf: support peeking at string contents virsh: remove limits on tree listing virsh: make tree listing more flexible snapshot: virsh indentation cleanup snapshot: new virsh function factored from snapshot-list snapshot: use new virsh function for snapshot-list src/libvirt_private.syms | 1 + src/util/buf.c | 22 +- src/util/buf.h | 1 + tests/virbuftest.c | 8 + tools/virsh.c | 735 +++++++++++++++++++++++++++------------------- 5 files changed, 466 insertions(+), 301 deletions(-) -- 1.7.10.2

2 22

Re: [libvirt] [Qemu-devel] [PATCH v2 1/4] qapi: Convert getfd and closefd
by Luiz Capitulino 13 Jun '12

13 Jun '12

On Fri, 8 Jun 2012 11:42:56 -0400 Corey Bryant <coreyb(a)linux.vnet.ibm.com> wrote: > v2: > - Convert getfd and closefd to QAPI (lcapitulino(a)redhat.com) > - Remove changes that returned fd from getfd (lcapitulino(a)redhat.com) > - Wrap hmp_* functions around qmp_* functions (kwolf(a)redhat.com) > - Move hmp_* functions to hmp.c (lcapitulino(a)redhat.com) > - Drop .user_print lines (lcapitulino(a)redhat.com) > - Use 'cmd' instead of 'cmd_new' for HMP (lcapitulino(a)redhat.com) > - Change QMP command existance back to 0.14 (lcapitulino(a)redhat.com) Btw, having the changelog like this is not nice because it becomes part of the history. It's better to move it after the '---' line, so that git ignores it. > > Signed-off-by: Corey Bryant <coreyb(a)linux.vnet.ibm.com> > --- > hmp-commands.hx | 6 ++---- > hmp.c | 18 ++++++++++++++++++ > hmp.h | 2 ++ > monitor.c | 34 ++++++++++++++++------------------ > qapi-schema.json | 29 +++++++++++++++++++++++++++++ > qmp-commands.hx | 6 ++---- > 6 files changed, 69 insertions(+), 26 deletions(-) > > diff --git a/hmp-commands.hx b/hmp-commands.hx > index f5d9d91..eea8b32 100644 > --- a/hmp-commands.hx > +++ b/hmp-commands.hx > @@ -1236,8 +1236,7 @@ ETEXI > .args_type = "fdname:s", > .params = "getfd name", > .help = "receive a file descriptor via SCM rights and assign it a name", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_getfd, > + .mhandler.cmd = hmp_getfd, > }, > > STEXI > @@ -1253,8 +1252,7 @@ ETEXI > .args_type = "fdname:s", > .params = "closefd name", > .help = "close a file descriptor previously passed via SCM rights", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_closefd, > + .mhandler.cmd = hmp_closefd, > }, > > STEXI > diff --git a/hmp.c b/hmp.c > index 2ce8cb9..6241856 100644 > --- a/hmp.c > +++ b/hmp.c > @@ -999,3 +999,21 @@ void hmp_netdev_del(Monitor *mon, const QDict *qdict) > qmp_netdev_del(id, &err); > hmp_handle_error(mon, &err); > } > + > +void hmp_getfd(Monitor *mon, const QDict *qdict) > +{ > + const char *fdname = qdict_get_str(qdict, "fdname"); > + Error *errp = NULL; > + > + qmp_getfd(fdname, &errp); > + hmp_handle_error(mon, &errp); > +} > + > +void hmp_closefd(Monitor *mon, const QDict *qdict) > +{ > + const char *fdname = qdict_get_str(qdict, "fdname"); > + Error *errp = NULL; > + > + qmp_closefd(fdname, &errp); > + hmp_handle_error(mon, &errp); > +} > diff --git a/hmp.h b/hmp.h > index 79d138d..8d2b0d7 100644 > --- a/hmp.h > +++ b/hmp.h > @@ -64,5 +64,7 @@ void hmp_device_del(Monitor *mon, const QDict *qdict); > void hmp_dump_guest_memory(Monitor *mon, const QDict *qdict); > void hmp_netdev_add(Monitor *mon, const QDict *qdict); > void hmp_netdev_del(Monitor *mon, const QDict *qdict); > +void hmp_getfd(Monitor *mon, const QDict *qdict); > +void hmp_closefd(Monitor *mon, const QDict *qdict); > > #endif > diff --git a/monitor.c b/monitor.c > index a3bc2c7..4c53cb6 100644 > --- a/monitor.c > +++ b/monitor.c > @@ -2182,48 +2182,46 @@ static void do_inject_mce(Monitor *mon, const QDict *qdict) > } > #endif > > -static int do_getfd(Monitor *mon, const QDict *qdict, QObject **ret_data) > +void qmp_getfd(const char *fdname, Error **errp) > { > - const char *fdname = qdict_get_str(qdict, "fdname"); > mon_fd_t *monfd; > int fd; > > - fd = qemu_chr_fe_get_msgfd(mon->chr); > + fd = qemu_chr_fe_get_msgfd(cur_mon->chr); > if (fd == -1) { > - qerror_report(QERR_FD_NOT_SUPPLIED); > - return -1; > + error_set(errp, QERR_FD_NOT_SUPPLIED); > + return; > } > > if (qemu_isdigit(fdname[0])) { > - qerror_report(QERR_INVALID_PARAMETER_VALUE, "fdname", > - "a name not starting with a digit"); > - return -1; > + error_set(errp, QERR_INVALID_PARAMETER_VALUE, "fdname", > + "a name not starting with a digit"); > + return; > } > > - QLIST_FOREACH(monfd, &mon->fds, next) { > + QLIST_FOREACH(monfd, &cur_mon->fds, next) { > if (strcmp(monfd->name, fdname) != 0) { > continue; > } > > close(monfd->fd); > monfd->fd = fd; > - return 0; > + return; > } > > monfd = g_malloc0(sizeof(mon_fd_t)); > monfd->name = g_strdup(fdname); > monfd->fd = fd; > > - QLIST_INSERT_HEAD(&mon->fds, monfd, next); > - return 0; > + QLIST_INSERT_HEAD(&cur_mon->fds, monfd, next); > + return; > } > > -static int do_closefd(Monitor *mon, const QDict *qdict, QObject **ret_data) > +void qmp_closefd(const char *fdname, Error **errp) > { > - const char *fdname = qdict_get_str(qdict, "fdname"); > mon_fd_t *monfd; > > - QLIST_FOREACH(monfd, &mon->fds, next) { > + QLIST_FOREACH(monfd, &cur_mon->fds, next) { > if (strcmp(monfd->name, fdname) != 0) { > continue; > } > @@ -2232,11 +2230,11 @@ static int do_closefd(Monitor *mon, const QDict *qdict, QObject **ret_data) > close(monfd->fd); > g_free(monfd->name); > g_free(monfd); > - return 0; > + return; > } > > - qerror_report(QERR_FD_NOT_FOUND, fdname); > - return -1; > + error_set(errp, QERR_FD_NOT_FOUND, fdname); > + return; > } > > static void do_loadvm(Monitor *mon, const QDict *qdict) > diff --git a/qapi-schema.json b/qapi-schema.json > index 3b6e346..6be1d90 100644 > --- a/qapi-schema.json > +++ b/qapi-schema.json > @@ -1862,3 +1862,32 @@ > # Since: 0.14.0 > ## > { 'command': 'netdev_del', 'data': {'id': 'str'} } > + > +## > +# @getfd: > +# > +# Receive a file descriptor via SCM rights and assign it a name > +# > +# @fdname: file descriptor name > +# > +# Returns: Nothing on success > +# If file descriptor was not received, FdNotSupplied > +# If @fdname is not valid, InvalidParameterType > +# > +# Since: 0.14.0 > +## > +{ 'command': 'getfd', 'data': {'fdname': 'str'} } > + > +## > +# @closefd: > +# > +# Close a file descriptor previously passed via SCM rights > +# > +# @fdname: file descriptor name > +# > +# Returns: Nothing on success > +# If @fdname is not found, FdNotFound > +# > +# Since: 0.14.0 > +## > +{ 'command': 'closefd', 'data': {'fdname': 'str'} } > diff --git a/qmp-commands.hx b/qmp-commands.hx > index 2e1a38e..f8c0f68 100644 > --- a/qmp-commands.hx > +++ b/qmp-commands.hx > @@ -873,8 +873,7 @@ EQMP > .args_type = "fdname:s", > .params = "getfd name", > .help = "receive a file descriptor via SCM rights and assign it a name", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_getfd, > + .mhandler.cmd_new = qmp_marshal_input_getfd, > }, > > SQMP > @@ -899,8 +898,7 @@ EQMP > .args_type = "fdname:s", > .params = "closefd name", > .help = "close a file descriptor previously passed via SCM rights", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_closefd, > + .mhandler.cmd_new = qmp_marshal_input_closefd, > }, > > SQMP

3 4

Re: [libvirt] [Qemu-devel] [PATCH v2 1/4] qapi: Convert getfd and closefd
by Luiz Capitulino 13 Jun '12

13 Jun '12

On Fri, 8 Jun 2012 11:42:56 -0400 Corey Bryant <coreyb(a)linux.vnet.ibm.com> wrote: > v2: > - Convert getfd and closefd to QAPI (lcapitulino(a)redhat.com) > - Remove changes that returned fd from getfd (lcapitulino(a)redhat.com) > - Wrap hmp_* functions around qmp_* functions (kwolf(a)redhat.com) > - Move hmp_* functions to hmp.c (lcapitulino(a)redhat.com) > - Drop .user_print lines (lcapitulino(a)redhat.com) > - Use 'cmd' instead of 'cmd_new' for HMP (lcapitulino(a)redhat.com) > - Change QMP command existance back to 0.14 (lcapitulino(a)redhat.com) > > Signed-off-by: Corey Bryant <coreyb(a)linux.vnet.ibm.com> > --- > hmp-commands.hx | 6 ++---- > hmp.c | 18 ++++++++++++++++++ > hmp.h | 2 ++ > monitor.c | 34 ++++++++++++++++------------------ > qapi-schema.json | 29 +++++++++++++++++++++++++++++ > qmp-commands.hx | 6 ++---- > 6 files changed, 69 insertions(+), 26 deletions(-) > > diff --git a/hmp-commands.hx b/hmp-commands.hx > index f5d9d91..eea8b32 100644 > --- a/hmp-commands.hx > +++ b/hmp-commands.hx > @@ -1236,8 +1236,7 @@ ETEXI > .args_type = "fdname:s", > .params = "getfd name", > .help = "receive a file descriptor via SCM rights and assign it a name", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_getfd, > + .mhandler.cmd = hmp_getfd, > }, > > STEXI > @@ -1253,8 +1252,7 @@ ETEXI > .args_type = "fdname:s", > .params = "closefd name", > .help = "close a file descriptor previously passed via SCM rights", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_closefd, > + .mhandler.cmd = hmp_closefd, > }, > > STEXI > diff --git a/hmp.c b/hmp.c > index 2ce8cb9..6241856 100644 > --- a/hmp.c > +++ b/hmp.c > @@ -999,3 +999,21 @@ void hmp_netdev_del(Monitor *mon, const QDict *qdict) > qmp_netdev_del(id, &err); > hmp_handle_error(mon, &err); > } > + > +void hmp_getfd(Monitor *mon, const QDict *qdict) > +{ > + const char *fdname = qdict_get_str(qdict, "fdname"); > + Error *errp = NULL; > + > + qmp_getfd(fdname, &errp); > + hmp_handle_error(mon, &errp); > +} > + > +void hmp_closefd(Monitor *mon, const QDict *qdict) > +{ > + const char *fdname = qdict_get_str(qdict, "fdname"); > + Error *errp = NULL; > + > + qmp_closefd(fdname, &errp); > + hmp_handle_error(mon, &errp); > +} > diff --git a/hmp.h b/hmp.h > index 79d138d..8d2b0d7 100644 > --- a/hmp.h > +++ b/hmp.h > @@ -64,5 +64,7 @@ void hmp_device_del(Monitor *mon, const QDict *qdict); > void hmp_dump_guest_memory(Monitor *mon, const QDict *qdict); > void hmp_netdev_add(Monitor *mon, const QDict *qdict); > void hmp_netdev_del(Monitor *mon, const QDict *qdict); > +void hmp_getfd(Monitor *mon, const QDict *qdict); > +void hmp_closefd(Monitor *mon, const QDict *qdict); > > #endif > diff --git a/monitor.c b/monitor.c > index a3bc2c7..4c53cb6 100644 > --- a/monitor.c > +++ b/monitor.c > @@ -2182,48 +2182,46 @@ static void do_inject_mce(Monitor *mon, const QDict *qdict) > } > #endif > > -static int do_getfd(Monitor *mon, const QDict *qdict, QObject **ret_data) > +void qmp_getfd(const char *fdname, Error **errp) > { > - const char *fdname = qdict_get_str(qdict, "fdname"); > mon_fd_t *monfd; > int fd; > > - fd = qemu_chr_fe_get_msgfd(mon->chr); > + fd = qemu_chr_fe_get_msgfd(cur_mon->chr); > if (fd == -1) { > - qerror_report(QERR_FD_NOT_SUPPLIED); > - return -1; > + error_set(errp, QERR_FD_NOT_SUPPLIED); > + return; > } > > if (qemu_isdigit(fdname[0])) { > - qerror_report(QERR_INVALID_PARAMETER_VALUE, "fdname", > - "a name not starting with a digit"); > - return -1; > + error_set(errp, QERR_INVALID_PARAMETER_VALUE, "fdname", > + "a name not starting with a digit"); > + return; > } > > - QLIST_FOREACH(monfd, &mon->fds, next) { > + QLIST_FOREACH(monfd, &cur_mon->fds, next) { > if (strcmp(monfd->name, fdname) != 0) { > continue; > } > > close(monfd->fd); > monfd->fd = fd; > - return 0; > + return; > } > > monfd = g_malloc0(sizeof(mon_fd_t)); > monfd->name = g_strdup(fdname); > monfd->fd = fd; > > - QLIST_INSERT_HEAD(&mon->fds, monfd, next); > - return 0; > + QLIST_INSERT_HEAD(&cur_mon->fds, monfd, next); > + return; You have a few of this kind of gratuitous return, please drop 'em. > } > > -static int do_closefd(Monitor *mon, const QDict *qdict, QObject **ret_data) > +void qmp_closefd(const char *fdname, Error **errp) > { > - const char *fdname = qdict_get_str(qdict, "fdname"); > mon_fd_t *monfd; > > - QLIST_FOREACH(monfd, &mon->fds, next) { > + QLIST_FOREACH(monfd, &cur_mon->fds, next) { > if (strcmp(monfd->name, fdname) != 0) { > continue; > } > @@ -2232,11 +2230,11 @@ static int do_closefd(Monitor *mon, const QDict *qdict, QObject **ret_data) > close(monfd->fd); > g_free(monfd->name); > g_free(monfd); > - return 0; > + return; > } > > - qerror_report(QERR_FD_NOT_FOUND, fdname); > - return -1; > + error_set(errp, QERR_FD_NOT_FOUND, fdname); > + return; > } > > static void do_loadvm(Monitor *mon, const QDict *qdict) > diff --git a/qapi-schema.json b/qapi-schema.json > index 3b6e346..6be1d90 100644 > --- a/qapi-schema.json > +++ b/qapi-schema.json > @@ -1862,3 +1862,32 @@ > # Since: 0.14.0 > ## > { 'command': 'netdev_del', 'data': {'id': 'str'} } > + > +## > +# @getfd: > +# > +# Receive a file descriptor via SCM rights and assign it a name > +# > +# @fdname: file descriptor name > +# > +# Returns: Nothing on success > +# If file descriptor was not received, FdNotSupplied > +# If @fdname is not valid, InvalidParameterType Please, add a note explaining that getfd automatically closes fds if fdname matches an existing name. Otherwise this conversion looks very good to me. > +# > +# Since: 0.14.0 > +## > +{ 'command': 'getfd', 'data': {'fdname': 'str'} } > + > +## > +# @closefd: > +# > +# Close a file descriptor previously passed via SCM rights > +# > +# @fdname: file descriptor name > +# > +# Returns: Nothing on success > +# If @fdname is not found, FdNotFound > +# > +# Since: 0.14.0 > +## > +{ 'command': 'closefd', 'data': {'fdname': 'str'} } > diff --git a/qmp-commands.hx b/qmp-commands.hx > index 2e1a38e..f8c0f68 100644 > --- a/qmp-commands.hx > +++ b/qmp-commands.hx > @@ -873,8 +873,7 @@ EQMP > .args_type = "fdname:s", > .params = "getfd name", > .help = "receive a file descriptor via SCM rights and assign it a name", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_getfd, > + .mhandler.cmd_new = qmp_marshal_input_getfd, > }, > > SQMP > @@ -899,8 +898,7 @@ EQMP > .args_type = "fdname:s", > .params = "closefd name", > .help = "close a file descriptor previously passed via SCM rights", > - .user_print = monitor_user_noop, > - .mhandler.cmd_new = do_closefd, > + .mhandler.cmd_new = qmp_marshal_input_closefd, > }, > > SQMP

2 1

[libvirt] [PATCH] network: fully support/use VIR_NETWORK_XML_INACTIVE flag
by Laine Stump 13 Jun '12

13 Jun '12

commit 52d064f42dbc857f4096dc60c0335395ffac73aa added VIR_NETWORK_XML_INACTIVE in order to allow suppressing the auto-generated list of VFs in network definitions, and a --inactive flag to virsh net-dumpxml to take advantage of the flag. However, it missed out on two oppurtunities: 1) Use INACTIVE to get the current config of the network as it exists on disk, rather than the currently active config. 2) Add INACTIVE to the flags used for the virsh net-edit command, so that it won't include the forward-pool interfaces that were autogenerated, and so that a re-edit of the network prior to restarting it will show any other edits made since the last restart of the network. (prior to this patch, if you edited a network a 2nd time without restarting, all of the previous edits would magically disappear). In order to fit with the new #define-based generic edit function in virsh.c, a new function vshNetworkGetXMLDesc() was added. This function first tries to call virNetworkGetXMLDesc with the INACTIVE flag added, then retries without if the first attempt fails (in the manner expected when the server doesn't support it)., --- src/network/bridge_driver.c | 8 +++++++- tools/virsh.c | 20 ++++++++++++++++++-- 2 files changed, 25 insertions(+), 3 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 5f3132c..79d3010 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -2533,6 +2533,7 @@ static char *networkGetXMLDesc(virNetworkPtr net, { struct network_driver *driver = net->conn->networkPrivateData; virNetworkObjPtr network; + virNetworkDefPtr def; char *ret = NULL; virCheckFlags(VIR_NETWORK_XML_INACTIVE, NULL); @@ -2547,7 +2548,12 @@ static char *networkGetXMLDesc(virNetworkPtr net, goto cleanup; } - ret = virNetworkDefFormat(network->def, flags); + if ((flags & VIR_NETWORK_XML_INACTIVE) && network->newDef) + def = network->newDef; + else + def = network->def; + + ret = virNetworkDefFormat(def, flags); cleanup: if (network) diff --git a/tools/virsh.c b/tools/virsh.c index 0453b95..1c9e44a 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -15806,13 +15806,29 @@ static const vshCmdOptDef opts_network_edit[] = { {NULL, 0, 0, NULL} }; +static char *vshNetworkGetXMLDesc(virNetworkPtr network) +{ + unsigned int flags = VIR_NETWORK_XML_INACTIVE; + char *doc = virNetworkGetXMLDesc(network, flags); + + if (!doc && last_error->code == VIR_ERR_INVALID_ARG) { + /* The server side libvirt doesn't support + * VIR_NETWORK_XML_INACTIVE, so retry without it. + */ + virFreeError(last_error); + last_error = NULL; + flags &= ~VIR_NETWORK_XML_INACTIVE; + doc = virNetworkGetXMLDesc(network, flags); + } + return doc; +} + static bool cmdNetworkEdit(vshControl *ctl, const vshCmd *cmd) { bool ret = false; virNetworkPtr network = NULL; virNetworkPtr network_edited = NULL; - unsigned int flags = 0; if (!vshConnectionUsability(ctl, ctl->conn)) goto cleanup; @@ -15821,7 +15837,7 @@ cmdNetworkEdit(vshControl *ctl, const vshCmd *cmd) if (network == NULL) goto cleanup; -#define EDIT_GET_XML virNetworkGetXMLDesc(network, flags) +#define EDIT_GET_XML vshNetworkGetXMLDesc(network) #define EDIT_NOT_CHANGED \ vshPrint(ctl, _("Network %s XML configuration not changed.\n"), \ virNetworkGetName (network)); \ -- 1.7.10.2

2 2

[libvirt] [PATCH] storage backend: Add sheepdog support
by Sebastian Wiedenroth 13 Jun '12

13 Jun '12

Signed-off-by: Sebastian Wiedenroth <wiedi(a)frubar.net> --- This patch adds support for sheepdog pools and volumes. Thanks go to Frank Spijkerman who started the initial work on this and Wido den Hollander for his RBD support on which this is partly based. configure.ac | 23 ++ docs/drivers.html.in | 1 + docs/schemas/storagepool.rng | 17 ++ docs/schemas/storagevol.rng | 3 +- docs/storage.html.in | 62 +++++ po/POTFILES.in | 1 + src/Makefile.am | 8 + src/conf/storage_conf.c | 17 +- src/conf/storage_conf.h | 1 + src/storage/storage_backend.c | 6 + src/storage/storage_backend_sheepdog.c | 323 +++++++++++++++++++++++++ src/storage/storage_backend_sheepdog.h | 39 +++ tests/Makefile.am | 13 + tests/storagebackendsheepdogtest.c | 196 +++++++++++++++ tests/storagepoolxml2xmlin/pool-sheepdog.xml | 8 + tests/storagepoolxml2xmlout/pool-sheepdog.xml | 11 + tests/storagepoolxml2xmltest.c | 1 + tests/storagevolxml2xmlin/vol-sheepdog.xml | 10 + tests/storagevolxml2xmlout/vol-sheepdog.xml | 17 ++ tests/storagevolxml2xmltest.c | 1 + tools/virsh.c | 3 + 21 files changed, 757 insertions(+), 4 deletions(-) create mode 100644 src/storage/storage_backend_sheepdog.c create mode 100644 src/storage/storage_backend_sheepdog.h create mode 100644 tests/storagebackendsheepdogtest.c create mode 100644 tests/storagepoolxml2xmlin/pool-sheepdog.xml create mode 100644 tests/storagepoolxml2xmlout/pool-sheepdog.xml create mode 100644 tests/storagevolxml2xmlin/vol-sheepdog.xml create mode 100644 tests/storagevolxml2xmlout/vol-sheepdog.xml diff --git a/configure.ac b/configure.ac index dda764d..774a9dd 100644 --- a/configure.ac +++ b/configure.ac @@ -1807,6 +1807,8 @@ AC_ARG_WITH([storage-disk], AC_HELP_STRING([--with-storage-disk], [with GPartd Disk backend for the storage driver @<:@default=check@:>@]),[],[with_storage_disk=check]) AC_ARG_WITH([storage-rbd], AC_HELP_STRING([--with-storage-rbd], [with RADOS Block Device backend for the storage driver @<:@default=check@:>@]),[],[with_storage_rbd=check]) +AC_ARG_WITH([storage-sheepdog], + AC_HELP_STRING([--with-storage-sheepdog], [with Sheepdog backend for the storage driver @<:@default=check@:>@]),[],[with_storage_sheepdog=check]) if test "$with_libvirtd" = "no"; then with_storage_dir=no @@ -1817,6 +1819,7 @@ if test "$with_libvirtd" = "no"; then with_storage_mpath=no with_storage_disk=no with_storage_rbd=no + with_storage_sheepdog=no fi if test "$with_storage_dir" = "yes" ; then AC_DEFINE_UNQUOTED([WITH_STORAGE_DIR], 1, [whether directory backend for storage driver is enabled]) @@ -1991,6 +1994,25 @@ fi AM_CONDITIONAL([WITH_STORAGE_RBD], [test "$with_storage_rbd" = "yes"]) AC_SUBST([LIBRBD_LIBS]) +if test "$with_storage_sheepdog" = "yes" || test "$with_storage_sheepdog" = "check"; then + AC_PATH_PROG([COLLIE], [collie], [], [$PATH:/sbin:/usr/sbin]) + + if test "$with_storage_sheepdog" = "yes" ; then + if test -z "$COLLIE" ; then AC_MSG_ERROR([We need collie for Sheepdog storage driver]) ; fi + else + if test -z "$COLLIE" ; then with_storage_sheepdog=no ; fi + + if test "$with_storage_sheepdog" = "check" ; then with_storage_sheepdog=yes ; fi + fi + + if test "$with_storage_sheepdog" = "yes" ; then + AC_DEFINE_UNQUOTED([WITH_STORAGE_SHEEPDOG], 1, [whether Sheepdog backend for storage driver is enabled]) + AC_DEFINE_UNQUOTED([COLLIE],["$COLLIE"],[Location of collie program]) + fi +fi +AM_CONDITIONAL([WITH_STORAGE_SHEEPDOG], [test "$with_storage_sheepdog" = "yes"]) + + LIBPARTED_CFLAGS= LIBPARTED_LIBS= if test "$with_storage_disk" = "yes" || @@ -2806,6 +2828,7 @@ AC_MSG_NOTICE([ SCSI: $with_storage_scsi]) AC_MSG_NOTICE([ mpath: $with_storage_mpath]) AC_MSG_NOTICE([ Disk: $with_storage_disk]) AC_MSG_NOTICE([ RBD: $with_storage_rbd]) +AC_MSG_NOTICE([Sheepdog: $with_storage_sheepdog]) AC_MSG_NOTICE([]) AC_MSG_NOTICE([Security Drivers]) AC_MSG_NOTICE([]) diff --git a/docs/drivers.html.in b/docs/drivers.html.in index 8ad2c33..90b6196 100644 --- a/docs/drivers.html.in +++ b/docs/drivers.html.in @@ -43,6 +43,7 @@ <li><strong><a href="storage.html#StorageBackendSCSI">SCSI backend</a></strong></li> <li><strong><a href="storage.html#StorageBackendMultipath">Multipath backend</a></strong></li> <li><strong><a href="storage.html#StorageBackendRBD">RBD (RADOS Block Device) backend</a></strong></li> + <li><strong><a href="storage.html#StorageBackendSheepdog">Sheepdog backend</a></strong></li> </ul> </body> </html> diff --git a/docs/schemas/storagepool.rng b/docs/schemas/storagepool.rng index 75b6b51..039798a 100644 --- a/docs/schemas/storagepool.rng +++ b/docs/schemas/storagepool.rng @@ -20,6 +20,7 @@ <ref name='poolscsi'/> <ref name='poolmpath'/> <ref name='poolrbd'/> + <ref name='poolsheepdog'/> </choice> </element> </define> @@ -115,6 +116,15 @@ <ref name='sourcerbd'/> </define> + <define name='poolsheepdog'> + <attribute name='type'> + <value>sheepdog</value> + </attribute> + <ref name='commonmetadata'/> + <ref name='sizing'/> + <ref name='sourcesheepdog'/> + </define> + <define name='sourceinfovendor'> <optional> <element name='vendor'> @@ -496,6 +506,13 @@ </element> </define> + <define name='sourcesheepdog'> + <element name='source'> + <ref name='sourceinfohost'/> + <ref name='sourceinfoname'/> + </element> + </define> + <define name='name'> <data type='string'> <param name="pattern">[a-zA-Z0-9_\+\-]+</param> diff --git a/docs/schemas/storagevol.rng b/docs/schemas/storagevol.rng index 8edb877..83148ca 100644 --- a/docs/schemas/storagevol.rng +++ b/docs/schemas/storagevol.rng @@ -67,7 +67,7 @@ <element name='target'> <optional> <element name='path'> - <ref name='absFilePath'/> + <data type='anyURI'/> </element> </optional> <ref name='format'/> @@ -144,6 +144,7 @@ <define name='formatfile'> <choice> + <value>unknown</value> <value>raw</value> <value>dir</value> <value>bochs</value> diff --git a/docs/storage.html.in b/docs/storage.html.in index b3484e8..8730164 100644 --- a/docs/storage.html.in +++ b/docs/storage.html.in @@ -110,6 +110,9 @@ <li> <a href="#StorageBackendRBD">RBD (RADOS Block Device) backend</a> </li> + <li> + <a href="#StorageBackendSheepdog">Sheepdog backend</a> + </li> </ul> <h2><a name="StorageBackendDir">Directory pool</a></h2> @@ -565,5 +568,64 @@ The RBD pool does not use the volume format type element. </p> + <h2><a name="StorageBackendSheepdog">Sheepdog pools</a></h2> + <p> + This provides a pool based on a Sheepdog Cluster. + Sheepdog is a distributed storage system for QEMU/KVM. + It provides highly available block level storage volumes that + can be attached to QEMU/KVM virtual machines. + + The cluster must already be formated. + + <span class="since">Since 0.9.13</span> + </p> + + <h3>Example pool input</h3> + <pre> + <pool type="sheepdog"> + <name>mysheeppool</name> + <source> + <name>mysheeppool</name> + <host name='localhost' port='7000'/> + </source> + </pool></pre> + + <h3>Example volume output</h3> + <pre> + <volume> + <name>myvol</name> + <key>sheep/myvol</key> + <source> + </source> + <capacity unit='bytes'>53687091200</capacity> + <allocation unit='bytes'>53687091200</allocation> + <target> + <path>sheepdog:myvol</path> + <format type='unknown'/> + <permissions> + <mode>00</mode> + <owner>0</owner> + <group>0</group> + </permissions> + </target> + </volume></pre> + + <h3>Example disk attachement</h3> + <p>Sheepdog images can be attached to Qemu guests. + Information about attaching a Sheepdog image to a + guest can be found + at the <a href="formatdomain.html#elementsDisks">format domain</a> + page.</p> + + <h3>Valid pool format types</h3> + <p> + The Sheepdog pool does not use the pool format type element. + </p> + + <h3>Valid volume format types</h3> + <p> + The Sheepdog pool does not use the volume format type element. + </p> + </body> </html> diff --git a/po/POTFILES.in b/po/POTFILES.in index 31246f7..33a2ace 100644 --- a/po/POTFILES.in +++ b/po/POTFILES.in @@ -108,6 +108,7 @@ src/storage/storage_backend_logical.c src/storage/storage_backend_mpath.c src/storage/storage_backend_rbd.c src/storage/storage_backend_scsi.c +src/storage/storage_backend_sheepdog.c src/storage/storage_driver.c src/test/test_driver.c src/uml/uml_conf.c diff --git a/src/Makefile.am b/src/Makefile.am index 60f5442..82238b9 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -506,6 +506,9 @@ STORAGE_DRIVER_DISK_SOURCES = \ STORAGE_DRIVER_RBD_SOURCES = \ storage/storage_backend_rbd.h storage/storage_backend_rbd.c +STORAGE_DRIVER_SHEEPDOG_SOURCES = \ + storage/storage_backend_sheepdog.h storage/storage_backend_sheepdog.c + STORAGE_HELPER_DISK_SOURCES = \ storage/parthelper.c @@ -1009,6 +1012,10 @@ libvirt_driver_storage_la_SOURCES += $(STORAGE_DRIVER_RBD_SOURCES) libvirt_driver_storage_la_LIBADD += $(LIBRBD_LIBS) endif +if WITH_STORAGE_SHEEPDOG +libvirt_driver_storage_la_SOURCES += $(STORAGE_DRIVER_SHEEPDOG_SOURCES) +endif + if WITH_NODE_DEVICES # Needed to keep automake quiet about conditionals if WITH_DRIVER_MODULES @@ -1107,6 +1114,7 @@ EXTRA_DIST += \ $(STORAGE_DRIVER_MPATH_SOURCES) \ $(STORAGE_DRIVER_DISK_SOURCES) \ $(STORAGE_DRIVER_RBD_SOURCES) \ + $(STORAGE_DRIVER_SHEEPDOG_SOURCES) \ $(NODE_DEVICE_DRIVER_SOURCES) \ $(NODE_DEVICE_DRIVER_HAL_SOURCES) \ $(NODE_DEVICE_DRIVER_UDEV_SOURCES) \ diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index bf4567f..8ca6b1e 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -52,7 +52,7 @@ VIR_ENUM_IMPL(virStoragePool, VIR_STORAGE_POOL_LAST, "dir", "fs", "netfs", "logical", "disk", "iscsi", - "scsi", "mpath", "rbd") + "scsi", "mpath", "rbd", "sheepdog") VIR_ENUM_IMPL(virStoragePoolFormatFileSystem, VIR_STORAGE_POOL_FS_LAST, @@ -206,6 +206,17 @@ static virStoragePoolTypeInfo poolTypeInfo[] = { .formatToString = virStoragePoolFormatDiskTypeToString, } }, + { .poolType = VIR_STORAGE_POOL_SHEEPDOG, + .poolOptions = { + .flags = (VIR_STORAGE_POOL_SOURCE_HOST | + VIR_STORAGE_POOL_SOURCE_NETWORK | + VIR_STORAGE_POOL_SOURCE_NAME), + }, + .volOptions = { + .defaultFormat = VIR_STORAGE_FILE_RAW, + .formatToString = virStoragePoolFormatDiskTypeToString, + } + }, { .poolType = VIR_STORAGE_POOL_MPATH, .volOptions = { .formatToString = virStoragePoolFormatDiskTypeToString, @@ -1011,9 +1022,9 @@ virStoragePoolDefFormat(virStoragePoolDefPtr def) { if (virStoragePoolSourceFormat(&buf, options, &def->source) < 0) goto cleanup; - /* RBD devices are not local block devs nor files, so it doesn't + /* RBD and Sheepdog devices are not local block devs nor files, so it doesn't * have a target */ - if (def->type != VIR_STORAGE_POOL_RBD) { + if (def->type != VIR_STORAGE_POOL_RBD && def->type != VIR_STORAGE_POOL_SHEEPDOG) { virBufferAddLit(&buf," <target>\n"); if (def->target.path) diff --git a/src/conf/storage_conf.h b/src/conf/storage_conf.h index 5733b57..dcf976f 100644 --- a/src/conf/storage_conf.h +++ b/src/conf/storage_conf.h @@ -121,6 +121,7 @@ enum virStoragePoolType { VIR_STORAGE_POOL_SCSI, /* SCSI HBA */ VIR_STORAGE_POOL_MPATH, /* Multipath devices */ VIR_STORAGE_POOL_RBD, /* RADOS Block Device */ + VIR_STORAGE_POOL_SHEEPDOG, /* Sheepdog device */ VIR_STORAGE_POOL_LAST, }; diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index e2e9b51..93964d6 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -80,6 +80,9 @@ #if WITH_STORAGE_RBD # include "storage_backend_rbd.h" #endif +#if WITH_STORAGE_SHEEPDOG +# include "storage_backend_sheepdog.h" +#endif #define VIR_FROM_THIS VIR_FROM_STORAGE @@ -109,6 +112,9 @@ static virStorageBackendPtr backends[] = { #if WITH_STORAGE_RBD &virStorageBackendRBD, #endif +#if WITH_STORAGE_SHEEPDOG + &virStorageBackendSheepdog, +#endif NULL }; diff --git a/src/storage/storage_backend_sheepdog.c b/src/storage/storage_backend_sheepdog.c new file mode 100644 index 0000000..b003856 --- /dev/null +++ b/src/storage/storage_backend_sheepdog.c @@ -0,0 +1,323 @@ +/* + * storage_backend_sheepdog.c: storage backend for Sheepdog handling + * + * Copyright (C) 2012 Wido den Hollander + * Copyright (C) 2012 Frank Spijkerman + * Copyright (C) 2012 Sebastian Wiedenroth + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Author: Wido den Hollander <wido(a)widodh.nl> + * Frank Spijkerman <frank.spijkerman(a)avira.com> + * Sebastian Wiedenroth <sebastian.wiedenroth(a)skylime.net> + */ + +#include <config.h> + +#include "virterror_internal.h" +#include "storage_backend_sheepdog.h" +#include "storage_conf.h" +#include "util/command.h" +#include "util.h" +#include "memory.h" +#include "logging.h" + +#define VIR_FROM_THIS VIR_FROM_STORAGE + +static int virStorageBackendSheepdogRefreshVol(virConnectPtr conn, + virStoragePoolObjPtr pool, + virStorageVolDefPtr vol); + +void virStorageBackendSheepdogAddHostArg(virCommandPtr cmd, + virStoragePoolObjPtr pool); + +int +virStorageBackendSheepdogParseNodeInfo(virStoragePoolDefPtr pool, + char *output) +{ + /* example output: + * 0 15245667872 117571104 0% + * Total 15245667872 117571104 0% 20972341 + */ + + const char *p, *next; + + pool->allocation = pool->capacity = pool->available = 0; + + p = output; + do { + char *end; + + if ((next = strchr(p, '\n'))) + ++next; + else + return -1; + + if (!STRPREFIX(p, "Total ")) + continue; + + p = p + 6; + + if (virStrToLong_ull(p, &end, 10, &pool->capacity) < 0) + return -1; + + if ((p = end + 1) > next) + return -1; + + if (virStrToLong_ull(p, &end, 10, &pool->allocation) < 0) + return -1; + + pool->available = pool->capacity - pool->allocation; + return 0; + + } while ((p = next)); + + return -1; +} + +void virStorageBackendSheepdogAddHostArg(virCommandPtr cmd, virStoragePoolObjPtr pool) +{ + const char *address = "localhost"; + int port = 7000; + if (pool->def->source.nhost > 0) { + if (pool->def->source.hosts[0].name != NULL) { + address = pool->def->source.hosts[0].name; + } + if (pool->def->source.hosts[0].port) { + port = pool->def->source.hosts[0].port; + } + } + virCommandAddArg(cmd, "-a"); + virCommandAddArgFormat(cmd, "%s", address); + virCommandAddArg(cmd, "-p"); + virCommandAddArgFormat(cmd, "%d", port); +} + + +static int +virStorageBackendSheepdogRefreshPool(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool) +{ + int ret; + + char *output = NULL; + + virCommandPtr cmd; + + cmd = virCommandNew(COLLIE); + virCommandAddArgList(cmd, "node", "info", "-r", NULL); + virStorageBackendSheepdogAddHostArg(cmd, pool); + virCommandSetOutputBuffer(cmd, &output); + ret = virCommandRun(cmd, NULL); + if (ret == 0) + ret = virStorageBackendSheepdogParseNodeInfo(pool->def, output); + + virCommandFree(cmd); + VIR_FREE(output); + return ret; +} + + +static int +virStorageBackendSheepdogDeleteVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool + ATTRIBUTE_UNUSED, + virStorageVolDefPtr vol, + unsigned int flags) +{ + + if (flags & VIR_STORAGE_VOL_DELETE_ZEROED) { + VIR_WARN("%s", _("This storage backend does not supported zeroed removal of volumes")); + } + + virCommandPtr cmd = virCommandNew(COLLIE); + + virCommandAddArgList(cmd, "vdi", "delete", vol->name, NULL); + virStorageBackendSheepdogAddHostArg(cmd, pool); + int ret = virCommandRun(cmd, NULL); + + virCommandFree(cmd); + return ret; +} + + +static int +virStorageBackendSheepdogCreateVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool + ATTRIBUTE_UNUSED, + virStorageVolDefPtr vol) +{ + + int ret; + + if (vol->target.encryption != NULL) { + virStorageReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("storage pool does not support encrypted volumes")); + return -1; + } + + virCommandPtr cmd = virCommandNew(COLLIE); + + virCommandAddArgList(cmd, "vdi", "create", vol->name, NULL); + virCommandAddArgFormat(cmd, "%llu", vol->capacity); + virStorageBackendSheepdogAddHostArg(cmd, pool); + ret = virCommandRun(cmd, NULL); + + virStorageBackendSheepdogRefreshVol(conn, pool, vol); + + virCommandFree(cmd); + return ret; +} + + +int +virStorageBackendSheepdogParseVdiList(virStorageVolDefPtr vol, + char *output) +{ + /* example output: + * s test 1 10 0 0 1336556634 7c2b25 + * s test 2 10 0 0 1336557203 7c2b26 + * = test 3 10 0 0 1336557216 7c2b27 + */ + + int id; + + const char *p, *next; + + vol->allocation = vol->capacity = 0; + + p = output; + do { + char *end; + + if ((next = strchr(p, '\n'))) + ++next; + + /* ignore snapshots */ + if (*p != '=') + continue; + + /* skip space */ + if (p + 2 < next) { + p += 2; + } else { + return -1; + } + + /* skip name */ + while (*p != '\0' + && (*p != ' ' || (*p == ' ' && (*(p - 1) == '\\')))) + ++p; + + if (virStrToLong_i(p, &end, 10, &id) < 0) + return -1; + + p = end + 1; + + if (virStrToLong_ull(p, &end, 10, &vol->capacity) < 0) + return -1; + + p = end + 1; + + if (virStrToLong_ull(p, &end, 10, &vol->allocation) < 0) + return -1; + + return 0; + } while ((p = next)); + + return -1; +} + +static int +virStorageBackendSheepdogRefreshVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool + ATTRIBUTE_UNUSED, + virStorageVolDefPtr vol) +{ + int ret; + + char *output = NULL; + + virCommandPtr cmd; + + cmd = virCommandNew(COLLIE); + virCommandAddArgList(cmd, "vdi", "list", vol->name, "-r", NULL); + virStorageBackendSheepdogAddHostArg(cmd, pool); + virCommandSetOutputBuffer(cmd, &output); + ret = virCommandRun(cmd, NULL); + + if (ret < 0) + goto cleanup; + + if ((ret = virStorageBackendSheepdogParseVdiList(vol, output)) < 0) + goto cleanup; + + vol->type = VIR_STORAGE_VOL_NETWORK; + + VIR_FREE(vol->key); + if (virAsprintf(&vol->key, "%s/%s", + pool->def->source.name, vol->name) == -1) { + virReportOOMError(); + goto cleanup; + } + + VIR_FREE(vol->target.path); + if (virAsprintf(&vol->target.path, "sheepdog:%s", vol->name) == -1) { + virReportOOMError(); + goto cleanup; + } + + + +cleanup: + virCommandFree(cmd); + return ret; +} + + +static int +virStorageBackendSheepdogResizeVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool + ATTRIBUTE_UNUSED, + virStorageVolDefPtr vol, + unsigned long long capacity, + unsigned int flags) +{ + + virCheckFlags(0, -1); + + virCommandPtr cmd = virCommandNew(COLLIE); + + virCommandAddArgList(cmd, "vdi", "resize", vol->name, NULL); + virCommandAddArgFormat(cmd, "%llu", capacity); + virStorageBackendSheepdogAddHostArg(cmd, pool); + int ret = virCommandRun(cmd, NULL); + + virCommandFree(cmd); + return ret; + +} + + + +virStorageBackend virStorageBackendSheepdog = { + .type = VIR_STORAGE_POOL_SHEEPDOG, + + .refreshPool = virStorageBackendSheepdogRefreshPool, + .createVol = virStorageBackendSheepdogCreateVol, + .refreshVol = virStorageBackendSheepdogRefreshVol, + .deleteVol = virStorageBackendSheepdogDeleteVol, + .resizeVol = virStorageBackendSheepdogResizeVol, +}; diff --git a/src/storage/storage_backend_sheepdog.h b/src/storage/storage_backend_sheepdog.h new file mode 100644 index 0000000..51ef7a4 --- /dev/null +++ b/src/storage/storage_backend_sheepdog.h @@ -0,0 +1,39 @@ +/* + * storage_backend_sheepog.h: storage backend for Sheepdog handling + * + * Copyright (C) 2012 Wido den Hollander + * Copyright (C) 2012 Frank Spijkerman + * Copyright (C) 2012 Sebastian Wiedenroth + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Author: Wido den Hollander <wido(a)widodh.nl> + * Frank Spijkerman <frank.spijkerman(a)avira.com> + * Sebastian Wiedenroth <sebastian.wiedenroth(a)skylime.net> + */ + +#ifndef __VIR_STORAGE_BACKEND_SHEEPDOG_H__ +# define __VIR_STORAGE_BACKEND_SHEEPDOG_H__ + +# include "storage_backend.h" + +int virStorageBackendSheepdogParseNodeInfo(virStoragePoolDefPtr pool, + char *output); +int virStorageBackendSheepdogParseVdiList(virStorageVolDefPtr vol, + char *output); + +extern virStorageBackend virStorageBackendSheepdog; + +#endif /* __VIR_STORAGE_BACKEND_SHEEPDOG_H__ */ diff --git a/tests/Makefile.am b/tests/Makefile.am index e3bd6d1..a466480 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -141,6 +141,10 @@ if WITH_NETWORK test_programs += networkxml2argvtest endif +if WITH_STORAGE_SHEEPDOG +test_programs += storagebackendsheepdogtest +endif + test_programs += nwfilterxml2xmltest test_programs += storagevolxml2xmltest storagepoolxml2xmltest @@ -398,6 +402,15 @@ else EXTRA_DIST += networkxml2argvtest.c endif +if WITH_STORAGE_SHEEPDOG +storagebackendsheepdogtest_SOURCES = \ + storagebackendsheepdogtest.c \ + testutils.c testutils.h +storagebackendsheepdogtest_LDADD = ../src/libvirt_driver_storage.la $(LDADDS) +else +EXTRA_DIST += storagebackendsheepdogtest.c +endif + nwfilterxml2xmltest_SOURCES = \ nwfilterxml2xmltest.c \ testutils.c testutils.h diff --git a/tests/storagebackendsheepdogtest.c b/tests/storagebackendsheepdogtest.c new file mode 100644 index 0000000..a8f62a4 --- /dev/null +++ b/tests/storagebackendsheepdogtest.c @@ -0,0 +1,196 @@ +#include <config.h> + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <stdint.h> +#include <string.h> + +#include <sys/types.h> +#include <fcntl.h> + +#include "internal.h" +#include "testutils.h" +#include "storage/storage_backend_sheepdog.h" + + +typedef struct { + const char *output; + int expected_return; + uint64_t expected_capacity; + uint64_t expected_allocation; +} collie_test; + + +static int +test_node_info_parser(collie_test test, char *poolxml) +{ + int ret = -1; + char *output = NULL; + char *poolXmlData = NULL; + virStoragePoolDefPtr pool = NULL; + + if (virtTestLoadFile(poolxml, &poolXmlData) < 0) + goto cleanup; + + if (!(pool = virStoragePoolDefParseString(poolXmlData))) + goto cleanup; + + output = strdup(test.output); + if(!output) { + goto cleanup; + } + + ret = virStorageBackendSheepdogParseNodeInfo(pool, output); + + if (ret != test.expected_return) + goto cleanup; + + if (ret != 0) { + ret = 0; + goto cleanup; + } + + if (pool->capacity == test.expected_capacity + && pool->allocation == test.expected_allocation) + ret = 0; + + cleanup: + VIR_FREE(output); + VIR_FREE(poolXmlData); + virStoragePoolDefFree(pool); + return ret; +} + +static int +test_vdi_list_parser(collie_test test, char *poolxml, char *volxml) +{ + int ret = -1; + char *poolXmlData = NULL; + char *volXmlData = NULL; + char *output = NULL; + virStoragePoolDefPtr pool = NULL; + virStorageVolDefPtr vol = NULL; + + if (virtTestLoadFile(poolxml, &poolXmlData) < 0) + goto cleanup; + if (virtTestLoadFile(volxml, &volXmlData) < 0) + goto cleanup; + + if (!(pool = virStoragePoolDefParseString(poolXmlData))) + goto cleanup; + + if (!(vol = virStorageVolDefParseString(pool, volXmlData))) + goto cleanup; + + output = strdup(test.output); + if(!output) { + goto cleanup; + } + + ret = virStorageBackendSheepdogParseVdiList(vol, output); + + + if (ret != test.expected_return) + goto cleanup; + + if (ret != 0) { + ret = 0; + goto cleanup; + } + + if (vol->capacity == test.expected_capacity + && vol->allocation == test.expected_allocation) + ret = 0; + + cleanup: + VIR_FREE(output); + VIR_FREE(poolXmlData); + VIR_FREE(volXmlData); + virStoragePoolDefFree(pool); + virStorageVolDefFree(vol); + return ret; +} + + +static int +mymain(void) +{ + int ret = -1; + char *poolxml = NULL; + char *volxml = NULL; + + collie_test node_info_tests[] = { + {"", -1, 0, 0}, + {"Total 15245667872 117571104 0% 20972341\n", 0, 15245667872, 117571104}, + {"To", -1, 0, 0}, + {"asdf\nasdf", -1, 0, 0}, + {"Total ", -1, 0, 0}, + {"Total 1", -1, 0, 0}, + {"Total 1\n", -1, 0, 0}, + {"Total 1 ", -1, 0, 0}, + {"Total 1 2", -1, 0, 0}, + {"Total 1 2 ", -1, 0, 0}, + {"Total 1 2\n", 0, 1, 2}, + {"Total 1 2 \n", 0, 1, 2}, + {"Total a 2 \n", -1, 0, 0}, + {"Total 1 b \n", -1, 0, 0}, + {"Total a b \n", -1, 0, 0}, + {"stuff\nTotal 1 2 \n", 0, 1, 2}, + {"0 1 2 3\nTotal 1 2 \n", 0, 1, 2}, + {NULL, 0, 0, 0} + }; + + collie_test vdi_list_tests[] = { + {"", -1, 0, 0}, + {"= test 3 10 20 0 1336557216 7c2b27\n", 0, 10, 20}, + {"= test\\ with\\ spaces 3 10 20 0 1336557216 7c2b27\n", 0, 10, 20}, + {"s test 1 10 20 0 1336556634 7c2b25\n= test 3 50 60 0 1336557216 7c2b27\n", 0, 50, 60}, + {"=", -1, 0, 0}, + {"= test", -1, 0, 0}, + {"= test ", -1, 0, 0}, + {"= test 1", -1, 0, 0}, + {"= test 1 ", -1, 0, 0}, + {"= test 1 2", -1, 0, 0}, + {"= test 1 2 ", -1, 0, 0}, + {"= test 1 2 3", -1, 0, 0}, + {NULL, 0, 0, 0} + }; + + collie_test *test = node_info_tests; + + if (virAsprintf(&poolxml, "%s/storagepoolxml2xmlin/pool-sheepdog.xml", + abs_srcdir) < 0) + goto cleanup; + + if (virAsprintf(&volxml, "%s/storagevolxml2xmlin/vol-sheepdog.xml", + abs_srcdir) < 0) + goto cleanup; + + while (test->output != NULL) { + ret = test_node_info_parser(*test, poolxml); + virtTestResult("node_info_parser", ret, NULL); + ++test; + if (ret < 0) + return EXIT_FAILURE; + } + + test = vdi_list_tests; + + while (test->output != NULL) { + ret = test_vdi_list_parser(*test, poolxml, volxml); + virtTestResult("vdi_list_parser", ret, NULL); + ++test; + if (ret < 0) + return EXIT_FAILURE; + } + + ret = 0; + + cleanup: + VIR_FREE(poolxml); + VIR_FREE(volxml); + return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE; +} + +VIRT_TEST_MAIN(mymain) diff --git a/tests/storagepoolxml2xmlin/pool-sheepdog.xml b/tests/storagepoolxml2xmlin/pool-sheepdog.xml new file mode 100644 index 0000000..1287047 --- /dev/null +++ b/tests/storagepoolxml2xmlin/pool-sheepdog.xml @@ -0,0 +1,8 @@ +<pool type='sheepdog'> + <name>sheepdog</name> + <uuid>65fcba04-5b13-bd93-cff3-52ce48e11ad7</uuid> + <source> + <host name='localhost' port='7000'/> + <name>sheepdog</name> + </source> +</pool> diff --git a/tests/storagepoolxml2xmlout/pool-sheepdog.xml b/tests/storagepoolxml2xmlout/pool-sheepdog.xml new file mode 100644 index 0000000..000c068 --- /dev/null +++ b/tests/storagepoolxml2xmlout/pool-sheepdog.xml @@ -0,0 +1,11 @@ +<pool type='sheepdog'> + <name>sheepdog</name> + <uuid>65fcba04-5b13-bd93-cff3-52ce48e11ad7</uuid> + <capacity unit='bytes'>0</capacity> + <allocation unit='bytes'>0</allocation> + <available unit='bytes'>0</available> + <source> + <host name='localhost' port='7000'/> + <name>sheepdog</name> + </source> +</pool> diff --git a/tests/storagepoolxml2xmltest.c b/tests/storagepoolxml2xmltest.c index d73fc8a..8cac978 100644 --- a/tests/storagepoolxml2xmltest.c +++ b/tests/storagepoolxml2xmltest.c @@ -93,6 +93,7 @@ mymain(void) DO_TEST("pool-mpath"); DO_TEST("pool-iscsi-multiiqn"); DO_TEST("pool-iscsi-vendor-product"); + DO_TEST("pool-sheepdog"); return ret==0 ? EXIT_SUCCESS : EXIT_FAILURE; } diff --git a/tests/storagevolxml2xmlin/vol-sheepdog.xml b/tests/storagevolxml2xmlin/vol-sheepdog.xml new file mode 100644 index 0000000..49e221c --- /dev/null +++ b/tests/storagevolxml2xmlin/vol-sheepdog.xml @@ -0,0 +1,10 @@ +<volume> + <name>test2</name> + <source> + </source> + <capacity unit='bytes'>1024</capacity> + <allocation unit='bytes'>0</allocation> + <target> + <path>sheepdog:test2</path> + </target> +</volume> diff --git a/tests/storagevolxml2xmlout/vol-sheepdog.xml b/tests/storagevolxml2xmlout/vol-sheepdog.xml new file mode 100644 index 0000000..2f19af8 --- /dev/null +++ b/tests/storagevolxml2xmlout/vol-sheepdog.xml @@ -0,0 +1,17 @@ +<volume> + <name>test2</name> + <key>(null)</key> + <source> + </source> + <capacity unit='bytes'>1024</capacity> + <allocation unit='bytes'>0</allocation> + <target> + <path>sheepdog:test2</path> + <format type='unknown'/> + <permissions> + <mode>0600</mode> + <owner>4294967295</owner> + <group>4294967295</group> + </permissions> + </target> +</volume> diff --git a/tests/storagevolxml2xmltest.c b/tests/storagevolxml2xmltest.c index 37c92cd..ee85988 100644 --- a/tests/storagevolxml2xmltest.c +++ b/tests/storagevolxml2xmltest.c @@ -112,6 +112,7 @@ mymain(void) DO_TEST("pool-disk", "vol-partition"); DO_TEST("pool-logical", "vol-logical"); DO_TEST("pool-logical", "vol-logical-backing"); + DO_TEST("pool-sheepdog", "vol-sheepdog"); return ret==0 ? EXIT_SUCCESS : EXIT_FAILURE; } diff --git a/tools/virsh.c b/tools/virsh.c index 744b629..3f24a34 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -20229,6 +20229,9 @@ vshShowVersion(vshControl *ctl ATTRIBUTE_UNUSED) #ifdef WITH_STORAGE_RBD vshPrint(ctl, " RBD"); #endif +#ifdef WITH_STORAGE_SHEEPDOG + vshPrint(ctl, " Sheepdog"); +#endif vshPrint(ctl, "\n"); vshPrint(ctl, "%s", _(" Miscellaneous:")); -- 1.7.9.4

2 2

[libvirt] [libvirt-glib] build: allow building with newer glibc-headers and -O0
by Christophe Fergeau 13 Jun '12

13 Jun '12

Commit log and autoconf snippet written by Eric Blake for libvirt. glibc 2.15 (on Fedora 17) coupled with explicit disabling of optimization during development dies a painful death: In file included from /usr/include/limits.h:27:0, from /usr/lib/gcc/x86_64-redhat-linux/4.7.0/include/limits.h:169,. from /usr/lib/gcc/x86_64-redhat-linux/4.7.0/include/syslimits.h:7, from /usr/lib/gcc/x86_64-redhat-linux/4.7.0/include/limits.h:34, from util/bitmap.c:26: /usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires compiling with optimization (-O) [-Werror=cpp] cc1: all warnings being treated as errors Work around this by only conditionally defining _FORTIFY_SOURCE, in the case where glibc can actually use it. The trick is using AH_VERBATIM instead of AC_DEFINE. * m4/virt-compile-warnings.m4 (LIBOSINFO_COMPILE_WARNINGS): Squelch _FORTIFY_SOURCE when needed to avoid glibc #warnings. --- m4/virt-compile-warnings.m4 | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/m4/virt-compile-warnings.m4 b/m4/virt-compile-warnings.m4 index de542cc..99d214e 100644 --- a/m4/virt-compile-warnings.m4 +++ b/m4/virt-compile-warnings.m4 @@ -84,8 +84,13 @@ AC_DEFUN([LIBVIRT_GLIB_COMPILE_WARNINGS],[ gl_WARN_ADD([-Wframe-larger-than=4096]) # Use improved glibc headers - AC_DEFINE([_FORTIFY_SOURCE], [2], - [enable compile-time and run-time bounds-checking, and some warnings]) + AH_VERBATIM([FORTIFY_SOURCE], + [/* Enable compile-time and run-time bounds-checking, and some warnings, + without upsetting newer glibc. */ + #if defined __OPTIMIZE__ && __OPTIMIZE__ + # define _FORTIFY_SOURCE 2 + #endif + ]) # Extra special flags dnl -fstack-protector stuff passes gl_WARN_ADD with gcc -- 1.7.10.2

2 1

[libvirt] [sandbox]console-rpc: fix segfault on null stdin
by Radu Caragea 13 Jun '12

13 Jun '12

If gvir_sandbox_console_attach is called with NULL as the stdin parameter it results in a segfault in the console-rpc module from libvirt-sandbox. Calling with NULL is of course useful when running something noninteractive and you don't want it to grab the stdin or pass any fd whatsoever.

2 1

Re: [libvirt] [Qemu-devel] [PATCH v2 0/4] file descriptor passing using passfd
by Kevin Wolf 13 Jun '12

13 Jun '12

Am 08.06.2012 17:42, schrieb Corey Bryant: > libvirt's sVirt security driver provides SELinux MAC isolation for > Qemu guest processes and their corresponding image files. In other > words, sVirt uses SELinux to prevent a QEMU process from opening > files that do not belong to it. > > sVirt provides this support by labeling guests and resources with > security labels that are stored in file system extended attributes. > Some file systems, such as NFS, do not support the extended > attribute security namespace, and therefore cannot support sVirt > isolation. > > A solution to this problem is to provide fd passing support, where > libvirt opens files and passes file descriptors to QEMU. This, > along with SELinux policy to prevent QEMU from opening files, can > provide image file isolation for NFS files stored on the same NFS > mount. > > This patch series adds the passfd QMP monitor command, which allows > an fd to be passed via SCM_RIGHTS, and returns the received file > descriptor. Support is also added to the block layer to allow QEMU > to dup the fd when the filename is of the /dev/fd/X format. This > is useful if MAC policy prevents QEMU from opening specific types > of files. > > One nice thing about this approach is that no new SELinux policy is > required to prevent open of NFS files (files with type nfs_t). The > virt_use_nfs boolean type simply needs to be set to false, and open > will be prevented (and dup will be allowed). For example: > > # setsebool virt_use_nfs 0 > # getsebool virt_use_nfs > virt_use_nfs --> off > > Corey Bryant (4): > qapi: Convert getfd and closefd > qapi: Add passfd QMP command > osdep: Enable qemu_open to dup pre-opened fd > block: Convert open calls to qemu_open > > block/raw-posix.c | 18 +++++++++--------- > block/raw-win32.c | 4 ++-- > block/vdi.c | 5 +++-- > block/vmdk.c | 21 +++++++++------------ > block/vpc.c | 2 +- > block/vvfat.c | 21 +++++++++++---------- > hmp-commands.hx | 6 ++---- > hmp.c | 18 ++++++++++++++++++ > hmp.h | 2 ++ > monitor.c | 36 ++++++++++++++++++++---------------- > osdep.c | 13 +++++++++++++ > qapi-schema.json | 44 ++++++++++++++++++++++++++++++++++++++++++++ > qmp-commands.hx | 33 +++++++++++++++++++++++++++++---- > 13 files changed, 163 insertions(+), 60 deletions(-) Looks good to me. If Luiz is okay with the QMP part, I'm going to apply this to the block branch. Corey, please make sure to check the host_floppy problem and send a patch if necessary. Kevin

2 1

Re: [libvirt] [Qemu-devel] [PATCH v2 4/4] block: Convert open calls to qemu_open
by Kevin Wolf 13 Jun '12

13 Jun '12

Am 08.06.2012 17:42, schrieb Corey Bryant: > This patch converts all block layer open calls to qemu_open. This > enables all block layer open paths to dup(X) a pre-opened file > descriptor if the filename is of the format /dev/fd/X. This is > useful if QEMU is restricted from opening certain files. > > Note that this adds the O_CLOEXEC flag to the changed open paths > when the O_CLOEXEC macro is defined. > > v2: > -Convert calls to qemu_open instead of file_open (kwolf(a)redhat.com) > -Mention introduction of O_CLOEXEC (kwolf(a)redhat.com) > > Signed-off-by: Corey Bryant <coreyb(a)linux.vnet.ibm.com> > @@ -950,7 +950,7 @@ static int floppy_probe_device(const char *filename) > if (strstart(filename, "/dev/fd", NULL)) > prio = 50; Good to have this context here. I think this has to be removed in another patch or all our file descriptors will become host_floppy... > > - fd = open(filename, O_RDONLY | O_NONBLOCK); > + fd = qemu_open(filename, O_RDONLY | O_NONBLOCK); > if (fd < 0) { > goto out; > } Kevin

2 1

[libvirt] [PATCH 00/12] rpc: Fix several issues with keepalive messages
by Jiri Denemark 13 Jun '12

13 Jun '12

So far, we were dropping non-blocking calls whenever sending them would block. In case a client is sending lots of stream calls (which are not supposed to generate any reply), the assumption that having other calls in a queue is sufficient to get a reply from the server doesn't work. I tried to fix this in b1e374a7ac56927cfe62435179bf0bba1e08b372 but failed and reverted that commit. While working on the proper fix, I discovered several other issues we had in handling keepalive messages in client RPC code. See individual patches for more details. As a nice bonus, the fixed version is shorter by one line than the current broken version :-) Jiri Denemark (12): client rpc: Improve debug messages in virNetClientIO client rpc: Use event loop for writing client rpc: Don't drop non-blocking calls client rpc: Just queue non-blocking call if another thread has the buck client rpc: Drop unused return value of virNetClientSendNonBlock rpc: Refactor keepalive timer code rpc: Add APIs for direct triggering of keepalive timer client rpc: Separate call creation from running IO loop rpc: Do not use timer for sending keepalive responses rpc: Remove unused parameter in virKeepAliveStopInternal server rpc: Remove APIs for manipulating filters on locked client client rpc: Send keepalive requests from IO event loop src/libvirt_probes.d | 2 +- src/rpc/virkeepalive.c | 233 ++++++++++++-------------- src/rpc/virkeepalive.h | 7 +- src/rpc/virnetclient.c | 368 +++++++++++++++++++++++------------------- src/rpc/virnetserverclient.c | 127 +++++++-------- 5 files changed, 368 insertions(+), 369 deletions(-) -- 1.7.10.2

3 31

[libvirt] [PATCH] build: fix 'make dist' on virgin checkout
by Eric Blake 13 Jun '12

13 Jun '12

'make dist' was depending on *protocol-structs files, which are stored in git but in turn depended on generated files. We still want to ship the protocol-structs files, but by renaming the tests to something not matching a file name, we separate 'make check' (which depends on the generated file) from 'make dist' (which only depends on the git files). After all, the tarball should never depend on a generated file not stored in git. I found one more case of a git file depending on a generated file, in a bogus virkeycode.c listing; but at least this one had no associated rules so it never broke 'make dist'. Reported by Wen Congyang. Latent bug has been present since commit 62dee6f, but only recently exposed by commit 7bff56a. * src/Makefile.am ($(srcdir)/util/virkeycode.c): Drop useless dependency. (BUILT_SOURCES): ...and build virkeymaps.h sooner. (PROTOCOL_STRUCTS): Rather than depend on the struct file... (check-local): ...convert things into a phony target of... (check-protocol): ...a new check. ($(srcdir)/remote_protocol-struct): Rename to isolate the distributed file from the conditional test. (PDWTAGS): Deal with rename. Swap to compare 'expected actual'. --- Posting now for a review. I'm still planning on running a 'make distcheck' on a virgin repository, if that passes and correctly runs the protocol syntax checks, then I will check this in under the build-breaker rule. Or, if someone ACKs this first... src/Makefile.am | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/src/Makefile.am b/src/Makefile.am index 60f5442..2bcebcf 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -113,12 +113,12 @@ UTIL_SOURCES = \ EXTRA_DIST += $(srcdir)/util/virkeymaps.h $(srcdir)/util/keymaps.csv \ $(srcdir)/util/virkeycode-mapgen.py +BUILT_SOURCES += $(srcdir)/util/virkeymaps.h + $(srcdir)/util/virkeymaps.h: $(srcdir)/util/keymaps.csv \ $(srcdir)/util/virkeycode-mapgen.py $(AM_V_GEN)$(PYTHON) $(srcdir)/util/virkeycode-mapgen.py <$(srcdir)/util/keymaps.csv >$@ -$(srcdir)/util/virkeycode.c: $(srcdir)/util/virkeycode.h $(srcdir)/util/virkeymaps.h - EXTRA_DIST += util/threads-pthread.c util/threads-win32.c # Internal generic driver infrastructure @@ -293,7 +293,7 @@ PDWTAGS = \ -e '}' \ < $(@F)-t1 > $(@F)-t3; \ case $$? in 8) rm -f $(@F)-t?; exit 0;; 0) ;; *) exit 1;; esac;\ - diff -u $(@F)-t3 $@; st=$$?; rm -f $(@F)-t?; exit $$st; \ + diff -u $(@)s $(@F)-t3; st=$$?; rm -f $(@F)-t?; exit $$st; \ fi; \ else \ echo 'WARNING: you lack pdwtags; skipping the $@ test' >&2; \ @@ -306,21 +306,24 @@ PROTOCOL_STRUCTS = \ $(srcdir)/virnetprotocol-structs \ $(srcdir)/virkeepaliveprotocol-structs if WITH_REMOTE +check-protocol: $(PROTOCOL_STRUCTS) $(PROTOCOL_STRUCTS:structs=struct) + # The .o file that pdwtags parses is created as a side effect of running # libtool; but from make's perspective we depend on the .lo file. -$(srcdir)/%_protocol-structs: libvirt_driver_remote_la-%_protocol.lo - $(PDWTAGS) -$(srcdir)/virnetprotocol-structs: libvirt_net_rpc_la-virnetprotocol.lo +$(srcdir)/remote_protocol-struct $(srcdir)/qemu_protocol-struct: \ + $(srcdir)/%-struct: libvirt_driver_remote_la-%.lo $(PDWTAGS) -$(srcdir)/virkeepaliveprotocol-structs: libvirt_net_rpc_la-virkeepaliveprotocol.lo +$(srcdir)/virnetprotocol-struct $(srcdir)/virkeepaliveprotocol-struct: \ + $(srcdir)/%-struct: libvirt_net_rpc_la-%.lo $(PDWTAGS) else !WITH_REMOTE -# These generated files must live in git, because they cannot be re-generated -# when configured --without-remote. -$(PROTOCOL_STRUCTS): +# The $(PROTOCOL_STRUCTS) files must live in git, because they cannot be +# re-generated when configured --without-remote. +check-protocol: endif EXTRA_DIST += $(PROTOCOL_STRUCTS) -check-local: $(PROTOCOL_STRUCTS) +check-local: check-protocol +.PHONY: check-protocol $(PROTOCOL_STRUCTS:structs=struct) # Mock driver, covering domains, storage, networks, etc TEST_DRIVER_SOURCES = \ -- 1.7.10.2

1 1