June 2011 - Devel - libvirt List Archives

[libvirt] Request to rename 'destroy' to something milder.
by Kashyap Chamarthy 14 Jun '11

14 Jun '11

(please cc me in response as I have not subscribed to this list) Hi all, A minor nitpick: Every-time I suggest someone to do a force shut-down a guest using 'virsh destroy foo' , the very first question I get is -- does it _destroy_ my data? This causes confusion to the inexperienced user and makes him/her suspect that the data/disk could be destroyed while running 'virsh destroy foo' Maybe replacing it to a milder name like 'poweroff' or something might help? Thanks, Kashyap Chamarthy

4 7

Re: [libvirt] [libvirt-users] request to add an application the app listing on libvirt site
by Eric Blake 14 Jun '11

14 Jun '11

[adding libvir-list for patch tracking] On 06/14/2011 10:17 AM, Ohad Levy wrote: >> the simplest is to actually send a patch against the file docs/ >> apps.html.in >> in the source tree. I would think "Ruby Objects for libvirt" should be >> added in the "libraries" section, and "Foreman" in the "provisioning" >> section. The pages are maintained in the source which also allow to >> package them easilly, I hope making an HTML patch isn't too complex, >> if not say so I will do it myself :-) >> >> thanks ! >> >> thanks! see attached patch. > Thanks for the patch. Your mailer sent both plain text and html renderings of the email body; that is overkill (plain text alone works just fine). > + <dl> > + <dt><a href="http://theforeman.org">Foreman</a></dt> > + <dd> > + Foreman is an open source web based application aimed to be a Single Address For All Machines Life Cycle Management, Foreman: > + That's a long line. I reformatted to break things to 80 columns where reasonably possible, and added your name to AUTHORS. Let me know if you prefer an alternate spelling there (for example, the commit was attributed to your gmail.com address, although I see that you also have a redhat.com address). Also, I saw some grammar nits. ACK and pushed as modified: diff --git i/docs/apps.html.in w/docs/apps.html.in index 69c9aaa..6b5afdb 100644 --- i/docs/apps.html.in +++ w/docs/apps.html.in @@ -15,7 +15,7 @@ To add an application not listed on this page, send a message to the <a href="contact.html">mailing list</a>, requesting it be added here, or simply send a patch against the documentation - in our git docs sub directory. + in the libvirt.git docs subdirectory. If your application uses libvirt as its API, the following graphic is available for your website to advertise support for libvirt: @@ -232,11 +232,11 @@ <dt><a href="https://github.com/ohadlevy/virt#readme">Ruby Libvirt Object bindings</a></dt> <dd> - Virt Allows using simple ruby objects to manipulate + Allows using simple ruby objects to manipulate hypervisors, guests, storage, network etc. It is based on top of the <a href="http://libvirt.org/ruby">native ruby - binding</a>. + bindings</a>. </dd> </dl> <h2><a name="livecd">LiveCD / Appliances</a></h2> -- Eric Blake eblake(a)redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org

1 0

[libvirt] [PATCH 0/3] more virCommand cleanups
by Eric Blake 14 Jun '11

14 Jun '11

When I first reviewed Cole's virCommand cleanups, I pointed out some possible later cleanups. Here they are. Eric Blake (3): command: reduce duplicated debug messages command: avoid double close storage: avoid an intermediate malloc src/storage/storage_backend.c | 20 ++++++------ src/util/command.c | 64 ++++++++++++++++++++-------------------- 2 files changed, 42 insertions(+), 42 deletions(-) -- 1.7.4.4

2 7

[libvirt] [PATCH] docs: improve VIR_DOMAIN_AFFECT_CURRENT description
by Eric Blake 14 Jun '11

14 Jun '11

* include/libvirt/libvirt.h.in (virDomainModificationImpact): Reword. --- include/libvirt/libvirt.h.in | 15 ++++++++++----- 1 files changed, 10 insertions(+), 5 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 76ad908..a7c5f72 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -144,11 +144,16 @@ typedef enum { /** * virDomainModificationImpact: * - * Several APIs take flags to determine whether a change to the domain - * affects just the running instance, just the persistent definition, - * or both. The use of VIR_DOMAIN_AFFECT_CURRENT will resolve to - * either VIR_DOMAIN_AFFECT_LIVE or VIR_DOMAIN_AFFECT_CONFIG according - * to current domain state. VIR_DOMAIN_AFFECT_LIVE requires a running + * Several modification APIs take flags to determine whether a change + * to the domain affects just the running instance, just the + * persistent definition, or both at the same time. The counterpart + * query APIs also take the same flags to determine whether to query + * the running instance or persistent definition, although both cannot + * be queried at once. + * + * The use of VIR_DOMAIN_AFFECT_CURRENT will resolve to either + * VIR_DOMAIN_AFFECT_LIVE or VIR_DOMAIN_AFFECT_CONFIG according to + * current domain state. VIR_DOMAIN_AFFECT_LIVE requires a running * domain, and VIR_DOMAIN_AFFECT_CONFIG requires a persistent domain * (whether or not it is running). */ -- 1.7.4.4

2 2

[libvirt] [PATCH] Fix errno return in safezero()
by Daniel P. Berrange 14 Jun '11

14 Jun '11

Most of the safezero() implementations return -1 on error, setting errno. The safezero() impl using posix_fallocate() though returned a positive errno value on error (due to the unusual API contract of posix_fallocate() compared to most syscall APIs). * src/util/util.c: Ensure safezero() returns -1 and sets errno on error. * src/storage/storage_backend.c: Change safezero != 0 to < 0 for detecting errors --- src/storage/storage_backend.c | 4 ++-- src/util/util.c | 6 +++++- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index d269e4c..63e3005 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -331,7 +331,7 @@ createRawFile(int fd, virStorageVolDefPtr vol, if (bytes > remain) bytes = remain; - if (safezero(fd, 0, vol->allocation - remain, bytes) != 0) { + if (safezero(fd, 0, vol->allocation - remain, bytes) < 0) { ret = -errno; virReportSystemError(errno, _("cannot fill file '%s'"), vol->target.path); @@ -340,7 +340,7 @@ createRawFile(int fd, virStorageVolDefPtr vol, remain -= bytes; } } else { /* No progress bars to be shown */ - if (safezero(fd, 0, 0, remain) != 0) { + if (safezero(fd, 0, 0, remain) < 0) { ret = -errno; virReportSystemError(errno, _("cannot fill file '%s'"), vol->target.path); diff --git a/src/util/util.c b/src/util/util.c index d00f065..dbfe2ee 100644 --- a/src/util/util.c +++ b/src/util/util.c @@ -136,7 +136,11 @@ safewrite(int fd, const void *buf, size_t count) #ifdef HAVE_POSIX_FALLOCATE int safezero(int fd, int flags ATTRIBUTE_UNUSED, off_t offset, off_t len) { - return posix_fallocate(fd, offset, len); + int ret = posix_fallocate(fd, offset, len); + if (ret == 0) + return 0; + errno = ret; + return -1; } #else -- 1.7.4.4

2 2

[libvirt] [PATCH] Add missing error reporting when loading mac filter config for QEMU
by Daniel P. Berrange 14 Jun '11

14 Jun '11

If the 'mac_filter' configuration parameter is enabled, and there is a failure to enable filtering, no error is reported back to the caller. Also fix some bogus whitespace indentation for hugetlbfs_mount * src/qemu/qemu_conf.c: Add missing error reporting --- src/qemu/qemu_conf.c | 24 ++++++++++++++---------- 1 files changed, 14 insertions(+), 10 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index b00b391..3d8aba4 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -378,16 +378,16 @@ int qemudLoadDriverConfig(struct qemud_driver *driver, } } - p = virConfGetValue (conf, "hugetlbfs_mount"); - CHECK_TYPE ("hugetlbfs_mount", VIR_CONF_STRING); - if (p && p->str) { - VIR_FREE(driver->hugetlbfs_mount); - if (!(driver->hugetlbfs_mount = strdup(p->str))) { - virReportOOMError(); - virConfFree(conf); - return -1; - } - } + p = virConfGetValue (conf, "hugetlbfs_mount"); + CHECK_TYPE ("hugetlbfs_mount", VIR_CONF_STRING); + if (p && p->str) { + VIR_FREE(driver->hugetlbfs_mount); + if (!(driver->hugetlbfs_mount = strdup(p->str))) { + virReportOOMError(); + virConfFree(conf); + return -1; + } + } p = virConfGetValue (conf, "mac_filter"); CHECK_TYPE ("mac_filter", VIR_CONF_LONG); @@ -398,12 +398,16 @@ int qemudLoadDriverConfig(struct qemud_driver *driver, virReportSystemError(errno, _("failed to enable mac filter in '%s'"), __FILE__); + virConfFree(conf); + return -1; } if ((errno = networkDisableAllFrames(driver))) { virReportSystemError(errno, _("failed to add rule to drop all frames in '%s'"), __FILE__); + virConfFree(conf); + return -1; } } -- 1.7.4.4

2 1

[libvirt] [PATCH] Add missing checks for NULL in domain lock manager
by Daniel P. Berrange 14 Jun '11

14 Jun '11

The domain lock manager forgot to include a bunch of checks for NULL which could occur on OOM * src/locking/domain_lock.c: Add checks for NULL --- src/locking/domain_lock.c | 49 ++++++++++++++++++++++++++++++++++---------- 1 files changed, 38 insertions(+), 11 deletions(-) diff --git a/src/locking/domain_lock.c b/src/locking/domain_lock.c index 771ed53..feb3f98 100644 --- a/src/locking/domain_lock.c +++ b/src/locking/domain_lock.c @@ -157,10 +157,13 @@ int virDomainLockProcessStart(virLockManagerPluginPtr plugin, virDomainObjPtr dom, bool paused) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, true); + virLockManagerPtr lock; int ret; int flags = VIR_LOCK_MANAGER_ACQUIRE_RESTRICT; + if (!(lock = virDomainLockManagerNew(plugin, dom, true))) + return -1; + if (paused) flags |= VIR_LOCK_MANAGER_ACQUIRE_REGISTER_ONLY; @@ -175,9 +178,13 @@ int virDomainLockProcessPause(virLockManagerPluginPtr plugin, virDomainObjPtr dom, char **state) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, true); - int ret = virLockManagerRelease(lock, state, 0); + virLockManagerPtr lock; + int ret; + + if (!(lock = virDomainLockManagerNew(plugin, dom, true))) + return -1; + ret = virLockManagerRelease(lock, state, 0); virLockManagerFree(lock); return ret; @@ -187,9 +194,13 @@ int virDomainLockProcessResume(virLockManagerPluginPtr plugin, virDomainObjPtr dom, const char *state) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, true); - int ret = virLockManagerAcquire(lock, state, 0); + virLockManagerPtr lock; + int ret; + + if (!(lock = virDomainLockManagerNew(plugin, dom, true))) + return -1; + ret = virLockManagerAcquire(lock, state, 0); virLockManagerFree(lock); return ret; @@ -199,9 +210,13 @@ int virDomainLockProcessInquire(virLockManagerPluginPtr plugin, virDomainObjPtr dom, char **state) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, true); - int ret = virLockManagerInquire(lock, state, 0); + virLockManagerPtr lock; + int ret; + if (!(lock = virDomainLockManagerNew(plugin, dom, true))) + return -1; + + ret = virLockManagerInquire(lock, state, 0); virLockManagerFree(lock); return ret; @@ -212,9 +227,12 @@ int virDomainLockDiskAttach(virLockManagerPluginPtr plugin, virDomainObjPtr dom, virDomainDiskDefPtr disk) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, false); + virLockManagerPtr lock; int ret = -1; + if (!(lock = virDomainLockManagerNew(plugin, dom, false))) + return -1; + if (virDomainLockManagerAddDisk(lock, disk) < 0) goto cleanup; @@ -233,9 +251,12 @@ int virDomainLockDiskDetach(virLockManagerPluginPtr plugin, virDomainObjPtr dom, virDomainDiskDefPtr disk) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, false); + virLockManagerPtr lock; int ret = -1; + if (!(lock = virDomainLockManagerNew(plugin, dom, false))) + return -1; + if (virDomainLockManagerAddDisk(lock, disk) < 0) goto cleanup; @@ -255,9 +276,12 @@ int virDomainLockLeaseAttach(virLockManagerPluginPtr plugin, virDomainObjPtr dom, virDomainLeaseDefPtr lease) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, false); + virLockManagerPtr lock; int ret = -1; + if (!(lock = virDomainLockManagerNew(plugin, dom, false))) + return -1; + if (virDomainLockManagerAddLease(lock, lease) < 0) goto cleanup; @@ -276,9 +300,12 @@ int virDomainLockLeaseDetach(virLockManagerPluginPtr plugin, virDomainObjPtr dom, virDomainLeaseDefPtr lease) { - virLockManagerPtr lock = virDomainLockManagerNew(plugin, dom, false); + virLockManagerPtr lock; int ret = -1; + if (!(lock = virDomainLockManagerNew(plugin, dom, false))) + return -1; + if (virDomainLockManagerAddLease(lock, lease) < 0) goto cleanup; -- 1.7.4.4

2 1

[libvirt] Virt Tools Blog Planet (RSS feed aggregator)
by Daniel P. Berrange 14 Jun '11

14 Jun '11

The virt-tools.org website, launched last year, provides tutorials, videos, documentation, online help and roadmaps relevant to libvirt, libguestfs, gtk-vnc, spice, other related libraries, and tools or applications like virt-manager & virt-install. The site goal is to inform & assist end users, system administrators & application developers who wish to learn about the capabilities of the virt tools stack. The focus of most content is the, state of the art, Linux native KVM hypervisor, but writing about using other hypervisors using virt tools is also welcome. Today we are launching a new part of the site, the Blog Planet, to aggregate feeds from people working on virt tools related projects: http://planet.virt-tools.org/index.html The site is seeded with a mere 2 blogs thus far, and so we are looking for content from other people working on libvirt, libguestfs, gtk-vnc, spice, and any other infrastructure/tools which integrate with them. The aim is for the Blog Planet to contain content directly useful to end users, system administrators and application developers. To that end, it is desirable if your blog feed can be filtered to just include content directly related to virt tools. Any wordpress users can trivially do this by assigning either categories or tags to their posts, which the provide dedicated RSS feeds; other blog software hopefully has similar capabilities. To be included on the planet, send (off-list) your RSS feed URL and a Hackergotchi face icon (upto an absolute max of 96x96 pixels in size). Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

1 0

[libvirt] [PATCH 1/2] qemu: Scrape stdout for virtio console pty
by Cole Robinson 14 Jun '11

14 Jun '11

Currently we forget to do this and have to fallback to info chardev (which also fails, see following patch) Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/qemu/qemu_process.c | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 1efe024..349c25a 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -989,6 +989,16 @@ qemuProcessFindCharDevicePTYs(virDomainObjPtr vm, } } + if (vm->def->console) { + virDomainChrDefPtr chr = vm->def->console; + if (chr->source.type == VIR_DOMAIN_CHR_TYPE_PTY && + chr->targetType == VIR_DOMAIN_CHR_CONSOLE_TARGET_TYPE_VIRTIO) { + if ((ret = qemuProcessExtractTTYPath(output, &offset, + &chr->source.data.file.path)) != 0) + return ret; + } + } + return 0; } -- 1.7.4.4

2 4

[libvirt] [PATCH] Add support for network filter code in LXC driver
by Daniel P. Berrange 14 Jun '11

14 Jun '11

The LXC driver networking uses veth device pairs. These can be easily hooked into the network filtering code. * src/lxc/lxc_driver.c: Add calls to setup/teardown nwfilter --- src/lxc/lxc_driver.c | 12 ++++++++++-- 1 files changed, 10 insertions(+), 2 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 8eb87a2..4d14466 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -52,7 +52,7 @@ #include "hooks.h" #include "files.h" #include "fdstream.h" - +#include "domain_nwfilter.h" #define VIR_FROM_THIS VIR_FROM_LXC @@ -1027,6 +1027,8 @@ static void lxcVmCleanup(lxc_driver_t *driver, vethDelete(vm->def->nets[i]->ifname); } + virDomainConfVMNWFilterTeardown(vm); + if (driver->cgroup && virCgroupForDomain(driver->cgroup, vm->def->name, &cgroup, 0) == 0) { virCgroupRemove(cgroup); @@ -1146,6 +1148,10 @@ static int lxcSetupInterfaces(virConnectPtr conn, if (vethInterfaceUpOrDown(parentVeth, 1) < 0) goto error_exit; + + if (def->nets[i]->filter && + virDomainConfNWFilterInstantiate(conn, def->nets[i]) < 0) + goto error_exit; } rc = 0; @@ -1538,8 +1544,10 @@ cleanup: vethDelete(veths[i]); VIR_FREE(veths[i]); } - if (rc != 0) + if (rc != 0) { VIR_FORCE_CLOSE(priv->monitor); + virDomainConfVMNWFilterTeardown(vm); + } VIR_FORCE_CLOSE(parentTty); VIR_FREE(logfile); return rc; -- 1.7.4.4

3 3