- Devel - libvirt List Archives

[libvirt] Problem in attaching a interface to a domain
by sarveswara rao mygapula 22 May '08

22 May '08

Hello, I am facing problems in creating a network bridge using libvirt. My requirement is, i need to create a bridge on the host machine. The bridge shd get ip address from dhcp server(instead of static). I am not getting what exactly the xml format that virsh is looking for? I am using the following xml for creating a bridge. <network> <name>local</name> <bridge name="virbr1" /> <source dev="eth1"/> <mac address="00:11:22:33:44:55"/> </network> The bridge is getting created but not with the specified device and mac address. Bridge is not getting ip address at all. Can you please tell me the xml format for my requirement? Thanks.

2 1

[libvirt] [PATCH] Free object after *Destroy in python bindings
by Cole Robinson 21 May '08

21 May '08

The patch below fixes an issue in the python bindings with the vir*Destroy methods. After the object is successfully destroyed, the payload is cleared, using 'self._o = None'. This unfortunately screws up virt object reference counts, as the payload should be free'd using the appropriate vir*Free function. Thanks, Cole diff --git a/python/generator.py b/python/generator.py index cb57bff..9421981 100755 --- a/python/generator.py +++ b/python/generator.py @@ -629,10 +629,10 @@ function_classes = {} function_classes["None"] = [] function_post = { - 'virDomainDestroy': "self._o = None", - 'virNetworkDestroy': "self._o = None", - 'virStoragePoolDestroy': "self._o = None", - 'virStorageVolDestroy': "self._o = None", + 'virDomainDestroy': "del(self)", + 'virNetworkDestroy': "del(self)", + 'virStoragePoolDestroy': "del(self)", + 'virStorageVolDestroy': "del(self)", } # Functions returning an integral type which need special rules to

3 12

[libvirt] [PATCH] Fix destroy command memory leaks
by Cole Robinson 21 May '08

21 May '08

Some pieces of libvirt currently assume that the vir*Destroy functions will free the passed object upon success. In practice none of the current drivers seem to do this, resulting in memory leaks. The attached patch fixes the leaks I could find, as well as changes the comments for virDomainDestroy and virNetworkDestroy in libvirt.c to reflect reality. I also added a couple debug statements to hash.c where domain reference counts can be printed as they are changed. Thanks, Cole diff --git a/qemud/remote.c b/qemud/remote.c index a97641a..725152e 100644 --- a/qemud/remote.c +++ b/qemud/remote.c @@ -940,9 +940,11 @@ remoteDispatchDomainDestroy (struct qemud_server *server ATTRIBUTE_UNUSED, return -2; } - if (virDomainDestroy (dom) == -1) + if (virDomainDestroy (dom) == -1) { + virDomainFree(dom); return -1; - /* No need to free dom - destroy does it for us */ + } + virDomainFree(dom); return 0; } diff --git a/src/hash.c b/src/hash.c index 79421aa..a014990 100644 --- a/src/hash.c +++ b/src/hash.c @@ -842,6 +842,9 @@ __virGetDomain(virConnectPtr conn, const char *name, const unsigned char *uuid) goto error; } conn->refs++; + DEBUG0("virGetDomain: New hash entry."); + } else { + DEBUG("virGetDomain: Existing hash entry, refs now %d", ret->refs+1); } ret->refs++; pthread_mutex_unlock(&conn->lock); diff --git a/src/libvirt.c b/src/libvirt.c index 97f6bc3..9f6df8e 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -1390,10 +1390,9 @@ virDomainLookupByName(virConnectPtr conn, const char *name) * @domain: a domain object * * Destroy the domain object. The running instance is shutdown if not down - * already and all resources used by it are given back to the hypervisor. - * The data structure is freed and should not be used thereafter if the - * call does not return an error. - * This function may requires privileged access + * already and all resources used by it are given back to the hypervisor. This + * does not free the associated virDomainPtr object. + * This function may require privileged access * * Returns 0 in case of success and -1 in case of failure. */ @@ -3502,10 +3501,9 @@ virNetworkCreate(virNetworkPtr network) * @network: a network object * * Destroy the network object. The running instance is shutdown if not down - * already and all resources used by it are given back to the hypervisor. - * The data structure is freed and should not be used thereafter if the - * call does not return an error. - * This function may requires privileged access + * already and all resources used by it are given back to the hypervisor. This + * does not free the associated virNetworkPtr object. + * This function may require privileged access * * Returns 0 in case of success and -1 in case of failure. */ diff --git a/src/virsh.c b/src/virsh.c index 45af630..234fc36 100644 --- a/src/virsh.c +++ b/src/virsh.c @@ -1468,9 +1468,9 @@ cmdDestroy(vshControl * ctl, vshCmd * cmd) } else { vshError(ctl, FALSE, _("Failed to destroy domain %s"), name); ret = FALSE; - virDomainFree(dom); } + virDomainFree(dom); return ret; } @@ -2433,9 +2433,9 @@ cmdNetworkDestroy(vshControl * ctl, vshCmd * cmd) } else { vshError(ctl, FALSE, _("Failed to destroy network %s"), name); ret = FALSE; - virNetworkFree(network); } + virNetworkFree(network); return ret; } @@ -3161,9 +3161,9 @@ cmdPoolDestroy(vshControl * ctl, vshCmd * cmd) } else { vshError(ctl, FALSE, _("Failed to destroy pool %s"), name); ret = FALSE; - virStoragePoolFree(pool); } + virStoragePoolFree(pool); return ret; }

3 4

Re: [libvirt] Problems in using Storage APIs in libvirt
by Amudhan Gunasekaran 21 May '08

21 May '08

Hi Chris Lalancette, Thanks for your quick reply. I just removed the already installed libvirt from my machine and did a *make install* in my CVS version of the code (which is in my machine at: /root/libvirt). The *make install* completed without any problems. But when I tried to compile the code, I get other error messages. I actually use the command: gcc `pkg-config --cflags --libs libvirt` -o myexample myexample.c The error messages I got are: Package libvirt was not found in the pkg-config search path. Perhaps you should add the directory containing `libvirt.pc' to the PKG_CONFIG_PATH environment variable No package 'libvirt' found myexample.c:9:29: error: libvirt/libvirt.h: No such file or directory Previously, I used the same command to compile and it compiled without any problems (if I had not used storage related apis). I think I am doing some silly mistakes. Please help me to rectify it. Thanks, Amudhan. >>> Chris Lalancette <clalance(a)redhat.com> 05/22/08 12:10 AM >>> Amudhan Gunasekaran wrote: > Hi, > > I am getting the following the error when I try to use ANY storage related APIs. > > undefined reference to '<method name>' for example, > > undefined reference to virConnectListStoragePools. > > I checked in the libvirt.h and it has the method definition. I checked the libvirt.c and it has the implementations. I can not figure out why I get this error. I could use APIs related the network, for example virConnectListNetworks. Where could I have gone wrong and what are the work-arounds to solve them? > > I am using SLES 10 SP2. This probably means that libvirtd and/or virsh is still using the old copy of the library. If you have a libvirt package installed, and then you built the the CVS libvirt by hand, the CVS version probably got installed somewhere to /usr/local/lib, which is in the search path after where libvirt is probably installed to /usr/lib. Try removing your libvirt package and doing a "make install" in the CVS version; you should then have everything matching up. Chris Lalancette

1 0

[libvirt] Problems in using Storage APIs in libvirt
by Amudhan Gunasekaran 21 May '08

21 May '08

Hi, I am getting the following the error when I try to use ANY storage related APIs. undefined reference to '<method name>' for example, undefined reference to virConnectListStoragePools. I checked in the libvirt.h and it has the method definition. I checked the libvirt.c and it has the implementations. I can not figure out why I get this error. I could use APIs related the network, for example virConnectListNetworks. Where could I have gone wrong and what are the work-arounds to solve them? I am using SLES 10 SP2. Thanks in advance, Amudhan.

2 1

[libvirt] [PATCH] A few simple "copy-paste" errors
by Anton Protopopov 21 May '08

21 May '08

Hi, I found few simple errors in docs/formatdomain.html while scrolling it; the appended patch fixes them. Anton

2 1

[libvirt] PATCH: Fix removal of iptables FORWARD rules
by Daniel P. Berrange 21 May '08

21 May '08

The previous patch to add routed networking broke the removal of one of the FORWARD rules at shutdown. It was adding /sbin/iptables --table filter --insert FORWARD --destination 192.168.122.0/255.255.255.0 --out-interface virbr0 --match state --state ESTABLISHED,RELATED --jump ACCEPT But trying to remove /sbin/iptables --table filter --delete FORWARD --destination 192.168.122.0/255.255.255.0 --out-interface virbr0 --jump ACCEPT which wasn't matching on the state flags. This patch makes it use the correct removal code Dan. Index: src/qemu_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_driver.c,v retrieving revision 1.76 diff -u -r1.76 qemu_driver.c --- src/qemu_driver.c 16 May 2008 16:51:30 -0000 1.76 +++ src/qemu_driver.c 20 May 2008 20:41:03 -0000 @@ -1209,12 +1213,20 @@ struct qemud_network *network) { if (network->def->forward) { iptablesRemoveForwardMasquerade(driver->iptables, - network->def->network, - network->def->forwardDev); - iptablesRemoveForwardAllowIn(driver->iptables, - network->def->network, - network->bridge, - network->def->forwardDev); + network->def->network, + network->def->forwardDev); + + if (network->def->forwardMode == QEMUD_NET_FORWARD_NAT) + iptablesRemoveForwardAllowRelatedIn(driver->iptables, + network->def->network, + network->bridge, + network->def->forwardDev); + else if (network->def->forwardMode == QEMUD_NET_FORWARD_ROUTE) + iptablesRemoveForwardAllowIn(driver->iptables, + network->def->network, + network->bridge, + network->def->forwardDev); + iptablesRemoveForwardAllowOut(driver->iptables, network->def->network, network->bridge, -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] virsh create Guest Crashing
by Kenneth Nagin 21 May '08

21 May '08

An image that comes up successfully is crash when be created with virsh: This is libvirt create config file: <domain type='xen' > <name>ken-30</name> <os> <type>linux</type> <kernel>/data/images/boot/vmlinuz-2.6.21.7-3.fc8xen</kernel> <initrd>/data/images/boot/2.6.21.7-3.fc8xen-fedora.fc8.img</initrd> </os> <memory>131072</memory> <vcpu>1</vcpu> <devices> <disk type='file'> <source file='/data/images/nagin/c1/ken-30/fedora.fc8.img.sd'/> <target dev='sda1'/> </disk> <disk type='file'> <source file='/data/images/nagin/c1/ken-30/fedora.swap'/>. <target dev='sda2'/> </disk> <interface type='bridge'> <source bridge='virbr0'/> <script path='/etc/xen/scripts/vif-bridge'/> </interface> </devices> </domain> This the original xen configuration file: kernel = "/data/images/boot/vmlinuz-2.6.21.7-3.fc8xen" ramdisk = "/data/images/boot/2.6.21.7-3.fc8xen-fedora.fc8.img">. memory = 128 name = "ken-30" vif = [ 'bridge=virbr0' ] disk = ['tap:aio:/data/images/nagin/c1/ken-30/fedora.fc8.img,xvda1,w','tap:aio:/data/images/nagin/c1/ken-30/fedora.swap,xvda2,w'] root = "/dev/xvda1 ro" vfb = [ "type=vnc, vncdisplay=30" ] It crashes shortly after creation. This is the snapshot of the xen log: [2008-05-19 12:00:56 4937] DEBUG (XendDomainInfo:1040) XendDomainInfo.handleShutdownWatch [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices vif. [2008-05-19 12:00:56 4937] DEBUG (DevController:155) Waiting for 0. [2008-05-19 12:00:56 4937] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/vif/59/0/hotplug-status. [2008-05-19 12:00:56 4937] DEBUG (DevController:608) hotplugStatusCallback 1. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices usb. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices vbd. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices irq. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices vkbd. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices vfb.. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices console. [2008-05-19 12:00:56 4937] DEBUG (DevController:155) Waiting for 0. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices pci. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices ioports. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices tap. [2008-05-19 12:00:56 4937] DEBUG (DevController:155) Waiting for 51713. [2008-05-19 12:00:56 4937] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/tap/59/51713/hotplug-status. [2008-05-19 12:00:56 4937] DEBUG (DevController:608) hotplugStatusCallback 1. [2008-05-19 12:00:56 4937] DEBUG (DevController:155) Waiting for 51714. [2008-05-19 12:00:56 4937] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/tap/59/51714/hotplug-status. [2008-05-19 12:00:56 4937] DEBUG (DevController:608) hotplugStatusCallback 1. [2008-05-19 12:00:56 4937] DEBUG (DevController:150) Waiting for devices vtpm. [2008-05-19 12:00:56 4937] INFO (XendDomain:1130) Domain ken-30 (59) unpaused. [2008-05-19 12:01:00 4937] WARNING (XendDomainInfo:1203) Domain has crashed: name=ken-30 id=59. [2008-05-19 12:01:00 4937] DEBUG (XendDomainInfo:1821) XendDomainInfo.destroyDomain(59) I've changed the XML configuration many different ways but nothing seems to help. Any suggestions? --Kenneth Nagin

4 6

[libvirt] Domain XML format using defined storage volume
by Stefan de Konink 21 May '08

21 May '08

I have now a 'netapp'-pool, and lun-0..4. They all have a path. <disk type='block'> <source dev='?path?'> <target dev='xvda' bus='xen'/> </disk> Now I wonder, is it possible to have here somethinglike storage://netapp/lun-0? Or should I use the path that comes out of the vol-list? >From the standpoint: we make it easy for the user I would prefer not to provide a Linux specific path. Stefan

2 15

[libvirt] PATCH: Write pidfile earlier in startup
by Daniel P. Berrange 20 May '08

20 May '08

When used with the --daemon arg, libvirtd will write out a pid file to /var/run/libvirtd.pid and exit if this file already exists. Unfortuantely it does this quite late in its startup procedure - in particular *after* the call to qemudInitialize(), which in turns initializes the drivers, which in turn autostarts all the VMs and networks. So if you start a 2nd libvirtd instance it would do autostart before it got to checking for existing PID file. This is clearly very bad because the same VM would be started twice resulting in data corruption. Fortunately the Red Hat / Fedora initscript also checks the PID file before even starting the daemon, but this can affect people starting the daemon directly. So this patch makes the goDaemon() / pidfile creation the very first thing the daemon does after parsing command line arguments. This also results in greatly increased startup time for OS, since the initscript doesn't have to wait for it to auto-start all the VMs & networks before it daemonizes. It also initializes the signal handlers before calling qemudInitialize() since these really need to be setup before we start any VMs / child procs. Dan. Index: qemud/qemud.c =================================================================== RCS file: /data/cvs/libvirt/qemud/qemud.c,v retrieving revision 1.100 diff -u -p -r1.100 qemud.c --- qemud/qemud.c 15 May 2008 06:12:32 -0000 1.100 +++ qemud/qemud.c 16 May 2008 15:58:43 -0000 @@ -2143,6 +2143,24 @@ int main(int argc, char **argv) { } } + if (godaemon) { + openlog("libvirtd", 0, 0); + if (qemudGoDaemon() < 0) { + qemudLog(QEMUD_ERR, _("Failed to fork as daemon: %s"), + strerror(errno)); + goto error1; + } + + /* Choose the name of the PID file. */ + if (!pid_file) { + if (REMOTE_PID_FILE[0] != '\0') + pid_file = REMOTE_PID_FILE; + } + + if (pid_file && qemudWritePidFile (pid_file) < 0) + goto error1; + } + if (pipe(sigpipe) < 0 || qemudSetNonBlock(sigpipe[0]) < 0 || qemudSetNonBlock(sigpipe[1]) < 0 || @@ -2154,6 +2172,19 @@ int main(int argc, char **argv) { } sigwrite = sigpipe[1]; + sig_action.sa_sigaction = sig_handler; + sig_action.sa_flags = SA_SIGINFO; + sigemptyset(&sig_action.sa_mask); + + sigaction(SIGHUP, &sig_action, NULL); + sigaction(SIGINT, &sig_action, NULL); + sigaction(SIGQUIT, &sig_action, NULL); + sigaction(SIGTERM, &sig_action, NULL); + sigaction(SIGCHLD, &sig_action, NULL); + + sig_action.sa_handler = SIG_IGN; + sigaction(SIGPIPE, &sig_action, NULL); + if (!(server = qemudInitialize(sigpipe[0]))) { ret = 2; goto error1; @@ -2175,37 +2206,6 @@ int main(int argc, char **argv) { sockdirname); } - if (godaemon) { - openlog("libvirtd", 0, 0); - if (qemudGoDaemon() < 0) { - qemudLog(QEMUD_ERR, _("Failed to fork as daemon: %s"), - strerror(errno)); - goto error1; - } - - /* Choose the name of the PID file. */ - if (!pid_file) { - if (REMOTE_PID_FILE[0] != '\0') - pid_file = REMOTE_PID_FILE; - } - - if (pid_file && qemudWritePidFile (pid_file) < 0) - goto error1; - } - - sig_action.sa_sigaction = sig_handler; - sig_action.sa_flags = SA_SIGINFO; - sigemptyset(&sig_action.sa_mask); - - sigaction(SIGHUP, &sig_action, NULL); - sigaction(SIGINT, &sig_action, NULL); - sigaction(SIGQUIT, &sig_action, NULL); - sigaction(SIGTERM, &sig_action, NULL); - sigaction(SIGCHLD, &sig_action, NULL); - - sig_action.sa_handler = SIG_IGN; - sigaction(SIGPIPE, &sig_action, NULL); - if (virEventAddHandleImpl(sigpipe[0], POLLIN, qemudDispatchSignalEvent, -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 4