July 2009 - Devel - libvirt List Archives

[libvirt] Pre-0.7.0 release for testing
by Daniel Veillard 30 Jul '09

30 Jul '09

Considering the large amount of new code and features expected for 0.7.0 at the end of the week, I made a pre-release at ftp://libvirt.org/libvirt/libvirt-0.7.0-0.1.gitf055724.tar.gz if people have time to test and report, that would be great ! I expect to make the actual release this thursday, unless there is too many issues in which case I would postpone to beginning of next week (or whenever main problems get fixed...) thanks ! Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 4

[libvirt] Fix -Werror build error regarding to '#ifdef IFF_VNET_HDR'
by Maximilian Wilhelm 30 Jul '09

30 Jul '09

Hi! Our build bot found a build error when building libVirt with -Werror introduced in b4f62abbf1191c8fbab3306b4bf2f2567e18067f kvm/virtio: Set IFF_VNET_HDR when setting up tap fds The attached patch kind of fixes the build error but I'm not sure, if this is the final solution for this problem, though... Ciao Max -- The real problem with C++ for kernel modules is: the language just sucks. -- Linus Torvalds

2 3

[libvirt] [PATCH] Drop the leading lib from libvirt. Based on testing by thomas@scripty.at this is a better way to load the libarary
by Bryan Kearney 30 Jul '09

30 Jul '09

--- src/main/java/org/libvirt/jna/Libvirt.java | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/main/java/org/libvirt/jna/Libvirt.java b/src/main/java/org/libvirt/jna/Libvirt.java index 33914ac..62f7ba3 100644 --- a/src/main/java/org/libvirt/jna/Libvirt.java +++ b/src/main/java/org/libvirt/jna/Libvirt.java @@ -20,7 +20,7 @@ public interface Libvirt extends Library { public int authCallback(virConnectCredential cred, int ncred, Pointer cbdata); } - Libvirt INSTANCE = (Libvirt) Native.loadLibrary("libvirt", Libvirt.class); + Libvirt INSTANCE = (Libvirt) Native.loadLibrary("virt", Libvirt.class); // Constants we need public static int VIR_UUID_BUFLEN = 16; -- 1.6.0.6

3 2

[libvirt] RFC: libvirt java bindings based on JNA
by Bryan Kearney 30 Jul '09

30 Jul '09

I would like to get some comments on an initial cut of the java bindings on top of JNA[1]. They are not 100% complete (see below), but they are good enough to be criticized. In addition, if there are features additions to the API which people are interested in, I would appreciate the feedback on that. JNA is a java implementation of the Foreign Function Interface [2]. The main benefit is that it removes the need for all the nasty C glue code. What replaces it is pure java mapping files. If you look in the org.libvirt.jna package in the code, you will see that the 2928 lines of jni code have been replaced by 427 lines of java code. To be fair, some of the business logic from the C code is now in the org.libvirt package... but you get the point. The original java API has remained the same. The open issues with the code are: 1) Authorization callback has not been tested. 2) GetUUID is not returning the correct int array. 3) Find by UUID is not working These all are mapping issues to be worked out. The rest of the code should be working, but I would appreciate it being exercised. The main drawback to this approach is that it requires the use of java 1.6. I looked on EPEL, F10, and F11.. and java 1.6 and the JNA jar file are available so I did not believe this was an issue. I will send a patch stream to the list, but you can also get the code via an SRPM [3] or via github [4]. -- bk [1] https://jna.dev.java.net/ [2] http://en.wikipedia.org/wiki/Foreign_function_interface [3] http://bkearney.fedorapeople.org/libvirt-java-0.3.0pre-1.fc10.src.rpm [4] http://github.com/bkearney/libvirt-java/tree/jna

3 11

[libvirt] PATCH: Always start dnsmasq to provide DNS
by Daniel P. Berrange 30 Jul '09

30 Jul '09

Currently we only start dnsmasq if DHCP ranges are configured. https://bugzilla.redhat.com/show_bug.cgi?id=508965 points out that this prevents use of DNS in guests using static IP config. Thus this patch makes us always start dnsmasq, regardless of DHCP config if we have an IP addr configured. There is no need for the --no-dhcp-interface= that BZ mentions, since that's just to allow you to restrict which interfaces dnsmasq does DHCP on - we've already restricted it to just out interface, and leaving out the DHCP ranges disables its DHCP server for every interface Regards, Daniel commit deb71927a3551b3a1554351ef098644aec978a09 Author: Daniel P. Berrange <berrange(a)redhat.com> Date: Thu Jul 30 14:52:31 2009 +0100 Allow dnsmasq to provide DNS without DHCP * src/network_driver.c: Always start dnsmasq to allow it to provide DNS, even if no DHCP ranges are enabled diff --git a/src/network_driver.c b/src/network_driver.c index 9621782..1683631 100644 --- a/src/network_driver.c +++ b/src/network_driver.c @@ -136,8 +136,9 @@ networkFindActiveConfigs(struct network_driver *driver) { brHasBridge(driver->brctl, obj->def->bridge) == 0) { obj->active = 1; - /* Finally try and read dnsmasq pid if any DHCP ranges are set */ - if (obj->def->nranges && + /* Finally try and read dnsmasq pid if any */ + if ((obj->def->ipAddress || + obj->def->nranges) && virFileReadPid(NETWORK_PID_DIR, obj->def->name, &obj->dnsmasqPid) == 0) { @@ -844,7 +845,8 @@ static int networkStartNetworkDaemon(virConnectPtr conn, goto err_delbr2; } - if (network->def->nranges && + if ((network->def->ipAddress || + network->def->nranges) && dhcpStartDhcpDaemon(conn, network) < 0) goto err_delbr2; -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] [PATCH] Remove MAX_TAP_ID, take 3
by Aron Griffis 30 Jul '09

30 Jul '09

This is a resend of take 2 to fix formatting problems in the patch. No other changes. As far as I can tell, there's no reason to format the device string in brAddTap(). Delegate the job to TUNSETIFF, thereby removing the loop and the MAX_TAP_ID artificial limit. This patch allows me to get 421 guests running before hitting other limits. Signed-off-by: Aron Griffis <aron.griffis(a)hp.com> diff --git a/src/bridge.c b/src/bridge.c index 0509afd..ec37192 100644 --- a/src/bridge.c +++ b/src/bridge.c @@ -49,8 +49,6 @@ #include "util.h" #include "logging.h" -#define MAX_TAP_ID 256 - #define JIFFIES_TO_MS(j) (((j)*1000)/HZ) #define MS_TO_JIFFIES(ms) (((ms)*HZ)/1000) @@ -466,76 +464,52 @@ brAddTap(brControl *ctl, int vnet_hdr, int *tapfd) { - int id, subst, fd; + int fd, len; + struct ifreq ifr = {0}; if (!ctl || !ctl->fd || !bridge || !ifname) return EINVAL; - subst = id = 0; - - if (strstr(*ifname, "%d")) - subst = 1; - if ((fd = open("/dev/net/tun", O_RDWR)) < 0) return errno; - if (vnet_hdr) - vnet_hdr = brProbeVnetHdr(fd); + ifr.ifr_flags = IFF_TAP|IFF_NO_PI; - do { - struct ifreq try; - int len; +#ifdef IFF_VNET_HDR + if (vnet_hdr && brProbeVnetHdr(fd)) + ifr.ifr_flags |= IFF_VNET_HDR; +#endif - memset(&try, 0, sizeof(struct ifreq)); + strncpy(ifr.ifr_name, *ifname, IFNAMSIZ-1); - try.ifr_flags = IFF_TAP|IFF_NO_PI; + if (ioctl(fd, TUNSETIFF, &ifr) < 0) + goto error; -#ifdef IFF_VNET_HDR - if (vnet_hdr) - try.ifr_flags |= IFF_VNET_HDR; -#endif + len = strlen(ifr.ifr_name); + if (len >= BR_IFNAME_MAXLEN - 1) { + errno = EINVAL; + goto error; + } - if (subst) { - len = snprintf(try.ifr_name, BR_IFNAME_MAXLEN, *ifname, id); - if (len >= BR_IFNAME_MAXLEN) { - errno = EADDRINUSE; - goto error; - } - } else { - len = strlen(*ifname); - if (len >= BR_IFNAME_MAXLEN - 1) { - errno = EINVAL; - goto error; - } - - strncpy(try.ifr_name, *ifname, len); - try.ifr_name[len] = '\0'; - } - - if (ioctl(fd, TUNSETIFF, &try) == 0) { - /* We need to set the interface MTU before adding it - * to the bridge, because the bridge will have its - * MTU adjusted automatically when we add the new interface. - */ - if ((errno = brSetInterfaceMtu(ctl, bridge, try.ifr_name))) - goto error; - if ((errno = brAddInterface(ctl, bridge, try.ifr_name))) - goto error; - if ((errno = brSetInterfaceUp(ctl, try.ifr_name, 1))) - goto error; - if (!tapfd && - (errno = ioctl(fd, TUNSETPERSIST, 1))) - goto error; - VIR_FREE(*ifname); - if (!(*ifname = strdup(try.ifr_name))) - goto error; - if (tapfd) - *tapfd = fd; - return 0; - } - - id++; - } while (subst && id <= MAX_TAP_ID); + /* We need to set the interface MTU before adding it + * to the bridge, because the bridge will have its + * MTU adjusted automatically when we add the new interface. + */ + if ((errno = brSetInterfaceMtu(ctl, bridge, ifr.ifr_name))) + goto error; + if ((errno = brAddInterface(ctl, bridge, ifr.ifr_name))) + goto error; + if ((errno = brSetInterfaceUp(ctl, ifr.ifr_name, 1))) + goto error; + if (!tapfd && + (errno = ioctl(fd, TUNSETPERSIST, 1))) + goto error; + VIR_FREE(*ifname); + if (!(*ifname = strdup(ifr.ifr_name))) + goto error; + if (tapfd) + *tapfd = fd; + return 0; error: close(fd);

4 9

[libvirt] VM creation failed : Permission denied : bind(unix:/var/run/libvirt/qemu//VM.monitor)
by Pierre-Gilles Mialon 30 Jul '09

30 Jul '09

I use Ovirt under Fedora 11 with the latest developpement package and the VM creation failed with : bind(unix:/var/run/libvirt/qemu//VM-4.monitor): Permission denied See the logs : On the Ovirt Node ==> /var/log/messages <== Jul 29 20:43:37 node62 kernel: device vnet0 entered promiscuous mode Jul 29 20:43:37 node62 kernel: breth0: port 2(vnet0) entering learning state Jul 29 20:43:37 node62 libvirtd: 20:43:37.873: info : qemudDispatchSignalEvent:370 : Received unexpected signal 17 Jul 29 20:43:37 node62 libvirtd: 20:43:37.877: info : qemudDispatchSignalEvent:370 : Received unexpected signal 17 Jul 29 20:43:37 node62 kernel: breth0: port 2(vnet0) entering disabled state Jul 29 20:43:37 node62 kernel: device vnet0 left promiscuous mode Jul 29 20:43:37 node62 kernel: breth0: port 2(vnet0) entering disabled state Jul 29 20:43:37 node62 libvirtd: 20:43:37.977: error : qemudReadLogOutput:816 : internal error Process exited while reading console log output Jul 29 20:43:37 node62 libvirtd: 20:43:37.977: error : qemudWaitForMonitor:1103 : internal error unable to start guest: bind(unix:/var/run/libvirt/qemu//VM-4.monitor): Permission denied#012qemu: could not open monitor device 'unix:/var/run/libvirt/qemu//VM-4.monitor,server,nowait'#012 Jul 29 20:43:38 node62 libvirtd: 20:43:38.073: error : qemudDomainLookupByUUID:2644 : Domain not found: no domain with matching uuid '492f836f-5123-e185-39c2-09c5dd43a7f6' Jul 29 20:43:38 node62 libvirt-qpid: Error: virDomainLookupByUUIDString Subsystem qemu: Domain not found: no domain with matching uuid '492f836f-5123- e185-39c2-09c5dd43a7f6' in NodeWrap.cpp:syncDomains:241 code: 42 ==> /var/log/libvirt/qemu/VM-4.log <== LC_ALL=C PATH=/sbin:/usr/sbin:/bin:/usr/bin HOME=/root USER=root LOGNAME=root /usr/bin/qemu-kvm -S -M pc -m 256 -smp 1 -name VM-4 -uuid 492f836f-5123- e185-39c2-09c5dd43a7f6 -monitor unix:/var/run/libvirt/qemu//VM-4.monitor,server,nowait -boot n -net nic,macaddr=00:16:3e:10:de:fe,vlan=0,name=nic.0 -net tap,fd=18,vlan=0,name=tap.0 -serial pty -parallel none -usb -vnc 0.0.0.0:0 - vga cirrus bind(unix:/var/run/libvirt/qemu//VM-4.monitor): Permission denied qemu: could not open monitor device 'unix:/var/run/libvirt/qemu//VM-4.monitor,server,nowait' ==> # rpm -qa | grep libvirt <== libvirt-0.7.0-0.2.gitf055724.fc11.x86_64 libvirt-python-0.7.0-0.2.gitf055724.fc11.x86_64 libvirt-client-0.7.0-0.2.gitf055724.fc11.x86_64 libvirt-qpid-0.2.17-0.fc11.x86_64 Regards -- Pierre-Gilles Mialon Linagora :: http://www.linagora.com Responsable hébergement :: Head of Hosting services pmialon(a)linagora.com :: +33.1 58 18 65 46

5 11

[libvirt] VMware ESX: Scheduler documentation and cleanup
by Matthias Bolte 30 Jul '09

30 Jul '09

Add some documentation about the CPU scheduler parameters and remove some old, unnecessary compensation code, since virsh uses the proposed parameter types now. Documentation in HTML form will be added soon. Matthias

2 1

[libvirt] vcpupin broken towards Xen using xend
by Henrik Persson 30 Jul '09

30 Jul '09

Hi, During the update of the cpu affinity map format, the old format was broken ( '[' was switched with ']'). Also, the mapstr might contain garbage since the string is not initiate (however buf incorrectly is). This small patch fixes both the problems. Regards, Henrik

2 1

[libvirt] PATCH: Fix RPM upgrades to create qemu user
by Daniel P. Berrange 30 Jul '09

30 Jul '09

If upgrading from F11 to F12 it is possible the qemu user/group does not yet exist. Since the libvirt RPM needs to deploy files owned by this user/group, regardless of whether QEMU is yet installed, we need to make sure to create this user/group in %pre, otherwise fiels will end up with wrong ownership. Daniel diff -u -p -r1.159 libvirt.spec --- libvirt.spec 29 Jul 2009 14:58:18 -0000 1.159 +++ libvirt.spec 30 Jul 2009 11:06:53 -0000 @@ -419,6 +419,18 @@ chmod 0644 $RPM_BUILD_ROOT%{_sysconfdir} %clean rm -fr %{buildroot} +%pre +%if 0%{?fedora} >= 12 +# Normally 'setup' adds this in /etc/passwd, but this is +# here for case of upgrades from earlier Fedora. This +# UID/GID pair is reserved for qemu:qemu +getent group kvm >/dev/null || groupadd -g 36 -r kvm +getent group qemu >/dev/null || groupadd -g 107 -r qemu +getent passwd qemu >/dev/null || \ + useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \ + -c "qemu user" qemu +%endif + %post %if %{with_libvirtd} -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1