December 2008 - Devel - libvirt List Archives

[libvirt] [PATCH] bridge: set MTU of tap iface to the same value of the bridge
by Eduardo Habkost 04 Dec '08

04 Dec '08

Hi, This is a follow-up to the RFC patch sent by Chris Wright, at: https://www.redhat.com/archives/libvir-list/2008-November/msg00225.html With this patch, instead of hardcoding the MTU to an high value, we set the MTU of the tap interface to the same MTU value set on the bridge interface. The current code uses the default tap MTU value and lowers the bridge interface MTU without any need, killing performance. I will send another patch later for allowing the tap MTU to be configured on the XML, but the behavior added by this patch will be kept as the default. The default behavior should be enough for most use cases where a larger MTU is wanted for the bridged interfaces. Signed-off-by: Eduardo Habkost <ehabkost(a)redhat.com> --- src/bridge.c | 98 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 98 insertions(+), 0 deletions(-) diff --git a/src/bridge.c b/src/bridge.c index 4090163..60334b6 100644 --- a/src/bridge.c +++ b/src/bridge.c @@ -276,6 +276,98 @@ brDeleteInterface(brControl *ctl ATTRIBUTE_UNUSED, #endif /** + * ifGetMtu + * @ctl: bridge control pointer + * @ifname: interface name get MTU for + * + * This function gets the @mtu value set for a given interface @ifname. + * + * Returns the MTU value in case of success. + * On error, returns -1 and sets errno accordingly + */ +static int ifGetMtu(brControl *ctl, const char *ifname) +{ + struct ifreq ifr; + int len; + + if (!ctl || !ifname) { + errno = EINVAL; + return -1; + } + + if ((len = strlen(ifname)) >= BR_IFNAME_MAXLEN) { + errno = EINVAL; + return -1; + } + + memset(&ifr, 0, sizeof(struct ifreq)); + + strncpy(ifr.ifr_name, ifname, len); + ifr.ifr_name[len] = '\0'; + + if (ioctl(ctl->fd, SIOCGIFMTU, &ifr)) + return -1; + + return ifr.ifr_mtu; + +} + +/** + * ifSetMtu: + * @ctl: bridge control pointer + * @ifname: interface name to set MTU for + * @mtu: MTU value + * + * This function sets the @mtu for a given interface @ifname. Typically + * used on a tap device to set up for Jumbo Frames. + * + * Returns 0 in case of success or an errno code in case of failure. + */ +static int ifSetMtu(brControl *ctl, const char *ifname, int mtu) +{ + struct ifreq ifr; + int len; + + if (!ctl || !ifname) + return EINVAL; + + if ((len = strlen(ifname)) >= BR_IFNAME_MAXLEN) + return EINVAL; + + memset(&ifr, 0, sizeof(struct ifreq)); + + strncpy(ifr.ifr_name, ifname, len); + ifr.ifr_name[len] = '\0'; + ifr.ifr_mtu = mtu; + + return ioctl(ctl->fd, SIOCSIFMTU, &ifr) == 0 ? 0 : errno; +} + +/** + * brSetInterfaceMtu + * @ctl: bridge control pointer + * @bridge: name of the bridge interface + * @ifname: name of the interface whose MTU we want to set + * + * Sets the interface mtu to the same MTU of the bridge + * + * Returns 0 in case of success or an errno code in case of failure. + */ +static int brSetInterfaceMtu(brControl *ctl, + const char *bridge, + const char *ifname) +{ + int mtu = ifGetMtu(ctl, bridge); + + fprintf(stderr, "mtu: %d\n", mtu); + + if (mtu < 0) + return errno; + + return ifSetMtu(ctl, ifname, mtu); +} + +/** * brAddTap: * @ctl: bridge control pointer * @bridge: the bridge name @@ -334,6 +426,12 @@ brAddTap(brControl *ctl, } 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))) -- 1.5.5.GIT

5 8

[libvirt] [PATCH] also look for /usr/bin/kvm
by Guido Günther 04 Dec '08

04 Dec '08

We do so in qemudCapsInitGuest() but not in qemudProbe(). O.k. to apply? -- Guido

3 3

[libvirt] is libvirt-python default in libvirt compilation from source?
by Shanmuga Rajan 03 Dec '08

03 Dec '08

I was trying to install libvirt 0.5.0 from source.. It seems it went well. but when i tried to import libvirt module in python i got error no module named libvirt then i searched entire system for libvirt.py then i came to know that python bindings for libvirt is not installed. is it that python bindings for libvirt not default in compilation....? if i have to install python bindings then what option should i use? i tried to find option from ./configure --help from libvirt source directory but i didnt get anything. Can any one help me in this. Thanks and Regards, Shan

2 1

[libvirt] Allowing <interface type="none"> for network interface definition
by Gihan Munasinghe 02 Dec '08

02 Dec '08

Hi I am writing a management stack for our platform using libvirt. We have many DomU images which all have PV drivers loaded for network.I want to start these DomU guests with only the PV network device being visible from within the DomU. This is a requirement in the platform and hopefully, I presume there are other people with the same requirement.. But when creating a server, libvirt sends (type ioemu) tag to xend. Which make all of our vm's to have 2 network cards. Using the normal xm command I can send type=none ( vif = [ "mac=00:16:3e:00:a5:57,bridge=eth0,script=vif-bridge,type=none"]) to xend. Which will send qeum-dm "-net none" switch which solves the problem. After doing some investigation, I did some code change in libvirt0.5.0 to support a attribute called "none" with in <interface type="">, that will send xend the tag (type none). Allowing qeum-dm to be configured with "-net none" switch. ;-) So you can define your network tag as <interface type='none'>   <mac address='00:16:3e:00:a5:57'/> <source bridge='eth0'/> <target dev='vif47.0'/> </interface> I this the proper way of doing it? :-\ Would this be a general functionality that libvirt community will be interested in (I think this add bit more flexibility to libvirt of the way user want to write their own management stack). If so could I contribute the patch to the mailing list. Looking forward to hearing from you all, I am new to libvirt therefore any comments will be greatly appreciated :-) Thanks Gihan -- Gihan Munasinghe R&D Team Leader XCalibre Communications Ltd. www.flexiscale.com

2 2

[libvirt] [PATCH 1/2] Java bindings for domain events
by David Lively 02 Dec '08

02 Dec '08

The attached patch (against libvirt-java) implements Java bindings for libvirt domain events. This version provides a libvirt EventImpl running in its own Java Thread, and provides per-Connect synchronization that makes using the bindings thread-safe. (Note the Domain, Network, StoragePool, and StorageVol methods also synchronize on their Connect object, as required by libvirt. I have similar changes for NodeDevice.java that need to be made when that code is checked in.) This version of the patch also implements and uses an enum class (DomainEvent.Type), as suggested by Tóth István. IMPORTANT: THIS PATCH WILL BREAK THINGS UNLESS THE NEXT [PATCH 2/2] IS APPLIED TO libvirt FIRST. Also, libvirt must be compiled WITH_PTHREADS for Java events to work. Dave

3 12

[libvirt] [PATCH] post-release patch for symbols
by Daniel Veillard 02 Dec '08

02 Dec '08

Assuming we release a 0.5.1 soon. Maybe the code with threading should be on a 0.6.0 considering the amount of changes though. 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/

3 3

[libvirt] review fallout: short memset
by Jim Meyering 02 Dec '08

02 Dec '08

While reviewing unrelated changes, I spotted a short memset: char **names; ... memset(names, 0, maxnames); That zeros out 1/4 or 1/8 of the memory than it should. It should be doing this: memset(names, 0, maxnames * sizeof (*names)); I checked all memset uses and found a total of 6 uses like that. This fixes them: >From 614b0064eadc3ecef0690d5c65bb531844a3091d Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 2 Dec 2008 14:49:03 +0100 Subject: [PATCH] fix inadequate initialization in storage and test drivers * src/storage_driver.c (storageListPools): Set all "names" entries to 0. (storageListDefinedPools, storagePoolListVolumes): Likewise. * src/test.c (testStoragePoolListVolumes): Likewise. --- src/storage_driver.c | 8 ++++---- src/test.c | 4 ++-- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/src/storage_driver.c b/src/storage_driver.c index 366820b..53388f1 100644 --- a/src/storage_driver.c +++ b/src/storage_driver.c @@ -347,7 +347,7 @@ storageListPools(virConnectPtr conn, free(names[i]); names[i] = NULL; } - memset(names, 0, nnames); + memset(names, 0, nnames * sizeof(*names)); return -1; } @@ -389,7 +389,7 @@ storageListDefinedPools(virConnectPtr conn, free(names[i]); names[i] = NULL; } - memset(names, 0, nnames); + memset(names, 0, nnames * sizeof(*names)); return -1; } @@ -880,7 +880,7 @@ storagePoolListVolumes(virStoragePoolPtr obj, return -1; } - memset(names, 0, maxnames); + memset(names, 0, maxnames * sizeof(*names)); for (i = 0 ; i < pool->volumes.count && n < maxnames ; i++) { if ((names[n++] = strdup(pool->volumes.objs[i]->name)) == NULL) { virStorageReportError(obj->conn, VIR_ERR_NO_MEMORY, @@ -895,7 +895,7 @@ storagePoolListVolumes(virStoragePoolPtr obj, for (n = 0 ; n < maxnames ; n++) VIR_FREE(names[i]); - memset(names, 0, maxnames); + memset(names, 0, maxnames * sizeof(*names)); return -1; } diff --git a/src/test.c b/src/test.c index 7998886..3e942da 100644 --- a/src/test.c +++ b/src/test.c @@ -1951,7 +1951,7 @@ testStoragePoolListVolumes(virStoragePoolPtr obj, POOL_IS_ACTIVE(privpool, -1); int i = 0, n = 0; - memset(names, 0, maxnames); + memset(names, 0, maxnames * sizeof(*names)); for (i = 0 ; i < privpool->volumes.count && n < maxnames ; i++) { if ((names[n++] = strdup(privpool->volumes.objs[i]->name)) == NULL) { testError(obj->conn, VIR_ERR_NO_MEMORY, "%s", _("name")); @@ -1965,7 +1965,7 @@ testStoragePoolListVolumes(virStoragePoolPtr obj, for (n = 0 ; n < maxnames ; n++) VIR_FREE(names[i]); - memset(names, 0, maxnames); + memset(names, 0, maxnames * sizeof(*names)); return -1; } -- 1.6.0.4.1044.g77718

3 2

[libvirt] [PATCH 2/2]: Call udevsettle in the appropriate places
by Chris Lalancette 02 Dec '08

02 Dec '08

Instead of relying solely on polling for /dev devices to appear in libvirt, we really should be synchronizing against udev. This is generally done by a call to udevsettle, which is exactly what this patch implements for the storage backends that are likely to create new /dev nodes. I believe I've read that even after udevsettle, you are not guaranteed that devices are all the way created, so we still need the polling in the rest of the sources, but this should give us a much better chance of things existing as we expect. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> Index: src/storage_backend.c =================================================================== RCS file: /data/cvs/libvirt/src/storage_backend.c,v retrieving revision 1.29 diff -u -r1.29 storage_backend.c --- a/src/storage_backend.c 17 Nov 2008 11:19:33 -0000 1.29 +++ b/src/storage_backend.c 26 Nov 2008 12:40:19 -0000 @@ -270,6 +270,20 @@ return 0; } +void virStorageBackendWaitForDevices(virConnectPtr conn) +{ + const char *const settleprog[] = { UDEVSETTLE, NULL }; + int exitstatus; + + /* + * NOTE: we ignore errors here; this is just to make sure that any device + * nodes that are being created finish before we try to + * scan them. If this fails (udevsettle exits badly, or doesn't exist + * at all), then we still have a fallback mechanism anyway. + */ + virRun(conn, settleprog, &exitstatus); +} + /* * Given a volume path directly in /dev/XXX, iterate over the * entries in the directory pool->def->target.path and find the Index: src/storage_backend.h =================================================================== RCS file: /data/cvs/libvirt/src/storage_backend.h,v retrieving revision 1.13 diff -u -r1.13 storage_backend.h --- a/src/storage_backend.h 17 Nov 2008 11:19:33 -0000 1.13 +++ b/src/storage_backend.h 26 Nov 2008 12:40:19 -0000 @@ -69,6 +69,8 @@ int fd, int withCapacity); +void virStorageBackendWaitForDevices(virConnectPtr conn); + char *virStorageBackendStablePath(virConnectPtr conn, virStoragePoolObjPtr pool, const char *devpath); Index: src/storage_backend_logical.c =================================================================== RCS file: /data/cvs/libvirt/src/storage_backend_logical.c,v retrieving revision 1.26 diff -u -r1.26 storage_backend_logical.c --- a/src/storage_backend_logical.c 17 Nov 2008 11:19:33 -0000 1.26 +++ b/src/storage_backend_logical.c 26 Nov 2008 12:40:19 -0000 @@ -470,6 +470,8 @@ }; int exitstatus; + virStorageBackendWaitForDevices(conn); + /* Get list of all logical volumes */ if (virStorageBackendLogicalFindLVs(conn, pool, NULL) < 0) { virStoragePoolObjClearVols(pool); Index: src/storage_backend_disk.c =================================================================== RCS file: /data/cvs/libvirt/src/storage_backend_disk.c,v retrieving revision 1.20 diff -u -r1.20 storage_backend_disk.c --- a/src/storage_backend_disk.c 17 Nov 2008 11:19:33 -0000 1.20 +++ b/src/storage_backend_disk.c 26 Nov 2008 12:40:19 -0000 @@ -262,6 +262,8 @@ VIR_FREE(pool->def->source.devices[0].freeExtents); pool->def->source.devices[0].nfreeExtent = 0; + virStorageBackendWaitForDevices(conn); + return virStorageBackendDiskReadPartitions(conn, pool, NULL); } Index: src/storage_backend_iscsi.c =================================================================== RCS file: /data/cvs/libvirt/src/storage_backend_iscsi.c,v retrieving revision 1.18 diff -u -r1.18 storage_backend_iscsi.c --- a/src/storage_backend_iscsi.c 17 Nov 2008 11:19:33 -0000 1.18 +++ b/src/storage_backend_iscsi.c 26 Nov 2008 12:40:19 -0000 @@ -603,6 +603,8 @@ pool->def->allocation = pool->def->capacity = pool->def->available = 0; + virStorageBackendWaitForDevices(conn); + if ((session = virStorageBackendISCSISession(conn, pool)) == NULL) goto cleanup; if (virStorageBackendISCSIRescanLUNs(conn, pool, session) < 0) Index: configure.in =================================================================== RCS file: /data/cvs/libvirt/configure.in,v retrieving revision 1.187 diff -u -r1.187 configure.in --- a/configure.in 25 Nov 2008 15:48:11 -0000 1.187 +++ b/configure.in 26 Nov 2008 12:40:19 -0000 @@ -115,11 +115,15 @@ [/sbin:/usr/sbin:/usr/local/sbin:$PATH]) AC_PATH_PROG([BRCTL], [brctl], [brctl], [/sbin:/usr/sbin:/usr/local/sbin:$PATH]) +AC_PATH_PROG([UDEVSETTLE], [udevsettle], [udevsettle], + [/sbin:/usr/sbin:/usr/local/sbin:$PATH]) AC_DEFINE_UNQUOTED([DNSMASQ],["$DNSMASQ"], [Location or name of the dnsmasq program]) AC_DEFINE_UNQUOTED([BRCTL],["$BRCTL"], [Location or name of the brctl program (see bridge-utils)]) +AC_DEFINE_UNQUOTED([UDEVSETTLE],["$UDEVSETTLE"], + [Location or name of the udevsettle program (see udev)]) dnl Specific dir for HTML output ? AC_ARG_WITH([html-dir], [AC_HELP_STRING([--with-html-dir=path],

5 13

[libvirt] Concerning umlInotifyEvent
by Ron Yorston 02 Dec '08

02 Dec '08

At the top of umlInotifyEvent in uml_driver.c there is this test: if (watch != driver->inotifyWatch) return; This doesn't seem to be correct. I have to comment out the test to get libvirtd to work with UML. Ron

2 3

[libvirt] libvirt 0.5.0 and KVM migration
by Mickaël Canévet 02 Dec '08

02 Dec '08

Hi, I just installed libvirt 0.5.0 on Debian Lenny with KVM 0.72 to try KVM migration support. When I try to migrate a VM from one node to the other one, I have this error message: libvir: error : this function is not supported by the hypervisor: virDomainMigrate Is it really supposed to work ? Do I have to upgrade KVM to a newer version ? Thank you for your answers and congratulation for your precious work. -- _______________________________________________________________________ Mickaël Canévet. European Molecular Biology Laboratory (EMBL) Grenoble Outstation. FRANCE _______________________________________________________________________

3 6