July 2009 - Devel - libvirt List Archives

[libvirt] turn off dnsmasq
by Mike Power 08 Jul '09

08 Jul '09

I have a series of kvm vms I am running via libvirt. I have bind9 and dhcp3-server installed and I am happy with their configuration. The problem I have is that dhcp3-server and dnsmasq are currently fighting over who hands out the ip address. My vms work when they get their ip from dhcp3-server, and they fail when dnsmasq hands out the ip address. Unfortunately I can not get dnsmasq to stop handing out ip addresses. This is the configuration of my default network: <network> <name>default</name> <bridge name="virbr%d" /> <forward/> <ip address="192.168.122.1" netmask="255.255.255.0"> </ip> </network> Notice in the ip tag the optional dhcp tag is missing. Thus dnsmasq I thought, was supposed to leave well enough alone. It doesn't. I can turn off dhcp3-server and start up a host. What I see in the syslog is: Jul 7 08:52:21 vmhost dnsmasq[3420]: DHCPREQUEST(virbr0) 192.168.122.244 52:54:00:8f:7c:a4 Jul 7 08:52:21 vmhost dnsmasq[3420]: DHCPACK(virbr0) 192.168.122.244 52:54:00:8f:7c:a4 host1000 Then when I run ifconfig on the host1000 vm, it has an ip address that it obviously got from dnsmasq. I just need dnsmasq to stop handing out IPs, how do I do that? Mike Power

2 1

[libvirt] [PATCH 3/4] fix another failing "make distcheck" (qemuhelptest)
by Jim Meyering 08 Jul '09

08 Jul '09

This fixes another "make distcheck" (or non-srcdir) build failure: >From f9523d014a590fb881c14397f8c3869f328d023c Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 8 Jul 2009 08:35:24 +0200 Subject: [PATCH 3/4] fix another failing "make distcheck" (qemuhelptest) It failed in a non-srcdir build because those 5 sample output files were not included in the distribution tarball. Include them. * tests/Makefile.am (qemuhelpdata, EXTRA_DIST): Include these: kvm-74, kvm-86, qemu-0.10.5, qemu-0.9.1, qemu-kvm-0.10.5. --- tests/Makefile.am | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index 3d98d69..f8bde34 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -36,6 +36,13 @@ LDADDS = \ $(LIBSOCKET) \ $(COVERAGE_LDFLAGS) +qemuhelpdata = \ + kvm-74 \ + kvm-86 \ + qemu-0.10.5 \ + qemu-0.9.1 \ + qemu-kvm-0.10.5 + EXTRA_DIST = \ oomtrace.pl \ test-lib.sh \ @@ -51,7 +58,8 @@ EXTRA_DIST = \ storagevolschematest \ storagevolschemadata \ nodedevschematest \ - nodedevschemadata + nodedevschemadata \ + $(patsubst %,qemuhelpdata/%,$(qemuhelpdata)) noinst_PROGRAMS = virshtest conftest \ nodeinfotest statstest qparamtest -- 1.6.3.3.524.g8586b

3 4

[libvirt] [PATCH 4/4] generate ChangeLog from git logs into distribution tarball
by Jim Meyering 08 Jul '09

08 Jul '09

This is what I started out working on, before I noticed the "make distcheck" failures. >From 9fd0624b8cee33b8c725464d289cc5ce33ffc1c3 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 8 Jul 2009 08:54:25 +0200 Subject: [PATCH 4/4] generate ChangeLog from git logs into distribution tarball No longer maintain a version-controlled ChangeLog file, but do continue to include a ChangeLog file in distribution tarball. * Makefile.am (gen-ChangeLog): New rule. (dist-hook): Depend on it. (EXTRA_DIST): Add ChangeLog-old. * bootstrap (modules): Add gitlog-to-changelog. * ChangeLog: Remove file. Renamed to... * ChangeLog-old: ...this. New file. * autogen.sh: Touch ChangeLog, to ensure it exists. For automake. --- ChangeLog => ChangeLog-old | 2 +- Makefile.am | 15 +++++++++++++++ autogen.sh | 3 +++ bootstrap | 1 + build-aux/.gitignore | 1 + 5 files changed, 21 insertions(+), 1 deletions(-) rename ChangeLog => ChangeLog-old (99%) diff --git a/ChangeLog b/ChangeLog-old similarity index 99% rename from ChangeLog rename to ChangeLog-old index b266de6..c3b41e2 100644 --- a/ChangeLog +++ b/ChangeLog-old @@ -1207,7 +1207,7 @@ Fri Apr 24 13:17:47 BST 2009 Daniel P. Berrange <berrange(a)redhat.com> Fix network config in XM driver * src/xm_internal.c: Fix setting of bridge & script parameters in VIF config - * src/xend_internal.h, src/xend_internal.c: Export the + * src/xend_internal.h, src/xend_internal.c: Export the default bridge script name to XM driver * tests/xmconfigdata/*: Update to take account of script= setting in VIF config diff --git a/Makefile.am b/Makefile.am index c251f96..f9efff5 100644 --- a/Makefile.am +++ b/Makefile.am @@ -9,6 +9,7 @@ SUBDIRS = gnulib/lib include src qemud tools proxy docs gnulib/tests \ ACLOCAL_AMFLAGS = -I m4 -I gnulib/m4 EXTRA_DIST = \ + ChangeLog-old \ libvirt.spec libvirt.spec.in \ mingw32-libvirt.spec.in \ libvirt.pc libvirt.pc.in \ @@ -49,3 +50,17 @@ clean-cov: # disable this check distuninstallcheck: + +dist-hook: gen-ChangeLog + +# Generate the ChangeLog file (with all entries since the switch to git) +# and insert it into the directory we're about to use to create a tarball. +gen_start_date = 2009-07-04 +.PHONY: gen-ChangeLog +gen-ChangeLog: + if test -d .git; then \ + $(top_srcdir)/build-aux/gitlog-to-changelog \ + --since=$(gen_start_date) > $(distdir)/cl-t; \ + rm -f $(distdir)/ChangeLog; \ + mv $(distdir)/cl-t $(distdir)/ChangeLog; \ + fi diff --git a/autogen.sh b/autogen.sh index 788729f..c026112 100755 --- a/autogen.sh +++ b/autogen.sh @@ -54,6 +54,9 @@ if test -z "$*"; then echo "to pass any to it, please specify them on the $0 command line." fi +# Automake requires that ChangeLog exist. +touch ChangeLog + autopoint --force #rm -rf m4 libtoolize --copy --force diff --git a/bootstrap b/bootstrap index b0de3df..0754d02 100755 --- a/bootstrap +++ b/bootstrap @@ -78,6 +78,7 @@ getaddrinfo gethostname getpass gettext +gitlog-to-changelog gnumakefile inet_pton ioctl diff --git a/build-aux/.gitignore b/build-aux/.gitignore index af6c366..62188c1 100644 --- a/build-aux/.gitignore +++ b/build-aux/.gitignore @@ -10,3 +10,4 @@ mkinstalldirs mktempd /useless-if-before-free /vc-list-files +/gitlog-to-changelog -- 1.6.3.3.524.g8586b

3 3

[libvirt] netcf error message...
by Shahar Klein 08 Jul '09

08 Jul '09

folowing the instructions in netcf I'm trying to run some basic interface configuration via libvirt patched with the netcf patches I get error messages - but the interface(eth1) seem to be managed correctly - do I have an aug(what is that?) problem? virsh # iface-define /root/libvirt_working/eth1.xml error: Failed to define interface from /root/libvirt_working/eth1.xml error: internal error could not get interface XML description (netcf: unspecified error - aug_save failed) virsh # iface-list Name State MAC Address -------------------------------------------- eth0 active 00:18:fe:28:7d:82 lo active 00:00:00:00:00:00 virsh # iface-start eth1 Interface eth1 started virsh # iface-list Name State MAC Address -------------------------------------------- eth0 active 00:18:fe:28:7d:82 eth1 active 00:18:fe:28:7d:83 lo active 00:00:00:00:00:00

1 0

[libvirt] kernel summit topic - 'containers end-game'
by Serge E. Hallyn 08 Jul '09

08 Jul '09

A topic on ksummit agenda is 'containers end-game and how do we get there'. So for starters, looking just at application (and system) containers, what do the libvirt and liblxc projects want to see in kernel support that is currently missing? Are there specific things that should be done soon to make containers more useful and usable? More generally, the topic raises the question... what 'end-games' are there? A few I can think of off-hand include: 1. resource control 2. lightweight virtual servers 3. (or 2.5) unprivileged containers/jail-on-steroids (lightweight virtual servers in which you might, just maybe, almost, be able to give away a root account, at least as much as you could do so with a kvm/qemu/xen partition) 4. checkpoint, restart, and migration For each end-game, what kernel pieces do we think are missing? For instance, people seem agreed that resource control needs io control :) Containers imo need a user namespace. I think there are quite a few network namespace exploiters who require sysfs directory tagging (or some equivalent) to allow us to migrate physical devices into network namespaces. And checkpoint/restart needs... checkpoint/restart. thanks, -serge

5 14

[libvirt] PATCH: Fix LXC container capabilities
by Daniel P. Berrange 08 Jul '09

08 Jul '09

My previous change to LXC container capabilties setup has a fairly stupid bug in it. The container init process starts off with no capabilities whatsoever :-( This was caused by a bogus capng_lock() call which meant that all capabilities were cleared when the init process was exec'd. The capng_lock call sets NOROOT & NROOT_LOCKED flags in the process secure bits. This is not neccessary for the init process - we have reduced the bounding set which is sufficient for our security goals. With the capng_lock() call removed, the init process gets its permitted and effective sets filled to match the bounding set which is the desired scenario. Daniel diff --git a/src/lxc_container.c b/src/lxc_container.c index 950dd50..b3978df 100644 --- a/src/lxc_container.c +++ b/src/lxc_container.c @@ -675,16 +675,9 @@ static int lxcContainerDropCapabilities(void) _("failed to apply capabilities: %d"), ret); return -1; } - - /* Need to prevent them regaining any caps on exec */ - if ((ret = capng_lock()) < 0) { - lxcError(NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("failed to lock capabilities: %d"), ret); - return -1; - } - #else - VIR_WARN0(_("libcap-ng support not compiled in, unable to clear capabilities")); + /* Don't warn, since this ends up going to the guest console and annoying admins + VIR_WARN0(_("libcap-ng support not compiled in, unable to clear capabilities")); */ #endif return 0; } -- |: 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] Cannot get network to work with LXC
by James Yu 08 Jul '09

08 Jul '09

Hi all, I tried to use the first xml config in http://libvirt.org/drvlxc.html to setup my first light weight container. However, I was unable to get the container's network to work. When I tried to give the interface an IP, I got the following error message: sh-4.0# /sbin/ifconfig veth1 192.168.0.1/24 SIOCSIFADDR: Permission denied SIOCSIFNETMASK: Permission denied SIOCGIFADDR: Cannot assign requested address SIOCSIFBROADCAST: Permission denied SIOCSIFFLAGS: Permission denied sh-4.0# Any advise ? This is a UTF-8 formatted mail ----------------------------------------------- James C.-C.Yu

3 2

[libvirt] PATCH: Video device config
by Tom Hughes 08 Jul '09

08 Jul '09

I recently posted a patch on the RedHat bugzilla (bug #467038) to add a model attribute to the graphics element so that the different graphics cards supported by QEMU could be selected. Apparently that approach has been rejected in the past however, and the preferred approach is to add a separate video element. I've taken the patch that Daniel Berrange posted here a few months ago that added the necessary XML handling for the new element and extended it by filling in the QEMU driver code to make it use the model details as well as fixing the default video model for XEN domains as per the list of outstanding issues given with the original patch. The updated patch is attached, and based on the list of outstanding tasks given before what remains to be done is: - Add RNG XML schemas & website docs - Make Xen drivers use this info for setting stdvga=1|0 config arg - Make VirtualBox use this info in whatever way it needs Tom -- Tom Hughes (tom(a)compton.nu) http://www.compton.nu/

3 3

[libvirt] [PATCH 2/4] fix failing "make distcheck"
by Jim Meyering 08 Jul '09

08 Jul '09

Make distcheck would fail without this: >From e2f91000a33168d2aaac89f94f1b11c633c579f0 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 8 Jul 2009 08:15:54 +0200 Subject: [PATCH 2/4] fix failing "make distcheck" * docs/Makefile.am (xml): Distribute testpool.xml and testvol.xml, which are referenced from testnode.xml. --- docs/Makefile.am | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index 43d04c3..4bdb827 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -55,7 +55,9 @@ xml = \ testdomfv0.xml \ testnetdef.xml \ testnetpriv.xml \ - testnode.xml + testnode.xml \ + testpool.xml \ + testvol.xml fig = \ libvirt-net-logical.fig \ -- 1.6.3.3.524.g8586b

3 2

[libvirt] [PATCH 1/4] Don't use mylibtool, (subsumed by automake's silent rules)
by Jim Meyering 08 Jul '09

08 Jul '09

Its use was causing a non-srcdir build to fail. * Makefile.am (EXTRA_DIST): Remove mylibtool. * configure.in: Don't use mylibtool. --- An alternative (if you don't want to use automake's silent-rules option) would be to keep mylibtool, but to use $(abs_top_srcdir) rather than $(top_srcdir) in that script. Makefile.am | 3 +-- configure.in | 4 ---- 2 files changed, 1 insertions(+), 6 deletions(-) diff --git a/Makefile.am b/Makefile.am index dd98ff2..c251f96 100644 --- a/Makefile.am +++ b/Makefile.am @@ -20,8 +20,7 @@ EXTRA_DIST = \ .x-sc_require_config_h \ .x-sc_prohibit_nonreentrant \ Makefile.nonreentrant \ - autogen.sh \ - mylibtool + autogen.sh man_MANS = virsh.1 diff --git a/configure.in b/configure.in index 1170958..a6f6754 100644 --- a/configure.in +++ b/configure.in @@ -59,10 +59,6 @@ dnl Support building Win32 DLLs (must appear *before* AM_PROG_LIBTOOL) AC_LIBTOOL_WIN32_DLL AM_PROG_LIBTOOL -dnl Override normal libtool in favour of our quiet version -LIBTOOL='$(SHELL) $(top_srcdir)/mylibtool' -AC_SUBST([LIBTOOL]) - AM_PROG_CC_C_O VERSION_SCRIPT_FLAGS=-Wl,--version-script= -- 1.6.3.3.524.g8586b

2 4