[libvirt] [PATCH]: Secure migration support for KVM
by Chris Lalancette
All,
Attached is the secure migration patch for libvirt. What this patch
implements is a new remote RPC call for secure migration. On the source of the
migration, we do a migration from the qemu process to the libvirtd on localhost.
As each read() in libvirtd completes, it issues an RPC message to the remote
libvirtd, using the standard libvirt RPC mechanisms. On the destination, we do
essentially the mirror; the libvirtd accepts the data from RPC, and then writes
it to a qemu container process listening on localhost.
In order to actually use this, the command-line is pretty complex. If you want
to use standard live migration, the command-line looks something like:
# virsh -c qemu+tls://source.example.org/system migrate --live guest
qemu+tls://dest.example.org/system
This says to a live migration of "guest" from "source.example.org" to
"dest.example.org", connecting to each of the remote libvirtd via TLS. Note
that in this model, the virsh process connects to the remote libvirtd's via the
-c argument (source) and the destination argument (dest).
To do secure live migration, this becomes:
# virsh -c qemu+tls://source.example.org/system migrate --live --secure guest
qemu+tls://dest.example.org/system qemu+tls://dest.example.org/system
This says to do a secure live migration of "guest" from "source.example.org" to
"dest.example.org". The virsh process figures out how to connect to the remote
libvirtd's via the -c argument (source) and the destination argument (dest).
The second "qemu+tls://dest.example.org/system" is the secure channel from the
source libvirtd to the destination libvirtd.
Signed-off-by: Chris Lalancette <clalance(a)redhat.com>
docs/apibuild.py | 1
include/libvirt/libvirt.h | 1
include/libvirt/libvirt.h.in | 1
qemud/remote.c | 34 +++
qemud/remote_dispatch_args.h | 1
qemud/remote_dispatch_prototypes.h | 7
qemud/remote_dispatch_table.h | 5
qemud/remote_protocol.c | 13 +
qemud/remote_protocol.h | 17 +
qemud/remote_protocol.x | 12 +
src/driver.h | 10 +
src/libvirt.c | 55 ++++-
src/libvirt_internal.h | 5
src/libvirt_private.syms | 1
src/lxc_driver.c | 1
src/openvz_driver.c | 1
src/qemu_driver.c | 367 ++++++++++++++++++++++++++++++++-----
src/remote_internal.c | 33 +++
src/test.c | 1
src/uml_driver.c | 1
src/virsh.c | 4
src/xen_unified.c | 1
22 files changed, 518 insertions(+), 54 deletions(-)
15 years, 7 months
[libvirt] Hyper-V driver advise request
by Matthias Bolte
Hello,
I mentioned Hyper-V in my last mail. We are currently investigating
Hyper-V support for libvirt. Basically there seem to be two options:
- use the DCE/RPC library from the Samba project for Hyper-V's WMI protocol
- write a Windows based Hyper-V driver and use it from a Windows based
libvirt client only or port libvirtd to Windows
We're currently not sure which way to go and if any of this will
actually work. So, we just want to ask for advise.
15 years, 7 months
[libvirt] Incomplete cleanup in storage backend for physical disks?
by Henrik Persson E
Hi all,
I have some questions related to the storage backend for physical disks.
If I try to create a new storage pool connected to a physical disk that
doesn't have a partition table I get the following error messages:
error: Failed to create pool from /root/physical_disk_pool.xml
error: internal error non-zero exit status from command 2
However, the pool is still exists in inactive state. If I check the code
some part of the pool is freed:
cleanup:
virStoragePoolDefFree(def);
if (pool) {
virStoragePoolObjUnlock(pool);
}
storageDriverUnlock(driver);
return ret;
but not the pool object itself. It feels like I end up with an
non-complete pool. What is the story here, shouldn't the pool object
also be freed in this case?
And what is the suggested to create and start up a storage pool
connected to a physical disk that doesn't have a partition table yet?
Regards,
Henrik
15 years, 7 months
[libvirt] Works: libvirt client on OS X 10.5.6
by Schley Andrew Kutz
Getting libvirt-0.6.3 (client) to compile on OS X
- Use MacPorts to install gnutls (and its several dependencies)
- Set environment variables:
export LDFLAGS="-L/opt/local/lib"
export CPPFLAGS="-I/opt/local/include"
export MACOSX_DEPLOYMENT_TARGET=10.4
- Configure
--prefix=/opt/libvirt/ --without-sasl --without-avahi --without-polkit
--without-python --without-xen --without-qemu --without-lxc --without-
openvz --without-libvirtd --without-uml
- Apply patches
src/pci.c
#ifndef MODPROBE
#define MODPROBE 0
#endif
src/virsh.c:5665
if (command_ret != 0 /* WEXITSTATUS (0) */) {
- Compile
The MACOSX_DEPLOYMENT_TARGET variable is very important, otherwise you
will get symbol errors when linking.
--
-a
"Ideally, a code library must be immediately usable by naive
developers, easily customized by more sophisticated developers, and
readily extensible by experts." -- L. Stein
15 years, 7 months
[libvirt] [PATCH]: Add a flag to let live migration be persistent
by Chris Lalancette
As pointed out by Łukasz Mierzwa <l.mierzwa(a)gmail.com>, it would be nice if
there was an option to automatically make a domain persistent on the destination
during a live migration. The attached patch adds this simple capability. Note
that this has to be applied on top of my previous secure migration patch,
otherwise you'll have conflicts.
Signed-off-by: Chris Lalancette <clalance(a)redhat.com>
15 years, 7 months
[libvirt] PATCH: Test the QEMU argv -> XML convertor
by Daniel P. Berrange
This patch adds a test case which validates the argv -> XML
convertor for QEMU. This re-uses the existing XML data files
from the XML -> argv test. This has a few complications
though because we can't roundtrip all the data. So I have a
couple really seriously evil regexes which splice out the
XML lines we know won't match - name, uuid, memory and
disk sharing modes. This isn't ideal, but I figure it is
better todo this and get coverage over 90% of the code, than
to have no test coverage. Thanks to Cole for this testsuite
suggestion
Daniel
diff -r 453b896e0e3f .hgignore
--- a/.hgignore Tue May 19 18:43:32 2009 +0100
+++ b/.hgignore Tue May 19 18:43:50 2009 +0100
@@ -270,6 +270,7 @@ tests/object-locking
tests/object-locking-files.txt
tests/object-locking.cmi
tests/object-locking.cmx
+tests/qemuargv2xmltest
tests/qemuxml2argvtest
tests/qemuxml2xmltest
tests/qparamtest
diff -r 453b896e0e3f tests/.cvsignore
--- a/tests/.cvsignore Tue May 19 18:43:32 2009 +0100
+++ b/tests/.cvsignore Tue May 19 18:43:50 2009 +0100
@@ -11,6 +11,7 @@ xmconfigtest
xencapstest
qemuxml2xmltest
qemuxml2argvtest
+qemuargv2xmltest
nodedevxml2xmltest
nodeinfotest
statstest
diff -r 453b896e0e3f tests/.gitignore
--- a/tests/.gitignore Tue May 19 18:43:32 2009 +0100
+++ b/tests/.gitignore Tue May 19 18:43:50 2009 +0100
@@ -11,6 +11,7 @@ xmconfigtest
xencapstest
qemuxml2xmltest
qemuxml2argvtest
+qemuargv2xmltest
nodedevxml2xmltest
nodeinfotest
statstest
diff -r 453b896e0e3f tests/Makefile.am
--- a/tests/Makefile.am Tue May 19 18:43:32 2009 +0100
+++ b/tests/Makefile.am Tue May 19 18:43:50 2009 +0100
@@ -61,7 +61,7 @@ noinst_PROGRAMS += xml2sexprtest sexpr2x
reconnect xmconfigtest xencapstest
endif
if WITH_QEMU
-noinst_PROGRAMS += qemuxml2argvtest qemuxml2xmltest
+noinst_PROGRAMS += qemuxml2argvtest qemuxml2xmltest qemuargv2xmltest
endif
if WITH_SECDRIVER_SELINUX
@@ -119,7 +119,7 @@ endif
endif
if WITH_QEMU
-TESTS += qemuxml2argvtest qemuxml2xmltest
+TESTS += qemuxml2argvtest qemuxml2xmltest qemuargv2xmltest
endif
if WITH_SECDRIVER_SELINUX
@@ -194,8 +194,13 @@ qemuxml2xmltest_SOURCES = \
qemuxml2xmltest.c testutilsqemu.c testutilsqemu.h \
testutils.c testutils.h
qemuxml2xmltest_LDADD = ../src/libvirt_driver_qemu.la $(LDADDS)
+
+qemuargv2xmltest_SOURCES = \
+ qemuargv2xmltest.c testutilsqemu.c testutilsqemu.h \
+ testutils.c testutils.h
+qemuargv2xmltest_LDADD = ../src/libvirt_driver_qemu.la $(LDADDS)
else
-EXTRA_DIST += qemuxml2argvtest.c qemuxml2xmltest.c testutilsqemu.c testutilsqemu.h
+EXTRA_DIST += qemuxml2argvtest.c qemuxml2xmltest.c qemuargv2xmltest.c testutilsqemu.c testutilsqemu.h
endif
nodedevxml2xmltest_SOURCES = \
diff -r 453b896e0e3f tests/qemuargv2xmltest.c
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/qemuargv2xmltest.c Tue May 19 18:43:50 2009 +0100
@@ -0,0 +1,235 @@
+#include <config.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+
+#include <sys/types.h>
+#include <fcntl.h>
+
+#ifdef WITH_QEMU
+
+#include "internal.h"
+#include "testutils.h"
+#include "qemu_conf.h"
+
+#include "testutilsqemu.h"
+
+static char *progname;
+static char *abs_srcdir;
+static struct qemud_driver driver;
+
+#define MAX_FILE 4096
+
+static int blankProblemElements(char *data)
+{
+ if (virtTestClearLineRegex("<name>[[:alnum:]]+</name>", data) < 0 ||
+ virtTestClearLineRegex("<uuid>([[:alnum:]]|-)+</uuid>", data) < 0 ||
+ virtTestClearLineRegex("<memory>[[:digit:]]+</memory>", data) < 0 ||
+ virtTestClearLineRegex("<currentMemory>[[:digit:]]+</currentMemory>", data) < 0 ||
+ virtTestClearLineRegex("<readonly/>", data) < 0 ||
+ virtTestClearLineRegex("<sharable/>", data) < 0)
+ return -1;
+ return 0;
+}
+
+static int testCompareXMLToArgvFiles(const char *xml,
+ const char *cmdfile) {
+ char xmlData[MAX_FILE];
+ char cmdData[MAX_FILE];
+ char *expectxml = &(xmlData[0]);
+ char *actualxml = NULL;
+ char *cmd = &(cmdData[0]);
+ int ret = -1;
+ virDomainDefPtr vmdef = NULL;
+
+ if (virtTestLoadFile(cmdfile, &cmd, MAX_FILE) < 0)
+ goto fail;
+ if (virtTestLoadFile(xml, &expectxml, MAX_FILE) < 0)
+ goto fail;
+
+ if (!(vmdef = qemuParseCommandLineString(NULL, cmd)))
+ goto fail;
+
+ if (!(actualxml = virDomainDefFormat(NULL, vmdef, 0)))
+ goto fail;
+
+ if (blankProblemElements(expectxml) < 0 ||
+ blankProblemElements(actualxml) < 0)
+ goto fail;
+
+ if (STRNEQ(expectxml, actualxml)) {
+ virtTestDifference(stderr, expectxml, actualxml);
+ goto fail;
+ }
+
+ ret = 0;
+
+ fail:
+ free(actualxml);
+ virDomainDefFree(vmdef);
+ return ret;
+}
+
+
+struct testInfo {
+ const char *name;
+ int extraFlags;
+ const char *migrateFrom;
+};
+
+static int testCompareXMLToArgvHelper(const void *data) {
+ const struct testInfo *info = data;
+ char xml[PATH_MAX];
+ char args[PATH_MAX];
+ snprintf(xml, PATH_MAX, "%s/qemuxml2argvdata/qemuxml2argv-%s.xml",
+ abs_srcdir, info->name);
+ snprintf(args, PATH_MAX, "%s/qemuxml2argvdata/qemuxml2argv-%s.args",
+ abs_srcdir, info->name);
+ return testCompareXMLToArgvFiles(xml, args);
+}
+
+
+
+static int
+mymain(int argc, char **argv)
+{
+ int ret = 0;
+ char cwd[PATH_MAX];
+
+ progname = argv[0];
+
+ if (argc > 1) {
+ fprintf(stderr, "Usage: %s\n", progname);
+ return (EXIT_FAILURE);
+ }
+
+ abs_srcdir = getenv("abs_srcdir");
+ if (!abs_srcdir)
+ abs_srcdir = getcwd(cwd, sizeof(cwd));
+
+ if ((driver.caps = testQemuCapsInit()) == NULL)
+ return EXIT_FAILURE;
+ if((driver.stateDir = strdup("/nowhere")) == NULL)
+ return EXIT_FAILURE;
+
+#define DO_TEST_FULL(name, extraFlags, migrateFrom) \
+ do { \
+ const struct testInfo info = { name, extraFlags, migrateFrom }; \
+ if (virtTestRun("QEMU ARGV-2-XML " name, \
+ 1, testCompareXMLToArgvHelper, &info) < 0) \
+ ret = -1; \
+ } while (0)
+
+#define DO_TEST(name, extraFlags) \
+ DO_TEST_FULL(name, extraFlags, NULL)
+
+ setenv("PATH", "/bin", 1);
+ setenv("USER", "test", 1);
+ setenv("LOGNAME", "test", 1);
+ setenv("HOME", "/home/test", 1);
+ unsetenv("TMPDIR");
+ unsetenv("LD_PRELOAD");
+ unsetenv("LD_LIBRARY_PATH");
+
+ /* Can't roundtrip vcpu cpuset attribute */
+ /*DO_TEST("minimal", QEMUD_CMD_FLAG_NAME);*/
+ DO_TEST("boot-cdrom", 0);
+ DO_TEST("boot-network", 0);
+ DO_TEST("boot-floppy", 0);
+ /* Can't roundtrip xenner arch */
+ /*DO_TEST("bootloader", 0);*/
+ DO_TEST("clock-utc", 0);
+ DO_TEST("clock-localtime", 0);
+ DO_TEST("disk-cdrom", 0);
+ DO_TEST("disk-cdrom-empty", QEMUD_CMD_FLAG_DRIVE);
+ DO_TEST("disk-floppy", 0);
+ DO_TEST("disk-many", 0);
+ DO_TEST("disk-virtio", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_BOOT);
+ DO_TEST("disk-xenvbd", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_BOOT);
+ DO_TEST("disk-drive-boot-disk", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_BOOT);
+ DO_TEST("disk-drive-boot-cdrom", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_BOOT);
+ DO_TEST("disk-drive-fmt-qcow", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_BOOT);
+ /* Can't roundtrip shareable+cache mode option */
+ /*DO_TEST("disk-drive-shared", QEMUD_CMD_FLAG_DRIVE);*/
+ /* Can't roundtrip v1 writethrough option */
+ /*DO_TEST("disk-drive-cache-v1-wt", QEMUD_CMD_FLAG_DRIVE);*/
+ DO_TEST("disk-drive-cache-v1-wb", QEMUD_CMD_FLAG_DRIVE);
+ DO_TEST("disk-drive-cache-v1-none", QEMUD_CMD_FLAG_DRIVE);
+ DO_TEST("disk-drive-cache-v2-wt", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_CACHE_V2);
+ DO_TEST("disk-drive-cache-v2-wb", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_CACHE_V2);
+ DO_TEST("disk-drive-cache-v2-none", QEMUD_CMD_FLAG_DRIVE |
+ QEMUD_CMD_FLAG_DRIVE_CACHE_V2);
+ DO_TEST("disk-usb", 0);
+ DO_TEST("graphics-vnc", 0);
+
+ driver.vncSASL = 1;
+ driver.vncSASLdir = strdup("/root/.sasl2");
+ DO_TEST("graphics-vnc-sasl", 0);
+ driver.vncTLS = 1;
+ driver.vncTLSx509verify = 1;
+ driver.vncTLSx509certdir = strdup("/etc/pki/tls/qemu");
+ DO_TEST("graphics-vnc-tls", 0);
+ driver.vncSASL = driver.vncTLSx509verify = driver.vncTLS = 0;
+ free(driver.vncSASLdir);
+ free(driver.vncTLSx509certdir);
+ driver.vncSASLdir = driver.vncTLSx509certdir = NULL;
+
+ DO_TEST("graphics-sdl", 0);
+ DO_TEST("graphics-sdl-fullscreen", 0);
+ DO_TEST("input-usbmouse", 0);
+ DO_TEST("input-usbtablet", 0);
+ /* Can't rountrip xenner arch */
+ /*DO_TEST("input-xen", 0);*/
+ DO_TEST("misc-acpi", 0);
+ DO_TEST("misc-no-reboot", 0);
+ DO_TEST("misc-uuid", QEMUD_CMD_FLAG_NAME |
+ QEMUD_CMD_FLAG_UUID | QEMUD_CMD_FLAG_DOMID);
+ DO_TEST("net-user", 0);
+ DO_TEST("net-virtio", 0);
+ DO_TEST("net-eth", 0);
+ DO_TEST("net-eth-ifname", 0);
+
+ DO_TEST("serial-vc", 0);
+ DO_TEST("serial-pty", 0);
+ DO_TEST("serial-dev", 0);
+ DO_TEST("serial-file", 0);
+ DO_TEST("serial-unix", 0);
+ DO_TEST("serial-tcp", 0);
+ DO_TEST("serial-udp", 0);
+ DO_TEST("serial-tcp-telnet", 0);
+ DO_TEST("serial-many", 0);
+ DO_TEST("parallel-tcp", 0);
+ DO_TEST("console-compat", 0);
+ DO_TEST("sound", 0);
+
+ DO_TEST("hostdev-usb-product", 0);
+ DO_TEST("hostdev-usb-address", 0);
+
+ DO_TEST("hostdev-pci-address", 0);
+
+ DO_TEST_FULL("restore-v1", QEMUD_CMD_FLAG_MIGRATE_KVM_STDIO, "stdio");
+ DO_TEST_FULL("restore-v2", QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC, "stdio");
+ DO_TEST_FULL("restore-v2", QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC, "exec:cat");
+ DO_TEST_FULL("migrate", QEMUD_CMD_FLAG_MIGRATE_QEMU_TCP, "tcp:10.0.0.1:5000");
+
+ virCapabilitiesFree(driver.caps);
+
+ return(ret==0 ? EXIT_SUCCESS : EXIT_FAILURE);
+}
+
+VIRT_TEST_MAIN(mymain)
+
+#else
+
+int main (void) { return (77); /* means 'test skipped' for automake */ }
+
+#endif /* WITH_QEMU */
diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-bootloader.args
--- a/tests/qemuxml2argvdata/qemuxml2argv-bootloader.args Tue May 19 18:43:32 2009 +0100
+++ b/tests/qemuxml2argvdata/qemuxml2argv-bootloader.args Tue May 19 18:43:50 2009 +0100
@@ -1,1 +1,1 @@
-LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu-kvm -S -M xenner -m 214 -smp 1 -nographic -monitor pty -no-acpi -bootloader /usr/bin/pygrub -cdrom /dev/cdrom -net none -serial none -parallel none -usb
+LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/xenner -S -M xenner -m 214 -smp 1 -nographic -monitor pty -no-acpi -bootloader /usr/bin/pygrub -cdrom /dev/cdrom -net none -serial none -parallel none -usb
diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-bootloader.xml
--- a/tests/qemuxml2argvdata/qemuxml2argv-bootloader.xml Tue May 19 18:43:32 2009 +0100
+++ b/tests/qemuxml2argvdata/qemuxml2argv-bootloader.xml Tue May 19 18:43:50 2009 +0100
@@ -13,7 +13,7 @@
<on_reboot>restart</on_reboot>
<on_crash>destroy</on_crash>
<devices>
- <emulator>/usr/bin/qemu-kvm</emulator>
+ <emulator>/usr/bin/xenner</emulator>
<disk type='block' device='cdrom'>
<source dev='/dev/cdrom'/>
<target dev='hdc' bus='ide'/>
diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-disk-cdrom-empty.xml
--- a/tests/qemuxml2argvdata/qemuxml2argv-disk-cdrom-empty.xml Tue May 19 18:43:32 2009 +0100
+++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-cdrom-empty.xml Tue May 19 18:43:50 2009 +0100
@@ -24,4 +24,3 @@
</disk>
</devices>
</domain>
-
diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml
--- a/tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml Tue May 19 18:43:32 2009 +0100
+++ b/tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml Tue May 19 18:43:50 2009 +0100
@@ -18,7 +18,7 @@
<source dev='/dev/HostVG/QEMUGuest2'/>
<target dev='hda' bus='ide'/>
</disk>
- <hostdev mode='subsystem' type='pci' managed='no'>
+ <hostdev mode='subsystem' type='pci' managed='yes'>
<source>
<address domain='0x0000' bus='0x06' slot='0x12' function='0x5'/>
</source>
diff -r 453b896e0e3f tests/testutils.c
--- a/tests/testutils.c Tue May 19 18:43:32 2009 +0100
+++ b/tests/testutils.c Tue May 19 18:43:50 2009 +0100
@@ -18,6 +18,7 @@
#ifndef WIN32
#include <sys/wait.h>
#endif
+#include <regex.h>
#include <unistd.h>
#include <string.h>
#include <fcntl.h>
@@ -457,3 +458,48 @@ cleanup:
virResetLastError();
return ret;
}
+
+
+int virtTestClearLineRegex(const char *pattern,
+ char *str)
+{
+ regex_t reg;
+ char *lineStart = str;
+ char *lineEnd = strchr(str, '\n');
+
+ if (regcomp(®, pattern, REG_EXTENDED | REG_NOSUB) != 0)
+ return -1;
+
+ while (lineStart) {
+ int ret;
+ if (lineEnd)
+ *lineEnd = '\0';
+
+
+ ret = regexec(®, lineStart, 0, NULL, 0);
+ //fprintf(stderr, "Match %d '%s' '%s'\n", ret, lineStart, pattern);
+ if (ret == 0) {
+ if (lineEnd) {
+ memmove(lineStart, lineEnd + 1, strlen(lineEnd+1) + 1);
+ /* Don't update lineStart - just iterate again on this
+ location */
+ lineEnd = strchr(lineStart, '\n');
+ } else {
+ *lineStart = '\0';
+ lineStart = NULL;
+ }
+ } else {
+ if (lineEnd) {
+ *lineEnd = '\n';
+ lineStart = lineEnd + 1;
+ lineEnd = strchr(lineStart, '\n');
+ } else {
+ lineStart = NULL;
+ }
+ }
+ }
+
+ regfree(®);
+
+ return 0;
+}
diff -r 453b896e0e3f tests/testutils.h
--- a/tests/testutils.h Tue May 19 18:43:32 2009 +0100
+++ b/tests/testutils.h Tue May 19 18:43:50 2009 +0100
@@ -27,6 +27,8 @@ int virtTestCaptureProgramOutput(const c
char **buf,
int buflen);
+int virtTestClearLineRegex(const char *pattern,
+ char *string);
int virtTestDifference(FILE *stream,
const char *expect,
--
|: 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 :|
15 years, 7 months
[libvirt] kvm migration
by Łukasz Mierzwa
Hi,
can someone explain me how migration works in libvirt? I got 2 machines with
ubuntu 9.04 (libvirt 0.6.1 and kvm 84), disk images are stored on nfs share so
both machines can access them. When I run live migration (virsh migrate --live
domain uri) my domain is migrated to second host but:
1. domain is defined on both hosts after migration, it does not disappear from
original host
2. when I shutdown this domain on second host it gets undefined
so I guess that migration is only temporary but this is not what I want. When
I migrate domain to other host I want it to stay there, how can I do that?
I can't find anything about it in documentation on libvirt website, google does
not seem to know anything either.
Łukasz Mierzwa
15 years, 7 months
[libvirt] [RFC][PATCH] lxc: fix for ns cgroups subsystem
by Ryota Ozaki
Hi,
lxc does not work if ns cgroups subsystem is enabled because
of two factors; one is that ns has a special rule to create
a group[*] unlike other subsystems and the other is lxc
controller creates a new namespace for /dev/pts prior to
create a new group for a domain. Unfortunately the new
namespace breaks the rule of ns and that prevents a lxc
controller from creating a new group.
This patch addresses the problem by creating a new group
before creating a new namespace (i.e. call unshare syscall).
Note that this patch is only for the case ns is enabled and
current code works well if it disabled. However, I think
this patch makes sense because not just a few users know
much about cgroups and likely to enable all of subsystems
without notions (i.e. mount cgroups without any options).
[*] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f...
Thanks,
ozaki-r
Signed-off-by: Ryota Ozaki <ozaki.ryota(a)gmail.com>
>From 46531182708dc3eb132b14ce2f23fbc639430176 Mon Sep 17 00:00:00 2001
From: Ryota Ozaki <ozaki.ryota(a)gmail.com>
Date: Fri, 8 May 2009 05:31:03 +0900
Subject: [PATCH] lxc: fix for ns cgroups subsystem
lxc does not work if ns cgroups subsystem is enabled because
of two factors; one is that ns has a special rule to create
a group[*] unlike other subsystems and the other is lxc
controller creates a new namespace for /dev/pts prior to
create a new group for a domain. Unfortunately the new
namespace breaks the rule of ns and that prevents a lxc
controller from creating a new group.
This patch addresses the problem by creating a new group
before creating a new namespace (i.e. call unshare syscall).
Note that this patch is only for the case ns is enabled and
current code works well if it disabled. However, I think
this patch makes sense because not just a few users know
much about cgroups and likely to enable all of subsystems
without notions (i.e. mount cgroups without any options).
[*] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f...
---
src/lxc_controller.c | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/lxc_controller.c b/src/lxc_controller.c
index e0fb05d..1231817 100644
--- a/src/lxc_controller.c
+++ b/src/lxc_controller.c
@@ -458,6 +458,9 @@ lxcControllerRun(virDomainDefPtr def,
goto cleanup;
}
+ if (lxcSetContainerResources(def) < 0)
+ goto cleanup;
+
root = virDomainGetRootFilesystem(def);
/*
@@ -543,9 +546,6 @@ lxcControllerRun(virDomainDefPtr def,
}
- if (lxcSetContainerResources(def) < 0)
- goto cleanup;
-
if ((container = lxcContainerStart(def,
nveths,
veths,
--
1.6.0.6
15 years, 7 months
