Compare commits

..

8 Commits

Author SHA1 Message Date
Laine Stump 40080a09a5 make commandtest more robust to fix broken build 2011-07-06 11:10:52 -04:00
Laine Stump 05aa1933ee make commandtest more robust 2011-07-06 11:08:36 -04:00
Peter Robinson 6051cec412 add ARM to NUMA platform exlcludes 2011-07-06 15:45:06 +01:00
Laine Stump 84c34151e4 update to libvirt-0.8.8-5
Fix for CVE-2011-2178, regression introduced in disk probe logic,
Bug 709775

Fix for CVE-2011-2511, integer overflow in VirDomainGetVcpus,
Bug 717204

Add several build and runtime dependencies to specfile
Bug 680270
2011-07-05 18:36:14 -04:00
Laine Stump 2915aa73af Fix for CVE-2011-1486, error reporting in libvirtd is not thread safe, bug 693457 2011-04-05 15:51:36 -04:00
Daniel Veillard dedb223721 Fix for CVE-2011-1146, missing checks on read-only connections 2011-03-15 09:43:40 +08:00
Daniel P. Berrange 2243a44969 Add patch to fix -kernel boot with latest QEMU 2011-02-21 14:35:08 +00:00
Daniel Veillard 95a9c60969 Release of libvirt-0.8.8 upstream
- expose new API for sysinfo extraction
- cgroup blkio weight support
- smartcard device support
- qemu: Support per-device boot ordering
- Various improvements and bug fixes
Daniel
2011-02-17 12:24:52 -05:00
23 changed files with 3124 additions and 2966 deletions
-31
View File
@@ -1,31 +0,0 @@
[suppress_function]
symbol_version_regexp = LIBVIRT_PRIVATE.*
soname_regexp = libvirt\\.so.*
[suppress_function]
symbol_version_regexp = LIBVIRT_ADMIN_PRIVATE.*
soname_regexp = libvirt-admin\\.so.*
[suppress_variable]
symbol_version_regexp = LIBVIRT_PRIVATE.*
soname_regexp = libvirt\\.so.*
[suppress_variable]
symbol_version_regexp = LIBVIRT_ADMIN_PRIVATE.*
soname_regexp = libvirt-admin\\.so.*
[suppress_function]
symbol_version_regexp = .*
soname_regexp = libvirt_storage_.*\\.so.*
[suppress_variable]
symbol_version_regexp = .*
soname_regexp = libvirt_storage_.*\\.so.*
[suppress_function]
symbol_version_regexp = .*
soname_regexp = libvirt_driver_.*\\.so.*
[suppress_variable]
symbol_version_regexp = .*
soname_regexp = libvirt_driver_.*\\.so.*
+4 -1
View File
@@ -2,4 +2,7 @@
*.rpm
i686
x86_64
libvirt-*.tar.xz
libvirt-*.tar.gz
/libvirt-0.8.4.tar.gz
/libvirt-0.8.5.tar.gz
/libvirt-0.8.7.tar.gz
@@ -1,41 +0,0 @@
From 3499354e12a1c1832bf4030693a64e03ceb79d05 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Wed, 5 Jun 2024 11:16:21 +0100
Subject: [PATCH] interface: fix udev reference leak with invalid flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
The udevInterfaceGetXMLDesc method takes a reference on the udev
driver as its first action. If the virCheckFlags() condition
fails, however, this reference is never released.
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
src/interface/interface_backend_udev.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/interface/interface_backend_udev.c b/src/interface/interface_backend_udev.c
index fdf11a8318..e1a50389c9 100644
--- a/src/interface/interface_backend_udev.c
+++ b/src/interface/interface_backend_udev.c
@@ -1027,12 +1027,14 @@ static char *
udevInterfaceGetXMLDesc(virInterfacePtr ifinfo,
unsigned int flags)
{
- struct udev *udev = udev_ref(driver->udev);
+ struct udev *udev = NULL;
g_autoptr(virInterfaceDef) ifacedef = NULL;
char *xmlstr = NULL;
virCheckFlags(VIR_INTERFACE_XML_INACTIVE, NULL);
+ udev = udev_ref(driver->udev);
+
/* Recursively build up the interface XML based on the requested
* interface name
*/
--
2.45.1
@@ -1,49 +0,0 @@
From 98f1cf88fa7e0f992d93f376418fbfb3996a9690 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Fri, 17 May 2024 14:55:24 +0100
Subject: [PATCH] rpc: avoid leak of GSource in use for interrupting main loop
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
We never release the reference on the GSource created for
interrupting the main loop, nor do we remove it from the
main context if our thread is woken up prior to the wakeup
callback firing.
This can result in a leak of GSource objects, along with an
ever growing list of GSources attached to the main context,
which will gradually slow down execution of the loop, as
several operations are O(N) for the number of attached GSource
objects.
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
src/rpc/virnetclient.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c
index 147b0d661a..6d424eb599 100644
--- a/src/rpc/virnetclient.c
+++ b/src/rpc/virnetclient.c
@@ -1946,7 +1946,7 @@ static int virNetClientIO(virNetClient *client,
/* Check to see if another thread is dispatching */
if (client->haveTheBuck) {
/* Force other thread to wakeup from poll */
- GSource *wakeup = g_idle_source_new();
+ g_autoptr(GSource) wakeup = g_idle_source_new();
g_source_set_callback(wakeup, virNetClientIOWakeup, client->eventLoop, NULL);
g_source_attach(wakeup, client->eventCtx);
@@ -1968,6 +1968,7 @@ static int virNetClientIO(virNetClient *client,
return -1;
}
+ g_source_destroy(wakeup);
VIR_DEBUG("Woken up from sleep head=%p call=%p",
client->waitDispatch, thiscall);
/* Three reasons we can be woken up
--
2.45.1
@@ -1,99 +0,0 @@
From 8074d64dc2eca846d6a61efe1a9b7428a0ce1dd1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Tue, 30 Apr 2024 11:51:15 +0100
Subject: [PATCH] rpc: ensure temporary GSource is removed from client event
loop
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Users are seeing periodic segfaults from libvirt client apps,
especially thread heavy ones like virt-manager. A typical
stack trace would end up in the virNetClientIOEventFD method,
with illegal access to stale stack data. eg
==238721==ERROR: AddressSanitizer: stack-use-after-return on address 0x75cd18709788 at pc 0x75cd3111f907 bp 0x75cd181ff550 sp 0x75cd181ff548
WRITE of size 4 at 0x75cd18709788 thread T11
#0 0x75cd3111f906 in virNetClientIOEventFD /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:1634:15
#1 0x75cd3210d198 (/usr/lib/libglib-2.0.so.0+0x5a198) (BuildId: 0a2311dfbbc6c215dc36f4b6bdd2b4b6fbae55a2)
#2 0x75cd3216c3be (/usr/lib/libglib-2.0.so.0+0xb93be) (BuildId: 0a2311dfbbc6c215dc36f4b6bdd2b4b6fbae55a2)
#3 0x75cd3210ddc6 in g_main_loop_run (/usr/lib/libglib-2.0.so.0+0x5adc6) (BuildId: 0a2311dfbbc6c215dc36f4b6bdd2b4b6fbae55a2)
#4 0x75cd3111a47c in virNetClientIOEventLoop /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:1722:9
#5 0x75cd3111a47c in virNetClientIO /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:2002:10
#6 0x75cd3111a47c in virNetClientSendInternal /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:2170:11
#7 0x75cd311198a8 in virNetClientSendWithReply /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:2198:11
#8 0x75cd31111653 in virNetClientProgramCall /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclientprogram.c:318:9
#9 0x75cd31241c8f in callFull /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/remote/remote_driver.c:6054:10
#10 0x75cd31241c8f in call /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/remote/remote_driver.c:6076:12
#11 0x75cd31241c8f in remoteNetworkGetXMLDesc /usr/src/debug/libvirt/libvirt-10.2.0/build/src/remote/remote_client_bodies.h:5959:9
#12 0x75cd31410ff7 in virNetworkGetXMLDesc /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/libvirt-network.c:952:15
The root cause is a bad assumption in the virNetClientIOEventLoop
method. This method is run by whichever thread currently owns the
buck, and is responsible for handling I/O. Inside a for(;;) loop,
this method creates a temporary GSource, adds it to the event loop
and runs g_main_loop_run(). When I/O is ready, the GSource callback
(virNetClientIOEventFD) will fire and call g_main_loop_quit(), and
return G_SOURCE_REMOVE which results in the temporary GSource being
destroyed. A g_autoptr() will then remove the last reference.
What was overlooked, is that a second thread can come along and
while it can't enter virNetClientIOEventLoop, it will register an
idle source that uses virNetClientIOWakeup to interrupt the
original thread's 'g_main_loop_run' call. When this happens the
virNetClientIOEventFD callback never runs, and so the temporary
GSource is not destroyed. The g_autoptr() will remove a reference,
but by virtue of still being attached to the event context, there
is an extra reference held causing GSource to be leaked. The
next time 'g_main_loop_run' is called, the original GSource will
trigger its callback, and access data that was allocated on the
stack by the previous thread, and likely SEGV.
To solve this, the thread calling 'g_main_loop_run' must call
g_source_destroy, immediately upon return, to guarantee that
the temporary GSource is removed.
CVE-2024-4418
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reported-by: Martin Shirokov <shirokovmartin@gmail.com>
Tested-by: Martin Shirokov <shirokovmartin@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
src/rpc/virnetclient.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c
index 68098b1c8d..147b0d661a 100644
--- a/src/rpc/virnetclient.c
+++ b/src/rpc/virnetclient.c
@@ -1657,7 +1657,7 @@ static int virNetClientIOEventLoop(virNetClient *client,
#endif /* !WIN32 */
int timeout = -1;
virNetMessage *msg = NULL;
- g_autoptr(GSource) G_GNUC_UNUSED source = NULL;
+ g_autoptr(GSource) source = NULL;
GIOCondition ev = 0;
struct virNetClientIOEventData data = {
.client = client,
@@ -1721,6 +1721,18 @@ static int virNetClientIOEventLoop(virNetClient *client,
g_main_loop_run(client->eventLoop);
+ /*
+ * If virNetClientIOEventFD ran, this GSource will already be
+ * destroyed due to G_SOURCE_REMOVE. It is harmless to re-destroy
+ * it, since we still own a reference.
+ *
+ * If virNetClientIOWakeup ran, it will have interrupted the
+ * g_main_loop_run call, before virNetClientIOEventFD could
+ * run, and thus the GSource is still registered, and we need
+ * to destroy it since it is referencing stack memory for 'data'
+ */
+ g_source_destroy(source);
+
#ifndef WIN32
ignore_value(pthread_sigmask(SIG_SETMASK, &oldmask, NULL));
#endif /* !WIN32 */
--
2.45.1
@@ -1,35 +0,0 @@
From adfdb79f1e01401349e1321d0f5059d7b6489f00 Mon Sep 17 00:00:00 2001
Message-ID: <adfdb79f1e01401349e1321d0f5059d7b6489f00.1724763718.git.crobinso@redhat.com>
From: Martin Kletzander <mkletzan@redhat.com>
Date: Fri, 16 Aug 2024 13:56:51 +0200
Subject: [PATCH 1/3] virarptable: Properly calculate rtattr length
Content-type: text/plain
Use convenience macro which does almost the same thing we were doing,
but also pads out the payload length to a multiple of NLMSG_ALIGNTO (4)
bytes.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
Signed-off-by: Cole Robinson <crobinso@redhat.com>
---
src/util/virarptable.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/util/virarptable.c b/src/util/virarptable.c
index 299dddd664..d8e41c5a86 100644
--- a/src/util/virarptable.c
+++ b/src/util/virarptable.c
@@ -102,8 +102,7 @@ virArpTableGet(void)
return table;
VIR_WARNINGS_NO_CAST_ALIGN
- parse_rtattr(tb, NDA_MAX, NDA_RTA(r),
- nh->nlmsg_len - NLMSG_LENGTH(sizeof(*r)));
+ parse_rtattr(tb, NDA_MAX, NDA_RTA(r), NLMSG_PAYLOAD(nh, sizeof(*r)));
VIR_WARNINGS_RESET
if (tb[NDA_DST] == NULL || tb[NDA_LLADDR] == NULL)
--
2.46.0
@@ -1,42 +0,0 @@
From 137779b894858bd958ea575cec260a0559b31e48 Mon Sep 17 00:00:00 2001
Message-ID: <137779b894858bd958ea575cec260a0559b31e48.1724763718.git.crobinso@redhat.com>
In-Reply-To: <adfdb79f1e01401349e1321d0f5059d7b6489f00.1724763718.git.crobinso@redhat.com>
References: <adfdb79f1e01401349e1321d0f5059d7b6489f00.1724763718.git.crobinso@redhat.com>
From: Martin Kletzander <mkletzan@redhat.com>
Date: Fri, 16 Aug 2024 13:59:15 +0200
Subject: [PATCH 2/3] virarptable: Fix check for message length
Content-type: text/plain
The previous check was all wrong since it calculated the how long would
the netlink message be if the netlink header was the payload and then
subtracted that from the whole message length, a variable that was not
used later in the code. This check can fail if there are no additional
payloads, struct rtattr in particular, which we are parsing later,
however the RTA_OK macro would've caught that anyway.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
Signed-off-by: Cole Robinson <crobinso@redhat.com>
---
src/util/virarptable.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/util/virarptable.c b/src/util/virarptable.c
index d8e41c5a86..45ee76766f 100644
--- a/src/util/virarptable.c
+++ b/src/util/virarptable.c
@@ -81,10 +81,9 @@ virArpTableGet(void)
for (; NLMSG_OK(nh, msglen); nh = NLMSG_NEXT(nh, msglen)) {
VIR_WARNINGS_RESET
struct ndmsg *r = NLMSG_DATA(nh);
- int len = nh->nlmsg_len;
void *addr;
- if ((len -= NLMSG_LENGTH(sizeof(*nh))) < 0) {
+ if (nh->nlmsg_len < NLMSG_SPACE(sizeof(*r))) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("wrong nlmsg len"));
goto cleanup;
--
2.46.0
@@ -1,54 +0,0 @@
From df2cefb31dab2fa56e0864fbd2b8ad468dee22c0 Mon Sep 17 00:00:00 2001
Message-ID: <df2cefb31dab2fa56e0864fbd2b8ad468dee22c0.1724763718.git.crobinso@redhat.com>
In-Reply-To: <adfdb79f1e01401349e1321d0f5059d7b6489f00.1724763718.git.crobinso@redhat.com>
References: <adfdb79f1e01401349e1321d0f5059d7b6489f00.1724763718.git.crobinso@redhat.com>
From: Martin Kletzander <mkletzan@redhat.com>
Date: Fri, 16 Aug 2024 14:02:48 +0200
Subject: [PATCH 3/3] virarptable: End parsing earlier in case of NLMSG_DONE
Content-type: text/plain
Check for the last multipart message right as the first thing. The
presumption probably was that the last message might still contain a
payload we want to parse. However that cannot be true since that would
have to be a type RTM_NEWNEIGH. This was not caught because older
kernels were note sending NLMSG_DONE and probably relied on the fact
that the parsing just stops after all the messages are walked through,
which the NLMSG_OK macro successfully did.
Resolves: https://issues.redhat.com/browse/RHEL-52449
Resolves: https://bugzilla.redhat.com/2302245
Fixes: a176d67cdfaf5b8237a7e3a80d8be0e6bdf2d8fd
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
Signed-off-by: Cole Robinson <crobinso@redhat.com>
---
src/util/virarptable.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/util/virarptable.c b/src/util/virarptable.c
index 45ee76766f..20d11f97b0 100644
--- a/src/util/virarptable.c
+++ b/src/util/virarptable.c
@@ -83,6 +83,9 @@ virArpTableGet(void)
struct ndmsg *r = NLMSG_DATA(nh);
void *addr;
+ if (nh->nlmsg_type == NLMSG_DONE)
+ break;
+
if (nh->nlmsg_len < NLMSG_SPACE(sizeof(*r))) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("wrong nlmsg len"));
@@ -97,9 +100,6 @@ virArpTableGet(void)
(!(r->ndm_state == NUD_STALE || r->ndm_state == NUD_REACHABLE)))
continue;
- if (nh->nlmsg_type == NLMSG_DONE)
- return table;
-
VIR_WARNINGS_NO_CAST_ALIGN
parse_rtattr(tb, NDA_MAX, NDA_RTA(r), NLMSG_PAYLOAD(nh, sizeof(*r)));
VIR_WARNINGS_RESET
--
2.46.0
+21
View File
@@ -0,0 +1,21 @@
# Makefile for source rpm: libvirt
# $Id$
NAME := libvirt
SPECFILE = $(firstword $(wildcard *.spec))
define find-makefile-common
for d in common ../common ../../common ; do if [ -f $$d/Makefile.common ] ; then if [ -f $$d/CVS/Root -a -w $$d/Makefile.common ] ; then cd $$d ; cvs -Q update ; fi ; echo "$$d/Makefile.common" ; break ; fi ; done
endef
MAKEFILE_COMMON := $(shell $(find-makefile-common))
ifeq ($(MAKEFILE_COMMON),)
# attempt a checkout
define checkout-makefile-common
test -f CVS/Root && { cvs -Q -d $$(cat CVS/Root) checkout common && echo "common/Makefile.common" ; } || { echo "ERROR: I can't figure out how to checkout the 'common' module." ; exit -1 ; } >&2
endef
MAKEFILE_COMMON := $(shell $(checkout-makefile-common))
endif
include $(MAKEFILE_COMMON)
-31
View File
@@ -1,31 +0,0 @@
From 4c5b2e1e0d0d0cbbf8c6ed28ce77d055d5974f7f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1n=20Tomko?= <jtomko@redhat.com>
Date: Wed, 6 Mar 2024 17:26:40 +0100
Subject: [PATCH] qemu: virtiofs: set correct label when creating the socket
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Use svirt_t instead of virtd_t, since virtd_t is not available in the
session mode and qemu with svirt_t won't be able to talk to unconfined_t
socket.
Signed-off-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/qemu/qemu_virtiofs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/qemu/qemu_virtiofs.c b/src/qemu/qemu_virtiofs.c
index 15dea3bb57f..d80cddd3ba9 100644
--- a/src/qemu/qemu_virtiofs.c
+++ b/src/qemu/qemu_virtiofs.c
@@ -102,7 +102,7 @@ qemuVirtioFSOpenChardev(virQEMUDriver *driver,
chrdev->data.nix.listen = true;
chrdev->data.nix.path = g_strdup(socket_path);
- if (qemuSecuritySetDaemonSocketLabel(driver->securityManager, vm->def) < 0)
+ if (qemuSecuritySetSocketLabel(driver->securityManager, vm->def) < 0)
goto cleanup;
fd = qemuOpenChrChardevUNIXSocket(chrdev);
if (fd < 0) {
@@ -0,0 +1,40 @@
From e03899ff772cb753f02ecc99c81776a95c8e3d59 Mon Sep 17 00:00:00 2001
From: Osier Yang <jyang@redhat.com>
Date: Fri, 18 Feb 2011 13:45:13 +0800
Subject: [PATCH 2/6] Requires gettext for client package
https://bugzilla.redhat.com/show_bug.cgi?id=680270
libvirt-client is missing some dependencies
libvirt-guests invokes functions in gettext.sh, so we need to
require gettext package in spec file.
Demo with the fix:
% rpm -q gettext
package gettext is not installed
% rpm -ivh libvirt-client-0.8.8-1.fc14.x86_64.rpm
error: Failed dependencies:
gettext is needed by libvirt-client-0.8.8-1.fc14.x86_64
* libvirt.spec.in
---
libvirt.spec.in | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/libvirt.spec.in b/libvirt.spec.in
index d4208e8..c08b186 100644
--- a/libvirt.spec.in
+++ b/libvirt.spec.in
@@ -415,6 +415,8 @@ Requires: ncurses
# So remote clients can access libvirt over SSH tunnel
# (client invokes 'nc' against the UNIX socket on the server)
Requires: nc
+# Needed by libvirt-guests init script.
+Requires: gettext
%if %{with_sasl}
Requires: cyrus-sasl
# Not technically required, but makes 'out-of-box' config
--
1.7.3.4
@@ -0,0 +1,51 @@
From 2c2ae4c48c7e57fd233f1b9475fb6ecbab04804a Mon Sep 17 00:00:00 2001
From: Jiri Denemark <jdenemar@redhat.com>
Date: Fri, 25 Mar 2011 16:45:45 +0100
Subject: [PATCH 2/2] daemon: Avoid resetting errors before they are reported
https://bugzilla.redhat.com/show_bug.cgi?id=690733
Commit f44bfb7 was supposed to make sure no additional libvirt API (esp.
*Free) is called before remoteDispatchConnError() is called on error.
However, the patch missed two instances.
(cherry picked from commit 55cc591fc18e87b29febf78dc5b424b7c12f7349)
---
daemon/remote.c | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/daemon/remote.c b/daemon/remote.c
index 554e75e..159430e 100644
--- a/daemon/remote.c
+++ b/daemon/remote.c
@@ -4868,12 +4868,13 @@ remoteDispatchStoragePoolListVolumes (struct qemud_server *server ATTRIBUTE_UNUS
ret->names.names_len =
virStoragePoolListVolumes (pool,
ret->names.names_val, args->maxnames);
- virStoragePoolFree(pool);
if (ret->names.names_len == -1) {
VIR_FREE(ret->names.names_val);
remoteDispatchConnError(rerr, conn);
+ virStoragePoolFree(pool);
return -1;
}
+ virStoragePoolFree(pool);
return 0;
}
@@ -4897,11 +4898,12 @@ remoteDispatchStoragePoolNumOfVolumes (struct qemud_server *server ATTRIBUTE_UNU
}
ret->num = virStoragePoolNumOfVolumes (pool);
- virStoragePoolFree(pool);
if (ret->num == -1) {
remoteDispatchConnError(rerr, conn);
+ virStoragePoolFree(pool);
return -1;
}
+ virStoragePoolFree(pool);
return 0;
}
--
1.7.3.4
@@ -0,0 +1,30 @@
From 29680e00f67bad9145387022ea0d3c307465d3dc Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Mon, 21 Feb 2011 10:43:29 -0700
Subject: [PATCH 4/6] build: add dependency on gnutls-utils
https://bugzilla.redhat.com/show_bug.cgi?id=680270
libvirt-client is missing some dependencies
* libvirt.spec.in (Requires): Add gnutls-utils, for virt-pki-validate.
Suggested by Daniel P. Berrange.
---
libvirt.spec.in | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/libvirt.spec.in b/libvirt.spec.in
index c08b186..23f4525 100644
--- a/libvirt.spec.in
+++ b/libvirt.spec.in
@@ -417,6 +417,8 @@ Requires: ncurses
Requires: nc
# Needed by libvirt-guests init script.
Requires: gettext
+# Needed by virt-pki-validate script.
+Requires: gnutls-utils
%if %{with_sasl}
Requires: cyrus-sasl
# Not technically required, but makes 'out-of-box' config
--
1.7.3.4
+27
View File
@@ -0,0 +1,27 @@
commit efc2594b4e0cbcdd6947fafeeed41accd5b611e0
Author: Jim Fehlig <jfehlig@novell.com>
Date: Thu Feb 17 14:22:55 2011 -0700
Do not add drive 'boot=on' param when a kernel is specified
libvirt-tck was failing several domain tests [1] with qemu 0.14, which
is now less tolerable of specifying 2 bootroms with the same boot index [2].
Drop the 'boot=on' param if kernel has been specfied.
[1] https://www.redhat.com/archives/libvir-list/2011-February/msg00559.html
[2] http://lists.nongnu.org/archive/html/qemu-devel/2011-02/msg01892.html
diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c
index 371a7ed..0db2843 100644
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@@ -3173,7 +3173,7 @@ qemuBuildCommandLine(virConnectPtr conn,
int bootCD = 0, bootFloppy = 0, bootDisk = 0;
/* If QEMU supports boot=on for -drive param... */
- if (qemuCmdFlags & QEMUD_CMD_FLAG_DRIVE_BOOT) {
+ if (qemuCmdFlags & QEMUD_CMD_FLAG_DRIVE_BOOT && !def->os.kernel) {
for (i = 0 ; i < def->os.nBootDevs ; i++) {
switch (def->os.bootDevs[i]) {
case VIR_DOMAIN_BOOT_CDROM:
@@ -0,0 +1,32 @@
From 12509c09a55bd2ab171f9fa029fb94f297adc0a0 Mon Sep 17 00:00:00 2001
From: Daniel P. Berrange <berrange@redhat.com>
Date: Thu, 24 Feb 2011 12:12:27 +0000
Subject: [PATCH] Make commandtest more robust wrt its execution environment
When executed from cron, commandtest would fail to correctly
identify daemon processes. Set session ID and process group
IDs at startup to ensure we have a consistent environment to
run in.
* tests/commandtest.c: Call setsid() and setpgid()
---
tests/commandtest.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/tests/commandtest.c b/tests/commandtest.c
index 7157c51..dc2f8a1 100644
--- a/tests/commandtest.c
+++ b/tests/commandtest.c
@@ -730,6 +730,9 @@ mymain(int argc, char **argv)
if (chdir("/tmp") < 0)
return(EXIT_FAILURE);
+ setpgid(0, 0);
+ setsid();
+
/* Kill off any inherited fds that might interfere with our
* testing. */
fd = 3;
--
1.7.3.4
+95
View File
@@ -0,0 +1,95 @@
From: Guido Günther <agx@sigxcpu.org>
Date: Mon, 14 Mar 2011 02:56:28 +0000 (+0800)
Subject: Add missing checks for read only connections
X-Git-Url: http://libvirt.org/git/?p=libvirt.git;a=commitdiff_plain;h=71753cb7f7a16ff800381c0b5ee4e99eea92fed3;hp=13c00dde3171b3a38d23cceb3f9151cb6cac3dad
Add missing checks for read only connections
As pointed on CVE-2011-1146, some API forgot to check the read-only
status of the connection for entry point which modify the state
of the system or may lead to a remote execution using user data.
The entry points concerned are:
- virConnectDomainXMLToNative
- virNodeDeviceDettach
- virNodeDeviceReAttach
- virNodeDeviceReset
- virDomainRevertToSnapshot
- virDomainSnapshotDelete
* src/libvirt.c: fix the above set of entry points to error on read-only
connections
---
diff --git a/src/libvirt.c b/src/libvirt.c
index caa109d..713291f 100644
--- a/src/libvirt.c
+++ b/src/libvirt.c
@@ -3321,6 +3321,10 @@ char *virConnectDomainXMLToNative(virConnectPtr conn,
virDispatchError(NULL);
return NULL;
}
+ if (conn->flags & VIR_CONNECT_RO) {
+ virLibDomainError(VIR_ERR_OPERATION_DENIED, __FUNCTION__);
+ goto error;
+ }
if (nativeFormat == NULL || domainXml == NULL) {
virLibConnError(VIR_ERR_INVALID_ARG, __FUNCTION__);
@@ -9748,6 +9752,11 @@ virNodeDeviceDettach(virNodeDevicePtr dev)
return -1;
}
+ if (dev->conn->flags & VIR_CONNECT_RO) {
+ virLibConnError(VIR_ERR_OPERATION_DENIED, __FUNCTION__);
+ goto error;
+ }
+
if (dev->conn->driver->nodeDeviceDettach) {
int ret;
ret = dev->conn->driver->nodeDeviceDettach (dev);
@@ -9791,6 +9800,11 @@ virNodeDeviceReAttach(virNodeDevicePtr dev)
return -1;
}
+ if (dev->conn->flags & VIR_CONNECT_RO) {
+ virLibConnError(VIR_ERR_OPERATION_DENIED, __FUNCTION__);
+ goto error;
+ }
+
if (dev->conn->driver->nodeDeviceReAttach) {
int ret;
ret = dev->conn->driver->nodeDeviceReAttach (dev);
@@ -9836,6 +9850,11 @@ virNodeDeviceReset(virNodeDevicePtr dev)
return -1;
}
+ if (dev->conn->flags & VIR_CONNECT_RO) {
+ virLibConnError(VIR_ERR_OPERATION_DENIED, __FUNCTION__);
+ goto error;
+ }
+
if (dev->conn->driver->nodeDeviceReset) {
int ret;
ret = dev->conn->driver->nodeDeviceReset (dev);
@@ -13131,6 +13150,10 @@ virDomainRevertToSnapshot(virDomainSnapshotPtr snapshot,
}
conn = snapshot->domain->conn;
+ if (conn->flags & VIR_CONNECT_RO) {
+ virLibConnError(VIR_ERR_OPERATION_DENIED, __FUNCTION__);
+ goto error;
+ }
if (conn->driver->domainRevertToSnapshot) {
int ret = conn->driver->domainRevertToSnapshot(snapshot, flags);
@@ -13177,6 +13200,10 @@ virDomainSnapshotDelete(virDomainSnapshotPtr snapshot,
}
conn = snapshot->domain->conn;
+ if (conn->flags & VIR_CONNECT_RO) {
+ virLibConnError(VIR_ERR_OPERATION_DENIED, __FUNCTION__);
+ goto error;
+ }
if (conn->driver->domainSnapshotDelete) {
int ret = conn->driver->domainSnapshotDelete(snapshot, flags);
@@ -0,0 +1,115 @@
From 9388aeabcbb06ec93845b6d066148ad4cfe1dd9e Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 24 Jun 2011 12:16:05 -0600
Subject: [PATCH 6/6] remote: protect against integer overflow
https://bugzilla.redhat.com/show_bug.cgi?id=717204
CVE-2011-2511 - integer overflow in VirDomainGetVcpus
Integer overflow and remote code are never a nice mix.
This has existed since commit 56cd414.
* src/libvirt.c (virDomainGetVcpus): Reject overflow up front.
* src/remote/remote_driver.c (remoteDomainGetVcpus): Avoid overflow
on sending rpc.
* daemon/remote.c (remoteDispatchDomainGetVcpus): Avoid overflow on
receiving rpc.
(cherry picked from commit 774b21c163845170c9ffa873f5720d318812eaf6)
Conflicts:
daemon/remote.c
src/remote/remote_driver.c
Change to internal.h required to avoid backporting 89d994ad.
---
daemon/remote.c | 3 ++-
src/internal.h | 17 +++++++++++++++++
src/libvirt.c | 5 +++--
src/remote/remote_driver.c | 3 ++-
4 files changed, 24 insertions(+), 4 deletions(-)
diff --git a/daemon/remote.c b/daemon/remote.c
index 159430e..b707326 100644
--- a/daemon/remote.c
+++ b/daemon/remote.c
@@ -1722,7 +1722,8 @@ remoteDispatchDomainGetVcpus (struct qemud_server *server ATTRIBUTE_UNUSED,
return -1;
}
- if (args->maxinfo * args->maplen > REMOTE_CPUMAPS_MAX) {
+ if (INT_MULTIPLY_OVERFLOW(args->maxinfo, args->maplen) ||
+ args->maxinfo * args->maplen > REMOTE_CPUMAPS_MAX) {
virDomainFree(dom);
remoteDispatchFormatError (rerr, "%s", _("maxinfo * maplen > REMOTE_CPUMAPS_MAX"));
return -1;
diff --git a/src/internal.h b/src/internal.h
index e263684..f47b842 100644
--- a/src/internal.h
+++ b/src/internal.h
@@ -232,6 +232,23 @@
} \
} while (0)
+/* branch-specific: we don't want to update gnulib on the branch, so this
+ * backports just one required macro from newer gnulib's intprops.h.
+ * This version requires that both a and b are 'int', rather than
+ * the fully type-generic version from gnulib. */
+# define INT_MULTIPLY_OVERFLOW(a, b) \
+ ((b) < 0 \
+ ? ((a) < 0 \
+ ? (a) < INT_MAX / (b) \
+ : (b) == -1 \
+ ? 0 \
+ : INT_MIN / (b) < (a)) \
+ : (b) == 0 \
+ ? 0 \
+ : ((a) < 0 \
+ ? (a) < INT_MIN / (b) \
+ : INT_MAX / (b) < (a)))
+
/* divide value by size, rounding up */
# define VIR_DIV_UP(value, size) (((value) + (size) - 1) / (size))
diff --git a/src/libvirt.c b/src/libvirt.c
index 8c70a1f..d8ab8f8 100644
--- a/src/libvirt.c
+++ b/src/libvirt.c
@@ -40,6 +40,7 @@
#include "util.h"
#include "memory.h"
#include "configmake.h"
+#include "intprops.h"
#ifndef WITH_DRIVER_MODULES
# ifdef WITH_TEST
@@ -5363,8 +5364,8 @@ virDomainGetVcpus(virDomainPtr domain, virVcpuInfoPtr info, int maxinfo,
/* Ensure that domainGetVcpus (aka remoteDomainGetVcpus) does not
try to memcpy anything into a NULL pointer. */
- if ((cpumaps == NULL && maplen != 0)
- || (cpumaps && maplen <= 0)) {
+ if (!cpumaps ? maplen != 0
+ : (maplen <= 0 || INT_MULTIPLY_OVERFLOW(maxinfo, maplen))) {
virLibDomainError(VIR_ERR_INVALID_ARG, __FUNCTION__);
goto error;
}
diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c
index 4ca0d3b..c73452e 100644
--- a/src/remote/remote_driver.c
+++ b/src/remote/remote_driver.c
@@ -2850,7 +2850,8 @@ remoteDomainGetVcpus (virDomainPtr domain,
maxinfo, REMOTE_VCPUINFO_MAX);
goto done;
}
- if (maxinfo * maplen > REMOTE_CPUMAPS_MAX) {
+ if (INT_MULTIPLY_OVERFLOW(maxinfo, maplen) ||
+ maxinfo * maplen > REMOTE_CPUMAPS_MAX) {
remoteError(VIR_ERR_RPC,
_("vCPU map buffer length exceeds maximum: %d > %d"),
maxinfo * maplen, REMOTE_CPUMAPS_MAX);
--
1.7.3.4
@@ -0,0 +1,99 @@
From 775581ead9c0b6435e8a0dad2a6838909638e7b6 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Wed, 23 Mar 2011 10:30:49 -0600
Subject: [PATCH 5/6] rpm: add missing dependencies
manually adapted from upstream 206fc979b1656722b254e683d89b3e9fc4480c63
Among others, the missing radvd dependency showed up as:
error: Failed to start network ipv6net
error: Cannot find radvd - Possibly the package isn't installed: No such file
or directory
even when radvd was installed, because the RADVD preprocessor
symbol was missing at configure time.
* libvirt.spec.in (with_network): Add Build and BuildRequires for radvd
(BuildRequires): Add libxslt and augeas for docs and test.
(with_libvirtd): Add module-init-tools for modprobe.
(with_nwfilter): Add BuildRequires for ebtables.
---
libvirt.spec.in | 26 ++++++++++++++++++++++++--
1 files changed, 24 insertions(+), 2 deletions(-)
diff --git a/libvirt.spec.in b/libvirt.spec.in
index 23f4525..8ffb757 100644
--- a/libvirt.spec.in
+++ b/libvirt.spec.in
@@ -219,15 +219,21 @@ Requires: %{name}-client = %{version}-%{release}
# daemon is present
%if %{with_libvirtd}
Requires: bridge-utils
+# for modprobe of pci devices
+Requires: module-init-tools
+# for /sbin/ip
+Requires: iproute
%endif
%if %{with_network}
Requires: dnsmasq >= 2.41
+Requires: radvd
+%endif
+%if %{with_network} || %{with_nwfilter}
Requires: iptables
+Requires: iptables-ipv6
%endif
%if %{with_nwfilter}
Requires: ebtables
-Requires: iptables
-Requires: iptables-ipv6
%endif
# needed for device enumeration
%if %{with_hal}
@@ -295,10 +301,15 @@ BuildRequires: xmlrpc-c-devel >= 1.14.0
%endif
BuildRequires: libxml2-devel
BuildRequires: xhtml1-dtds
+BuildRequires: libxslt
BuildRequires: readline-devel
BuildRequires: ncurses-devel
BuildRequires: gettext
BuildRequires: gnutls-devel
+%if 0%{?fedora} >= 12 || 0%{?rhel} >= 6
+# for augparse, optionally used in testing
+BuildRequires: augeas
+%endif
%if %{with_hal}
BuildRequires: hal-devel
%endif
@@ -323,8 +334,15 @@ BuildRequires: libselinux-devel
%endif
%if %{with_network}
BuildRequires: dnsmasq >= 2.41
+BuildRequires: iptables
+BuildRequires: iptables-ipv6
+BuildRequires: radvd
+%endif
+%if %{with_nwfilter}
+BuildRequires: ebtables
%endif
BuildRequires: bridge-utils
+BuildRequires: module-init-tools
%if %{with_sasl}
BuildRequires: cyrus-sasl-devel
%endif
@@ -388,7 +406,11 @@ BuildRequires: libssh2-devel
BuildRequires: netcf-devel >= 0.1.4
%endif
%if %{with_esx}
+%if 0%{?fedora} >= 9 || 0%{?rhel} >= 6
BuildRequires: libcurl-devel
+%else
+BuildRequires: curl-devel
+%endif
%endif
%if %{with_audit}
BuildRequires: audit-libs-devel
--
1.7.3.4
@@ -0,0 +1,40 @@
From c2d77ade37ee917ca258cb24ffb130fc07bb95b4 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Thu, 26 May 2011 08:18:46 -0600
Subject: [PATCH 1/6] security: plug regression introduced in disk probe logic
This patch resolves:
https://bugzilla.redhat.com/show_bug.cgi?id=709775
CVE-2011-2178 - regression introduced in disk probe logic
Regression introduced in commit d6623003 (v0.8.8) - using the
wrong sizeof operand meant that security manager private data
was overlaying the allowDiskFormatProbing member of struct
_virSecurityManager. This reopens disk probing, which was
supposed to be prevented by the solution to CVE-2010-2238.
* src/security/security_manager.c
(virSecurityManagerGetPrivateData): Use correct offset.
---
src/security/security_manager.c | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/src/security/security_manager.c b/src/security/security_manager.c
index 0246dd8..6f0becd 100644
--- a/src/security/security_manager.c
+++ b/src/security/security_manager.c
@@ -107,7 +107,9 @@ virSecurityManagerPtr virSecurityManagerNew(const char *name,
void *virSecurityManagerGetPrivateData(virSecurityManagerPtr mgr)
{
- return ((char*)mgr) + sizeof(mgr);
+ /* This accesses the memory just beyond mgr, which was allocated
+ * via VIR_ALLOC_VAR earlier. */
+ return mgr + 1;
}
--
1.7.3.4
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,27 @@
From 9679cde15cabf95c7538c3b6929893ec68552d23 Mon Sep 17 00:00:00 2001
From: Dan Kenigsberg <danken@redhat.com>
Date: Sun, 20 Feb 2011 22:29:25 +0200
Subject: [PATCH 3/6] virt-pki-validate: behave when CERTTOOL is missing
https://bugzilla.redhat.com/show_bug.cgi?id=680270
libvirt-client is missing some dependencies
---
tools/virt-pki-validate.in | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/tools/virt-pki-validate.in b/tools/virt-pki-validate.in
index 207fa76..96659cf 100755
--- a/tools/virt-pki-validate.in
+++ b/tools/virt-pki-validate.in
@@ -14,7 +14,7 @@ PORT=16514
# First get certtool
#
CERTOOL=`which certtool 2>/dev/null`
-if [ ! -x $CERTOOL ]
+if [ ! -x "$CERTOOL" ]
then
echo "Could not locate the certtool program"
echo "make sure the gnutls-utils (or gnutls-bin) package is installed"
--
1.7.3.4
+1349 -2582
View File
File diff suppressed because it is too large Load Diff
+1 -1
View File
@@ -1 +1 @@
SHA512 (libvirt-10.1.0.tar.xz) = 08e73ae15de5681430b62db85ec9901242dca5e9a4ca9685614f4a67092c6e28f27f9187144b3ceb18ad6b40e6eb1a90b1a4b056b0888724d04a62002ee2bc48
ac9235576352b84b8cb17df7456bbdfc libvirt-0.8.8.tar.gz