Compare commits

..

9 Commits

Author SHA1 Message Date
Cole Robinson d2399d7bdf Fix qemu:///session connect race failures (bz #1271183)
driver: log missing modules as INFO, not WARN (bz #1274849)
2016-03-17 17:31:07 -04:00
Cole Robinson c76aa1c80f Fix XML validation with qemu commandline passthrough (bz #1292131)
Fix crash in libvirt_leasehelper (bz #1202350)
Generate consistent systemtap tapsets regardless of host arch (bz #1173641)
Fix qemu:///session error 'Transport endpoint is not connected' (bz #1271183)
Fix parallel VM start/top svirt errors on kernel/initrd (bz #1269975)
2016-01-20 20:04:34 -05:00
Cole Robinson 43c932192d Rebased to version 1.2.13.2
disk backend is not removed properly when disk frontent hotplug fails (bz #1265968)
Fix TPM cancel path on newer kernels (bz #1244895)
Remove timeout for libvirt-guests.service (bz #1195544)
CVE-2015-5313 libvirt: filesystem storage volume names path traversal flaw (bz #1291433)
Fix VM names with non-ascii (bz #1062943)
Fix backwards migration with graphics listen address (bz #1276883)
2015-12-23 19:09:14 -05:00
Cole Robinson 1b23098699 Fix polkit dep 2015-09-22 09:11:23 -04:00
Cole Robinson 925965e626 Add sanity checks for drive mirroring (bz #1263438) 2015-09-21 21:24:48 -04:00
Cole Robinson b8aa82790b Add patches 2015-06-06 11:17:32 -04:00
Cole Robinson ba6977e03f lxc network fixes (bz #1225591, bz #1225593, bz #1225594)
polkit: Allow password-less access for 'libvirt' group (bz #957300)
2015-06-06 11:12:42 -04:00
Cole Robinson e18130141c Rebased to version 1.2.13.1
Fix getVersion() after installing qemu (bz #1000116)
Fix autosocket setup with qemu:///session (bz #1044561, bz #1105274)
Ignore storage volumes with non-ascii in names (bz #1066564)
Don't generate invalid system nodedev XML (bz #1184131)
Fix crash via race when unrefing rpc identity object (bz #1203030)
Fix domcapabilities failure with ppc64le (bz #1209948)
Fix regression with 'virsh event ' (bz #1212620)
Add {Haswell,Broadwell}-noTSX CPU models (bz #1182650)
Don't lose VMs on libvirtd restart if qemu is uninstalled (bz #1099847)
Ignore storage volumes that libvirt can't open (bz #1103308)
2015-04-28 12:20:27 -04:00
Cole Robinson a99455b301 Fix LXC domain startup (bz #1210397)
Fix crash via identify object cleanup race (bz #1203030)
Fix race starting multiple session daemons (bz #1200149)
Fix change-media success messages
Strip invalid control codes from XML (bz #1066564, bz #1184131)
2015-04-15 14:37:39 -04:00
23 changed files with 3220 additions and 2440 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.*
+1 -1
View File
@@ -2,4 +2,4 @@
*.rpm
i686
x86_64
libvirt-*.tar.xz
libvirt-*.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
@@ -0,0 +1,38 @@
From: =?UTF-8?q?J=C3=A1n=20Tomko?= <jtomko@redhat.com>
Date: Thu, 17 Dec 2015 13:43:58 +0100
Subject: [PATCH] schema: interleave domain name and uuid with other elements
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Allow <name> and <uuid> anywhere under <domain>, not just at the top:
error:XML document failed to validate against schema: Unable to validate
doc against /usr/share/libvirt/schemas/domain.rng
Expecting an element name, got nothing
Invalid sequence in interleave
Element domain failed to validate content
Introduced with the first RelaxNG schema in commit c642103.
https://bugzilla.redhat.com/show_bug.cgi?id=1292131
(cherry picked from commit b4e0549febe416ffefc16f389423740d6d65fa74)
Signed-off-by: Ján Tomko <jtomko@redhat.com>
---
docs/schemas/domaincommon.rng | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng
index b252a17..48610ce 100644
--- a/docs/schemas/domaincommon.rng
+++ b/docs/schemas/domaincommon.rng
@@ -30,8 +30,8 @@
<define name="domain">
<element name="domain">
<ref name="hvs"/>
- <ref name="ids"/>
<interleave>
+ <ref name="ids"/>
<optional>
<ref name="title"/>
</optional>
@@ -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
@@ -0,0 +1,32 @@
From: =?UTF-8?q?J=C3=A1n=20Tomko?= <jtomko@redhat.com>
Date: Thu, 14 Jan 2016 14:31:17 +0100
Subject: [PATCH] leaseshelper: fix crash when no mac is specified
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
If dnsmasq specified DNSMASQ_IAID (so we're dealing with an IPv6
lease) but no DNSMASQ_MAC, we skip creation of the new lease object.
Also skip adding it to the leases array.
https://bugzilla.redhat.com/show_bug.cgi?id=1202350
(cherry picked from commit df9fe124d650bc438c531673492569da87523d20)
Signed-off-by: Ján Tomko <jtomko@redhat.com>
---
src/network/leaseshelper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/network/leaseshelper.c b/src/network/leaseshelper.c
index 2d528f7..6930310 100644
--- a/src/network/leaseshelper.c
+++ b/src/network/leaseshelper.c
@@ -439,7 +439,7 @@ main(int argc, char **argv)
case VIR_LEASE_ACTION_OLD:
case VIR_LEASE_ACTION_ADD:
- if (virJSONValueArrayAppend(leases_array_new, lease_new) < 0) {
+ if (lease_new && virJSONValueArrayAppend(leases_array_new, lease_new) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("failed to create json"));
goto cleanup;
@@ -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
@@ -0,0 +1,63 @@
From: Cole Robinson <crobinso@redhat.com>
Date: Tue, 19 Jan 2016 22:19:56 -0500
Subject: [PATCH] build: predictably generate systemtap tapsets (bz 1173641)
The generated output is dependent on perl hashtable ordering, which
gives different results for i686 and x86_64. Fix this by sorting
the hash keys before iterating over them
https://bugzilla.redhat.com/show_bug.cgi?id=1173641
(cherry picked from commit a1edb05c6028470aa24b74aa0f8d5fb5a181128a)
---
src/rpc/gensystemtap.pl | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/src/rpc/gensystemtap.pl b/src/rpc/gensystemtap.pl
index 2467300..7b80fbf 100755
--- a/src/rpc/gensystemtap.pl
+++ b/src/rpc/gensystemtap.pl
@@ -72,7 +72,7 @@ function libvirt_rpc_auth_name(type, verbose)
{
EOF
my $first = 1;
-foreach my $type (keys %auth) {
+foreach my $type (sort(keys %auth)) {
my $cond = $first ? "if" : "} else if";
$first = 0;
print " $cond (type == ", $type, ") {\n";
@@ -95,7 +95,7 @@ function libvirt_rpc_type_name(type, verbose)
{
EOF
$first = 1;
-foreach my $type (keys %type) {
+foreach my $type (sort(keys %type)) {
my $cond = $first ? "if" : "} else if";
$first = 0;
print " $cond (type == ", $type, ") {\n";
@@ -118,7 +118,7 @@ function libvirt_rpc_status_name(status, verbose)
{
EOF
$first = 1;
-foreach my $status (keys %status) {
+foreach my $status (sort(keys %status)) {
my $cond = $first ? "if" : "} else if";
$first = 0;
print " $cond (status == ", $status, ") {\n";
@@ -141,7 +141,7 @@ function libvirt_rpc_program_name(program, verbose)
{
EOF
$first = 1;
-foreach my $prog (keys %funcs) {
+foreach my $prog (sort(keys %funcs)) {
my $cond = $first ? "if" : "} else if";
$first = 0;
print " $cond (program == ", $funcs{$prog}->{id}, ") {\n";
@@ -165,7 +165,7 @@ function libvirt_rpc_procedure_name(program, version, proc, verbose)
{
EOF
$first = 1;
-foreach my $prog (keys %funcs) {
+foreach my $prog (sort(keys %funcs)) {
my $cond = $first ? "if" : "} else if";
$first = 0;
print " $cond (program == ", $funcs{$prog}->{id}, " && version == ", $funcs{$prog}->{version}, ") {\n";
@@ -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
@@ -0,0 +1,30 @@
From: "Daniel P. Berrange" <berrange@redhat.com>
Date: Fri, 3 Jul 2015 16:51:56 +0100
Subject: [PATCH] rpc: ensure daemon is spawn even if dead socket exists
The auto-spawn code would originally attempt to spawn the
daemon for both ENOENT and ECONNREFUSED errors from connect().
The various refactorings eventually lost this so we only
spawn the daemon on ENOENT. The result is if the daemon exits
uncleanly, so that the socket is left in the filesystem, we
will never be able to auto-spawn the daemon again.
(cherry picked from commit 406ee8c226d2197ba1aaecb9cf3ad2b6df31ae44)
---
src/rpc/virnetsocket.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c
index 51f94d4..6153e0e 100644
--- a/src/rpc/virnetsocket.c
+++ b/src/rpc/virnetsocket.c
@@ -610,7 +610,8 @@ int virNetSocketNewConnectUNIX(const char *path,
while (retries &&
connect(fd, &remoteAddr.data.sa, remoteAddr.len) < 0) {
- if (!(spawnDaemon && errno == ENOENT)) {
+ if (!(spawnDaemon && (errno == ENOENT ||
+ errno == ECONNREFUSED))) {
virReportSystemError(errno, _("Failed to connect socket to '%s'"),
path);
goto cleanup;
+48
View File
@@ -0,0 +1,48 @@
From: Cole Robinson <crobinso@redhat.com>
Date: Mon, 11 Jan 2016 20:01:24 -0500
Subject: [PATCH] rpc: socket: Minor cleanups
- Add some debugging
- Make the loop dependent only on retries
- Make it explicit that connect(2) success exits the loop
- Invert the error checking logic
(cherry picked from commit f102c7146ed7f6e04af0ad3bce302476239f2502)
---
src/rpc/virnetsocket.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c
index 6153e0e..dcff69e 100644
--- a/src/rpc/virnetsocket.c
+++ b/src/rpc/virnetsocket.c
@@ -548,6 +548,9 @@ int virNetSocketNewConnectUNIX(const char *path,
char *rundir = NULL;
int ret = -1;
+ VIR_DEBUG("path=%s spawnDaemon=%d binary=%s", path, spawnDaemon,
+ NULLSTR(binary));
+
memset(&localAddr, 0, sizeof(localAddr));
memset(&remoteAddr, 0, sizeof(remoteAddr));
@@ -608,10 +611,15 @@ int virNetSocketNewConnectUNIX(const char *path,
if (remoteAddr.data.un.sun_path[0] == '@')
remoteAddr.data.un.sun_path[0] = '\0';
- while (retries &&
- connect(fd, &remoteAddr.data.sa, remoteAddr.len) < 0) {
- if (!(spawnDaemon && (errno == ENOENT ||
- errno == ECONNREFUSED))) {
+ while (retries) {
+ if (connect(fd, &remoteAddr.data.sa, remoteAddr.len) == 0) {
+ VIR_DEBUG("connect() succeeded");
+ break;
+ }
+ VIR_DEBUG("connect() failed: retries=%d errno=%d", retries, errno);
+
+ if (!spawnDaemon ||
+ (errno != ENOENT && errno != ECONNREFUSED)) {
virReportSystemError(errno, _("Failed to connect socket to '%s'"),
path);
goto cleanup;
@@ -0,0 +1,40 @@
From: Cole Robinson <crobinso@redhat.com>
Date: Mon, 11 Jan 2016 20:08:45 -0500
Subject: [PATCH] rpc: socket: Explicitly error if we exceed retry count
When we autolaunch libvirtd for session URIs, we spin in a retry
loop waiting for the daemon to start and the connect(2) to succeed.
However if we exceed the retry count, we don't explicitly raise an
error, which can yield a slew of different error messages elsewhere
in the code.
Explicitly raise the last connect(2) failure if we run out of retries.
(cherry picked from commit 8da02d528068942303923fc4f935e77cccac9c7c)
---
src/rpc/virnetsocket.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c
index dcff69e..90951be 100644
--- a/src/rpc/virnetsocket.c
+++ b/src/rpc/virnetsocket.c
@@ -618,7 +618,9 @@ int virNetSocketNewConnectUNIX(const char *path,
}
VIR_DEBUG("connect() failed: retries=%d errno=%d", retries, errno);
+ retries--;
if (!spawnDaemon ||
+ retries == 0 ||
(errno != ENOENT && errno != ECONNREFUSED)) {
virReportSystemError(errno, _("Failed to connect socket to '%s'"),
path);
@@ -628,7 +630,6 @@ int virNetSocketNewConnectUNIX(const char *path,
if (virNetSocketForkDaemon(binary) < 0)
goto cleanup;
- retries--;
usleep(5000);
}
@@ -0,0 +1,43 @@
From: Cole Robinson <crobinso@redhat.com>
Date: Mon, 11 Jan 2016 20:13:38 -0500
Subject: [PATCH] rpc: socket: Don't repeatedly attempt to launch daemon
On every socket connect(2) attempt we were re-launching session
libvirtd, up to 100 times in 5 seconds.
This understandably caused some weird load races and intermittent
qemu:///session startup failures
https://bugzilla.redhat.com/show_bug.cgi?id=1271183
(cherry picked from commit 2eb7a975756d05a5b54ab4acf60083beb6161ac6)
---
src/rpc/virnetsocket.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c
index 90951be..2ee4b6e 100644
--- a/src/rpc/virnetsocket.c
+++ b/src/rpc/virnetsocket.c
@@ -547,6 +547,7 @@ int virNetSocketNewConnectUNIX(const char *path,
virSocketAddr remoteAddr;
char *rundir = NULL;
int ret = -1;
+ bool daemonLaunched = false;
VIR_DEBUG("path=%s spawnDaemon=%d binary=%s", path, spawnDaemon,
NULLSTR(binary));
@@ -627,8 +628,12 @@ int virNetSocketNewConnectUNIX(const char *path,
goto cleanup;
}
- if (virNetSocketForkDaemon(binary) < 0)
- goto cleanup;
+ if (!daemonLaunched) {
+ if (virNetSocketForkDaemon(binary) < 0)
+ goto cleanup;
+
+ daemonLaunched = true;
+ }
usleep(5000);
}
@@ -0,0 +1,57 @@
From: Jiri Denemark <jdenemar@redhat.com>
Date: Fri, 15 Jan 2016 10:55:58 +0100
Subject: [PATCH] security: Do not restore kernel and initrd labels
Kernel/initrd files are essentially read-only shareable images and thus
should be handled in the same way. We already use the appropriate label
for kernel/initrd files when starting a domain, but when a domain gets
destroyed we would remove the labels which would make other running
domains using the same files very unhappy.
https://bugzilla.redhat.com/show_bug.cgi?id=921135
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
(cherry picked from commit 68acc701bd449481e3206723c25b18fcd3d261b7)
---
src/security/security_dac.c | 8 --------
src/security/security_selinux.c | 8 --------
2 files changed, 16 deletions(-)
diff --git a/src/security/security_dac.c b/src/security/security_dac.c
index deb6980..d01215f 100644
--- a/src/security/security_dac.c
+++ b/src/security/security_dac.c
@@ -971,14 +971,6 @@ virSecurityDACRestoreSecurityAllLabel(virSecurityManagerPtr mgr,
virSecurityDACRestoreSecurityFileLabel(def->os.loader->nvram) < 0)
rc = -1;
- if (def->os.kernel &&
- virSecurityDACRestoreSecurityFileLabel(def->os.kernel) < 0)
- rc = -1;
-
- if (def->os.initrd &&
- virSecurityDACRestoreSecurityFileLabel(def->os.initrd) < 0)
- rc = -1;
-
if (def->os.dtb &&
virSecurityDACRestoreSecurityFileLabel(def->os.dtb) < 0)
rc = -1;
diff --git a/src/security/security_selinux.c b/src/security/security_selinux.c
index 6e67a86..2475a80 100644
--- a/src/security/security_selinux.c
+++ b/src/security/security_selinux.c
@@ -1953,14 +1953,6 @@ virSecuritySELinuxRestoreSecurityAllLabel(virSecurityManagerPtr mgr,
virSecuritySELinuxRestoreSecurityFileLabel(mgr, def->os.loader->nvram) < 0)
rc = -1;
- if (def->os.kernel &&
- virSecuritySELinuxRestoreSecurityFileLabel(mgr, def->os.kernel) < 0)
- rc = -1;
-
- if (def->os.initrd &&
- virSecuritySELinuxRestoreSecurityFileLabel(mgr, def->os.initrd) < 0)
- rc = -1;
-
if (def->os.dtb &&
virSecuritySELinuxRestoreSecurityFileLabel(mgr, def->os.dtb) < 0)
rc = -1;
@@ -0,0 +1,37 @@
From: Cole Robinson <crobinso@redhat.com>
Date: Tue, 15 Mar 2016 17:04:32 -0400
Subject: [PATCH] rpc: wait longer for session daemon to start up
https://bugzilla.redhat.com/show_bug.cgi?id=1271183
We only wait 0.5 seconds for the session daemon to start up and present
its socket, which isn't sufficient for many users. Bump up the sleep
interval and retry amount so we wait for a total of 5.0 seconds.
(cherry picked from commit ca0c06f4008154de55e0b3109885facd0bf02d32)
---
src/rpc/virnetsocket.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c
index 2ee4b6e..275f1f5 100644
--- a/src/rpc/virnetsocket.c
+++ b/src/rpc/virnetsocket.c
@@ -542,7 +542,7 @@ int virNetSocketNewConnectUNIX(const char *path,
char *lockpath = NULL;
int lockfd = -1;
int fd = -1;
- int retries = 100;
+ int retries = 500;
virSocketAddr localAddr;
virSocketAddr remoteAddr;
char *rundir = NULL;
@@ -635,7 +635,7 @@ int virNetSocketNewConnectUNIX(const char *path,
daemonLaunched = true;
}
- usleep(5000);
+ usleep(10000);
}
localAddr.len = sizeof(localAddr.data);
@@ -0,0 +1,27 @@
From: Jovanka Gulicoska <jovanka.gulicoska@gmail.com>
Date: Thu, 17 Mar 2016 20:02:20 +0100
Subject: [PATCH] driver: log missing modules as INFO, not WARN
Missing modules is a common expected scenario for most libvirt usage on
RPM distributions like Fedora, so it doesn't really warrant logging at
WARN level. Use INFO instead
https://bugzilla.redhat.com/show_bug.cgi?id=1274849
(cherry picked from commit 9a0c7f5f834185db9017c34aabc03ad99cf37bed)
---
src/driver.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/driver.c b/src/driver.c
index db03438..f926fe4 100644
--- a/src/driver.c
+++ b/src/driver.c
@@ -62,7 +62,7 @@ virDriverLoadModule(const char *name)
return NULL;
if (access(modfile, R_OK) < 0) {
- VIR_WARN("Module %s not accessible", modfile);
+ VIR_INFO("Module %s not accessible", modfile);
goto cleanup;
}
@@ -0,0 +1,126 @@
From: Cole Robinson <crobinso@redhat.com>
Date: Tue, 28 Apr 2015 17:38:00 -0400
Subject: [PATCH] polkit: Allow password-less access for 'libvirt' group
Many users, who admin their own machines, want to be able to access
system libvirtd via tools like virt-manager without having to enter
a root password. Just google 'virt-manager without password' and
you'll find many hits. I've read at least 5 blog posts over the years
describing slightly different ways of achieving this goal.
Let's finally add official support for this.
Install a polkit-1 rules file granting password-less auth for any user
in the new 'libvirt' group. Create the group on RPM install
https://bugzilla.redhat.com/show_bug.cgi?id=957300
(cherry picked from commit e94979e901517af9fdde358d7b7c92cc055dd50c)
---
daemon/Makefile.am | 13 +++++++++++++
daemon/libvirt.rules | 9 +++++++++
libvirt.spec.in | 15 +++++++++++++--
3 files changed, 35 insertions(+), 2 deletions(-)
create mode 100644 daemon/libvirt.rules
diff --git a/daemon/Makefile.am b/daemon/Makefile.am
index b95a79d..9c5ea37 100644
--- a/daemon/Makefile.am
+++ b/daemon/Makefile.am
@@ -53,6 +53,7 @@ EXTRA_DIST = \
libvirtd.init.in \
libvirtd.upstart \
libvirtd.policy.in \
+ libvirt.rules \
libvirtd.sasl \
libvirtd.service.in \
libvirtd.socket.in \
@@ -233,6 +234,8 @@ policyauth = auth_admin_keep_session
else ! WITH_POLKIT0
policydir = $(datadir)/polkit-1/actions
policyauth = auth_admin_keep
+rulesdir = $(datadir)/polkit-1/rules.d
+rulesfile = libvirt.rules
endif ! WITH_POLKIT0
endif WITH_POLKIT
@@ -263,9 +266,19 @@ if WITH_POLKIT
install-data-polkit::
$(MKDIR_P) $(DESTDIR)$(policydir)
$(INSTALL_DATA) libvirtd.policy $(DESTDIR)$(policydir)/org.libvirt.unix.policy
+if ! WITH_POLKIT0
+ $(MKDIR_P) $(DESTDIR)$(rulesdir)
+ $(INSTALL_DATA) $(srcdir)/$(rulesfile) $(DESTDIR)$(rulesdir)/50-libvirt.rules
+endif ! WITH_POLKIT0
+
uninstall-data-polkit::
rm -f $(DESTDIR)$(policydir)/org.libvirt.unix.policy
rmdir $(DESTDIR)$(policydir) || :
+if ! WITH_POLKIT0
+ rm -f $(DESTDIR)$(rulesdir)/50-libvirt.rules
+ rmdir $(DESTDIR)$(rulesdir) || :
+endif ! WITH_POLKIT0
+
else ! WITH_POLKIT
install-data-polkit::
uninstall-data-polkit::
diff --git a/daemon/libvirt.rules b/daemon/libvirt.rules
new file mode 100644
index 0000000..01a15fa
--- /dev/null
+++ b/daemon/libvirt.rules
@@ -0,0 +1,9 @@
+// Allow any user in the 'libvirt' group to connect to system libvirtd
+// without entering a password.
+
+polkit.addRule(function(action, subject) {
+ if (action.id == "org.libvirt.unix.manage" &&
+ subject.isInGroup("libvirt")) {
+ return polkit.Result.YES;
+ }
+});
diff --git a/libvirt.spec.in b/libvirt.spec.in
index dc327a2..a23629d 100644
--- a/libvirt.spec.in
+++ b/libvirt.spec.in
@@ -1631,9 +1631,9 @@ then
fi
%if %{with_libvirtd}
+%pre daemon
%if ! %{with_driver_modules}
%if %{with_qemu}
-%pre daemon
%if 0%{?fedora} || 0%{?rhel} >= 6
# We want soft static allocation of well-known ids, as disk images
# are commonly shared across NFS mounts by id rather than name; see
@@ -1647,11 +1647,21 @@ if ! getent passwd qemu >/dev/null; then
useradd -r -g qemu -G kvm -d / -s /sbin/nologin -c "qemu user" qemu
fi
fi
-exit 0
%endif
%endif
%endif
+ %if %{with_polkit}
+ %if 0%{?fedora} || 0%{?rhel} >= 6
+# 'libvirt' group is just to allow password-less polkit access to
+# libvirtd. The uid number is irrelevant, so we use dynamic allocation
+# described at the above link.
+getent group libvirt >/dev/null || groupadd -r libvirt
+ %endif
+ %endif
+
+exit 0
+
%post daemon
%if %{with_systemd}
@@ -1925,6 +1935,7 @@ exit 0
%if 0%{?fedora} || 0%{?rhel} >= 6
%{_datadir}/polkit-1/actions/org.libvirt.unix.policy
%{_datadir}/polkit-1/actions/org.libvirt.api.policy
+%{_datadir}/polkit-1/rules.d/50-libvirt.rules
%else
%{_datadir}/PolicyKit/policy/org.libvirt.unix.policy
%endif
+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) {
+2656 -2056
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
05f461cd499c628ef07822dd607c340a libvirt-1.2.13.2.tar.gz