pool/libvirt
SHA256
2
0
Fork 1
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Accepting request 480654 from home:jfehlig:branches:Virtualization

Browse Source 
- qemu: Fix monitor timeout with large memory VMs
  67dcb797-virTimeBackOffWait-sleepcap.patch
  85af0b80-qemu-adaptive-montimeout.patch
  bsc#1013113

OBS-URL: https://build.opensuse.org/request/show/480654
OBS-URL: https://build.opensuse.org/package/show/Virtualization/libvirt?expand=0&rev=591
This commit is contained in:
James Fehlig 2017-03-16 14:56:18 +00:00 committed by Git OBS Bridge
parent 04a0a5949c
commit 51e9f38d3f
5 changed files with 258 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

66
67dcb797-virTimeBackOffWait-sleepcap.patch Normal file
View File

@ -0,0 +1,66 @@
commit 67dcb797ed7f1fbb048aa47006576f424923933b
Author: Michal Privoznik <mprivozn@redhat.com>
Date: Mon Mar 13 11:05:08 2017 +0100
virTimeBackOffWait: Avoid long periods of sleep
While connecting to qemu monitor, the first thing we do is wait
for it to show up. However, we are doing it with some timeout to
avoid indefinite waits (e.g. when qemu doesn't create the monitor
socket at all). After beaa447a29 we are using exponential back
off timeout meaning, after the first connection attempt we wait
1ms, then 2ms, then 4 and so on. This allows us to bring down
wait time for small domains where qemu initializes quickly.
However, on the other end of this scale are some domains with
huge amounts of guest memory. Now imagine that we've gotten up to
wait time of 15 seconds. The next one is going to be 30 seconds,
and the one after that whole minute. Well, okay - with current
code we are not going to wait longer than 30 seconds in total,
but this is going to change in the next commit.
The exponential back off is usable only for first few iterations.
Then it needs to be caped (one second was chosen as the limit)
and switch to constant wait time.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Index: libvirt-3.1.0/src/util/virtime.c
===================================================================
--- libvirt-3.1.0.orig/src/util/virtime.c
+++ libvirt-3.1.0/src/util/virtime.c
@@ -390,6 +390,9 @@ virTimeBackOffStart(virTimeBackOffVar *v
return 0;
}
+
+#define VIR_TIME_BACKOFF_CAP 1000
+
/**
* virTimeBackOffWait
* @var: Timeout variable (with type virTimeBackOffVar *).
@@ -410,7 +413,9 @@ virTimeBackOffStart(virTimeBackOffVar *v
* The while loop that runs the body of the code repeatedly, with an
* exponential backoff. It first waits for first milliseconds, then
* runs the body, then waits for 2*first ms, then runs the body again.
- * Then 4*first ms, and so on.
+ * Then 4*first ms, and so on, up until wait time would reach
+ * VIR_TIME_BACK_OFF_CAP (whole second). Then it switches to constant
+ * waiting time of VIR_TIME_BACK_OFF_CAP.
*
* When timeout milliseconds is reached, the while loop ends.
*
@@ -429,8 +434,13 @@ virTimeBackOffWait(virTimeBackOffVar *va
if (t > var->limit_t)
return 0; /* ends the while loop */
+ /* Compute next wait time. Cap at VIR_TIME_BACKOFF_CAP
+ * to avoid long useless sleeps. */
next = var->next;
- var->next *= 2;
+ if (var->next < VIR_TIME_BACKOFF_CAP)
+ var->next *= 2;
+ else if (var->next > VIR_TIME_BACKOFF_CAP)
+ var->next = VIR_TIME_BACKOFF_CAP;
/* If sleeping would take us beyond the limit, then shorten the
* sleep. This is so we always run the body just before the final

171
85af0b80-qemu-adaptive-montimeout.patch Normal file
View File

@ -0,0 +1,171 @@
commit 85af0b803cd19a03f71bd01ab4e045552410368f
Author: Michal Privoznik <mprivozn@redhat.com>
Date: Sat Mar 11 07:23:42 2017 +0100
qemu: Adaptive timeout for connecting to monitor
There were couple of reports on the list (e.g. [1]) that guests
with huge amounts of RAM are unable to start because libvirt
kills qemu in the initialization phase. The problem is that if
guest is configured to use hugepages kernel has to zero them all
out before handing over to qemu process. For instance, 402GiB
worth of 1GiB pages took around 105 seconds (~3.8GiB/s). Since we
do not want to make the timeout for connecting to monitor
configurable, we have to teach libvirt to count with this
fact. This commit implements "1s per each 1GiB of RAM" approach
as suggested here [2].
1: https://www.redhat.com/archives/libvir-list/2017-March/msg00373.html
2: https://www.redhat.com/archives/libvir-list/2017-March/msg00405.html
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Index: libvirt-3.1.0/src/qemu/qemu_capabilities.c
===================================================================
--- libvirt-3.1.0.orig/src/qemu/qemu_capabilities.c
+++ libvirt-3.1.0/src/qemu/qemu_capabilities.c
@@ -4571,7 +4571,7 @@ virQEMUCapsInitQMPCommandRun(virQEMUCaps
cmd->vm->pid = cmd->pid;
if (!(cmd->mon = qemuMonitorOpen(cmd->vm, &cmd->config, true,
- &callbacks, NULL)))
+ 0, &callbacks, NULL)))
goto ignore;
virObjectLock(cmd->mon);
Index: libvirt-3.1.0/src/qemu/qemu_monitor.c
===================================================================
--- libvirt-3.1.0.orig/src/qemu/qemu_monitor.c
+++ libvirt-3.1.0/src/qemu/qemu_monitor.c
@@ -327,11 +327,13 @@ qemuMonitorDispose(void *obj)
static int
-qemuMonitorOpenUnix(const char *monitor, pid_t cpid)
+qemuMonitorOpenUnix(const char *monitor,
+ pid_t cpid,
+ unsigned long long timeout)
{
struct sockaddr_un addr;
int monfd;
- virTimeBackOffVar timeout;
+ virTimeBackOffVar timebackoff;
int ret = -1;
if ((monfd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) {
@@ -348,9 +350,9 @@ qemuMonitorOpenUnix(const char *monitor,
goto error;
}
- if (virTimeBackOffStart(&timeout, 1, 30*1000 /* ms */) < 0)
+ if (virTimeBackOffStart(&timebackoff, 1, timeout * 1000) < 0)
goto error;
- while (virTimeBackOffWait(&timeout)) {
+ while (virTimeBackOffWait(&timebackoff)) {
ret = connect(monfd, (struct sockaddr *) &addr, sizeof(addr));
if (ret == 0)
@@ -871,10 +873,30 @@ qemuMonitorOpenInternal(virDomainObjPtr
}
+#define QEMU_DEFAULT_MONITOR_WAIT 30
+
+/**
+ * qemuMonitorOpen:
+ * @vm: domain object
+ * @config: monitor configuration
+ * @json: enable JSON on the monitor
+ * @timeout: number of seconds to add to default timeout
+ * @cb: monitor event handles
+ * @opaque: opaque data for @cb
+ *
+ * Opens the monitor for running qemu. It may happen that it
+ * takes some time for qemu to create the monitor socket (e.g.
+ * because kernel is zeroing configured hugepages), therefore we
+ * wait up to default + timeout seconds for the monitor to show
+ * up after which a failure is claimed.
+ *
+ * Returns monitor object, NULL on error.
+ */
qemuMonitorPtr
qemuMonitorOpen(virDomainObjPtr vm,
virDomainChrSourceDefPtr config,
bool json,
+ unsigned long long timeout,
qemuMonitorCallbacksPtr cb,
void *opaque)
{
@@ -882,10 +904,14 @@ qemuMonitorOpen(virDomainObjPtr vm,
bool hasSendFD = false;
qemuMonitorPtr ret;
+ timeout += QEMU_DEFAULT_MONITOR_WAIT;
+
switch (config->type) {
case VIR_DOMAIN_CHR_TYPE_UNIX:
hasSendFD = true;
- if ((fd = qemuMonitorOpenUnix(config->data.nix.path, vm ? vm->pid : 0)) < 0)
+ if ((fd = qemuMonitorOpenUnix(config->data.nix.path,
+ vm ? vm->pid : 0,
+ timeout)) < 0)
return NULL;
break;
Index: libvirt-3.1.0/src/qemu/qemu_monitor.h
===================================================================
--- libvirt-3.1.0.orig/src/qemu/qemu_monitor.h
+++ libvirt-3.1.0/src/qemu/qemu_monitor.h
@@ -246,6 +246,7 @@ char *qemuMonitorUnescapeArg(const char
qemuMonitorPtr qemuMonitorOpen(virDomainObjPtr vm,
virDomainChrSourceDefPtr config,
bool json,
+ unsigned long long timeout,
qemuMonitorCallbacksPtr cb,
void *opaque)
ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(4);
Index: libvirt-3.1.0/src/qemu/qemu_process.c
===================================================================
--- libvirt-3.1.0.orig/src/qemu/qemu_process.c
+++ libvirt-3.1.0/src/qemu/qemu_process.c
@@ -1656,6 +1656,7 @@ qemuConnectMonitor(virQEMUDriverPtr driv
qemuDomainObjPrivatePtr priv = vm->privateData;
int ret = -1;
qemuMonitorPtr mon = NULL;
+ unsigned long long timeout = 0;
if (virSecurityManagerSetDaemonSocketLabel(driver->securityManager,
vm->def) < 0) {
@@ -1664,6 +1665,12 @@ qemuConnectMonitor(virQEMUDriverPtr driv
return -1;
}
+ /* When using hugepages, kernel zeroes them out before
+ * handing them over to qemu. This can be very time
+ * consuming. Therefore, add a second to timeout for each
+ * 1GiB of guest RAM. */
+ timeout = vm->def->mem.total_memory / (1024 * 1024);
+
/* Hold an extra reference because we can't allow 'vm' to be
* deleted until the monitor gets its own reference. */
virObjectRef(vm);
@@ -1674,6 +1681,7 @@ qemuConnectMonitor(virQEMUDriverPtr driv
mon = qemuMonitorOpen(vm,
priv->monConfig,
priv->monJSON,
+ timeout,
&monitorCallbacks,
driver);
Index: libvirt-3.1.0/tests/qemumonitortestutils.c
===================================================================
--- libvirt-3.1.0.orig/tests/qemumonitortestutils.c
+++ libvirt-3.1.0/tests/qemumonitortestutils.c
@@ -1175,6 +1175,7 @@ qemuMonitorTestNew(bool json,
if (!(test->mon = qemuMonitorOpen(test->vm,
&src,
json,
+ 0,
&qemuMonitorTestCallbacks,
driver)))
goto error;

8
libvirt.changes
View File

@ -1,3 +1,11 @@
-------------------------------------------------------------------
Thu Mar 16 14:23:16 UTC 2017 - jfehlig@suse.com
- qemu: Fix monitor timeout with large memory VMs
67dcb797-virTimeBackOffWait-sleepcap.patch
85af0b80-qemu-adaptive-montimeout.patch
bsc#1013113
-------------------------------------------------------------------
Tue Mar 7 22:49:32 UTC 2017 - mwilck@suse.com

4
libvirt.spec
View File

@ -317,6 +317,8 @@ Source4: libvirtd-relocation-server.fw
Source99: baselibs.conf
Source100: %{name}-rpmlintrc
# Upstream patches
Patch0: 67dcb797-virTimeBackOffWait-sleepcap.patch
Patch1: 85af0b80-qemu-adaptive-montimeout.patch
# Patches pending upstream review
Patch100: libxl-dom-reset.patch
Patch101: network-don-t-use-dhcp-authoritative-on-static-netwo.patch
@ -884,6 +886,8 @@ libvirt plugin for NSS for translating domain names into IP addresses.
%prep
%setup -q
%patch0 -p1
%patch1 -p1
%patch100 -p1
%patch101 -p1
%patch150 -p1

21
network-don-t-use-dhcp-authoritative-on-static-netwo.patch
View File

@ -17,11 +17,11 @@ Signed-off-by: Martin Wilck <mwilck@suse.com>
tests/networkxml2confdata/dhcp6host-routed-network.conf | 1 -
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index ae1589d8c..17c6f3a0f 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -1355,7 +1355,14 @@ networkDnsmasqConfContents(virNetworkObjPtr network,
Index: libvirt-3.1.0/src/network/bridge_driver.c
===================================================================
--- libvirt-3.1.0.orig/src/network/bridge_driver.c
+++ libvirt-3.1.0/src/network/bridge_driver.c
@@ -1398,7 +1398,14 @@ networkDnsmasqConfContents(virNetworkObj
if (VIR_SOCKET_ADDR_IS_FAMILY(&ipdef->address, AF_INET)) {
if (ipdef->nranges || ipdef->nhosts) {
virBufferAddLit(&configbuf, "dhcp-no-override\n");
@ -37,10 +37,10 @@ index ae1589d8c..17c6f3a0f 100644
}
if (ipdef->tftproot) {
diff --git a/tests/networkxml2confdata/dhcp6host-routed-network.conf b/tests/networkxml2confdata/dhcp6host-routed-network.conf
index 87a149880..5728ee430 100644
--- a/tests/networkxml2confdata/dhcp6host-routed-network.conf
+++ b/tests/networkxml2confdata/dhcp6host-routed-network.conf
Index: libvirt-3.1.0/tests/networkxml2confdata/dhcp6host-routed-network.conf
===================================================================
--- libvirt-3.1.0.orig/tests/networkxml2confdata/dhcp6host-routed-network.conf
+++ libvirt-3.1.0/tests/networkxml2confdata/dhcp6host-routed-network.conf
@@ -10,7 +10,6 @@ bind-dynamic
interface=virbr1
dhcp-range=192.168.122.1,static
@ -49,6 +49,3 @@ index 87a149880..5728ee430 100644
dhcp-range=2001:db8:ac10:fd01::1,static,64
dhcp-hostsfile=/var/lib/libvirt/dnsmasq/local.hostsfile
addn-hosts=/var/lib/libvirt/dnsmasq/local.addnhosts
--
2.11.0