[PATCH 1/2] Timeout QEMU monitor replies after 30 seconds

["Daniel P. Berrange" <berrange@xxxxxxxxxx>]

If the QEMU process has been stopped (kill -STOP/gdb), or the
QEMU process has live-locked itself, then we will never get a
reply from the monitor. We should not wait forever in this
case, but instead timeout after a reasonable amount of time.

NB if the host has high CPU load, or a single monitor command
intentionally takes a long time, then this will cause bogus
failures. In the case of high CPU load, arguably the guest
should have been migrated elsewhere, since you can't effectively
manage guests on a host if QEMU is taking > 30 seconds to reply
to simply commands. Since we use background migration, there
should not be any commands which take significant time to
execute any more

* src/qemu/qemu_monitor.c: Timeout waiting for reply after 30 seconds
---
 src/qemu/qemu_monitor.c |   21 ++++++++++++++++++---
 1 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c
index 89a3f64..d9b6600 100644
--- a/src/qemu/qemu_monitor.c
+++ b/src/qemu/qemu_monitor.c
@@ -781,10 +781,19 @@ char *qemuMonitorNextCommandID(qemuMonitorPtr mon)
 }
 
 
+/* Give up waiting for reply after 30 seconds */
+#define QEMU_MONITOR_WAIT_TIME (1000ull * 30)
+
 int qemuMonitorSend(qemuMonitorPtr mon,
                     qemuMonitorMessagePtr msg)
 {
     int ret = -1;
+    unsigned long long now;
+    unsigned long long then;
+
+    if (virTimeMs(&now) < 0)
+        return -1;
+    then = now + QEMU_MONITOR_WAIT_TIME;
 
     /* Check whether qemu quited unexpectedly */
     if (mon->lastError.code != VIR_ERR_OK) {
@@ -798,9 +807,15 @@ int qemuMonitorSend(qemuMonitorPtr mon,
     qemuMonitorUpdateWatch(mon);
 
     while (!mon->msg->finished) {
-        if (virCondWait(&mon->notify, &mon->lock) < 0) {
-            qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s",
-                            _("Unable to wait on monitor condition"));
+        if (virCondWaitUntil(&mon->notify, &mon->lock, then) < 0) {
+            if (errno == ETIMEDOUT)
+                qemuReportError(VIR_ERR_OPERATION_TIMEOUT,
+                                "%s", _("no reply received from qemu"));
+            else
+                virReportSystemError(errno,
+                                     "%s", _("cannot wait on monitor condition"));
+            /* Ensure no further monitor commands can be run */
+            virCopyLastError(&mon->lastError);
             goto cleanup;
         }
     }
-- 
1.7.4.4

--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list