From: Jiri Denemark <jdenemar@redhat.com>
Date: Mon, 17 Mar 2014 10:04:07 +0000 (+0100)
Subject: qemu: Return meaningful error when qemu dies early
X-Git-Tag: v1.2.3-rc1~96
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=cfa7ceab7735410c0427136236bf8bad10670816;p=thirdparty%2Flibvirt.git

qemu: Return meaningful error when qemu dies early

https://bugzilla.redhat.com/show_bug.cgi?id=844378

When qemu dies early after connecting to its monitor but before we
actually try to read something from the monitor, we would just fail
domain start with useless message:

    "An error occurred, but the cause is unknown"

This is because the real error gets reported in a monitor EOF handler
executing within libvirt's event loop.

The fix is to take any error set in qemuMonitor structure and propagate
it into the thread-local error when qemuMonitorClose is called and no
thread-local error is set.

Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
---

diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c
index 5f8a5f4cbb..48fc51cf87 100644
--- a/src/qemu/qemu_monitor.c
+++ b/src/qemu/qemu_monitor.c
@@ -928,6 +928,12 @@ void qemuMonitorClose(qemuMonitorPtr mon)
         virCondSignal(&mon->notify);
     }
 
+    /* Propagate existing monitor error in case the current thread has no
+     * error set.
+     */
+    if (mon->lastError.code != VIR_ERR_OK && !virGetLastError())
+        virSetError(&mon->lastError);
+
     virObjectUnlock(mon);
     virObjectUnref(mon);
 }