[pve-devel] [PATCH qemu-server 1/2] fix #5985: qmp client: increase timeout for {device, netdev, object}_{add, del} commands

[Fiona Ebner <f.ebner@proxmox.com>]

In the bug report, the user mentioned that 7 seconds was enough. For
the HMP 'drive_add' command, the used timeout is 1 minute and for the
'drive_del' command, the used timeout is 10 minutes, because IO might
need to be finished. While something similar might be true for certain
objects/devices, there were no issues reported with the *_del
operations using the default timeout until now and the callers can
still use a higher timeout if they know the specific device/object
requires it.

Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
---
 PVE/QMPClient.pm | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/PVE/QMPClient.pm b/PVE/QMPClient.pm
index abed1f01..b03f1da8 100644
--- a/PVE/QMPClient.pm
+++ b/PVE/QMPClient.pm
@@ -118,6 +118,15 @@ sub cmd {
 	    # the variance for Windows guests can be big. And there might be hook scripts
 	    # that are executed upon thaw, so use 3 minutes to be on the safe side.
 	    $timeout = 3 * 60;
+	} elsif (
+	    $cmd->{execute} eq 'device_add' ||
+	    $cmd->{execute} eq 'device_del' ||
+	    $cmd->{execute} eq 'netdev_add' ||
+	    $cmd->{execute} eq 'netdev_del' ||
+	    $cmd->{execute} eq 'object-add' ||
+	    $cmd->{execute} eq 'object-del'
+	) {
+	    $timeout = 60;
 	} elsif (
 	    $cmd->{execute} eq 'backup-cancel' ||
 	    $cmd->{execute} eq 'blockdev-snapshot-delete-internal-sync' ||
-- 
2.39.5



_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel