Bug 750037

Summary: vdagent is killed when client is connect to window 7 vm during startup on WAN (mouse doesn't work)
Product: Red Hat Enterprise Linux 8 Reporter: Yonit Halperin <yhalperi>
Component: spice-vdagent-winAssignee: Arnon Gilboa <agilboa>
Status: CLOSED ERRATA QA Contact: Desktop QE <desktop-qa-list>
Severity: medium Docs Contact:
Priority: unspecified    
Version: ---CC: acathrow, cpelland, dblechte, jbiddle, mbarta, mkrcmari, pvine, uril
Target Milestone: rcKeywords: ZStream
Target Release: ---   
Hardware: Unspecified   
OS: Windows   
Whiteboard:
Fixed In Version: vdagent-win-0.1-9 Doc Type: Bug Fix
Doc Text:
When connecting to a Windows 7 guest under WAN conditions the vdagent did not restart within the allocated timeout period of 3 seconds. This prevented the guest's mouse from responding to user input. An update has been made to ensure that the vdagent starts correctly to accept new session connections. As a result the guest's mouse will now respond correctly to user input under WAN conditions.
Story Points: ---
Clone Of:
: 805005 (view as bug list) Environment:
When connecting to a Windows 7 guest under WAN conditions the vdagent did not restart within the allocated timeout period of 3 seconds. This prevented the guest's mouse from responding to user input. An update has been made to ensure that the vdagent starts correctly to accept new session connections. As a result the guest's mouse will now respond correctly to user input under WAN conditions.
Last Closed: 2012-12-04 18:16:51 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 805005    
Attachments:
Description Flags
vdservice log
none
vdagent log
none
fix vdagent connection & termination handling none

Description Yonit Halperin 2011-10-30 09:33:43 UTC
Description of problem:
Under WAN conditions (e.g., 1Mbps), I connected spice client to a Windows7 vm, during the VM startup. The mouse didn't work. 
Investigate it with Arnon. Both server and client are in client mouse mode.
However, the agent is killed on the VM and is not restarted.
From the logs we saw, that vdservice attempts to kill the vdagent after a CONNECT event (as it should). It waits 3 seconds for the vdagent to be actually killed, and then it should restart it.
However, it took the agent more than 3 seconds to die, and the vdservice didn't restart it.
It is not totally clear why under WAN it takes longer to kill the agent.
I attached the vdservice and vdagent logs.


Version-Release number of selected component (if applicable):
vdagent-win-0.1-8
virtio-win-1.4.0
spice-client-0.8.2-7
spice-server-0.8.2-4

How reproducible:


Steps to Reproduce:
1.Start a Windows 7 vm
2. Connect to the vm during start up, with Spice client, Under WAN
  
Actual results:
No vdagent is running. i.e., mouse clicks don't work, copy-paste don't work, etc.

Comment 1 Yonit Halperin 2011-10-30 09:34:15 UTC
Created attachment 530821 [details]
vdservice log

Comment 2 Yonit Halperin 2011-10-30 09:34:39 UTC
Created attachment 530822 [details]
vdagent log

Comment 3 Arnon Gilboa 2011-11-21 10:25:54 UTC
Created attachment 534756 [details]
fix vdagent connection & termination handling

Comment 4 Arnon Gilboa 2011-11-22 13:12:05 UTC
This patch solves agent crash, and it's not risky from other perspectives.
Proposing to rhevm‑3.0.0.

Comment 5 Marian Krcmarik 2011-11-29 15:43:35 UTC
Switching back to ASSIGNED since the problem still occurs when connecting in early pahse of startup under WAn with low bandwith - ~100 Kbps.

We (Arnon and me) tested some scratch builds to determine the right solution.

Comment 6 Uri Lublin 2012-01-02 15:49:42 UTC
A Patch is available:
http://lists.freedesktop.org/archives/spice-devel/2011-December/006526.html

Comment 8 Uri Lublin 2012-08-16 11:50:48 UTC
Technical Notes are copied from bug 805005

Comment 12 errata-xmlrpc 2012-12-04 18:16:51 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2012-1502.html