Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 1216219 - kimchi novnc screen fails on websockify due to endian problems
Summary: kimchi novnc screen fails on websockify due to endian problems
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Fedora
Classification: Fedora
Component: python-websockify
Version: 22
Hardware: ppc64
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: Pádraig Brady
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-04-28 19:28 UTC by Jaromír Cápík
Modified: 2016-02-01 02:01 UTC (History)
7 users (show)

Fixed In Version: python-websockify-0.6.0-2.fc22
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2015-04-30 11:48:42 UTC
Type: Bug
Embargoed:

Attachments (Terms of Use)

Description Jaromír Cápík 2015-04-28 19:28:19 UTC 
Description of problem:
We're testing kimchi on ppc64 (big endian) and experiencing problems with novnc connections.
The kimchi upstream told me it's a known problem on big endian systems and provided me with the following link containing a patch against websockify.

http://lists.ovirt.org/pipermail/kimchi-devel/2014-August/007071.html

The thread states it's caused by numpy and the websockify patch is only temporary and will be removed once numpy is fixed. BUT, I tested the bitwise xor in numpy and it works correctly. That means the root cause really lies somewhere in the websockify component. I can confirm the attached patch fixes the issue.

Version-Release number of selected component (if applicable):
python-websockify-0.6.0-1.fc22.noarch
numpy-1.9.2-1.fc22.ppc64

How reproducible:
always

Steps to Reproduce:
1. Try to connect to kimchi guest screens


Actual results:
Error message : "Server disconnected (code: 1000, reason: Target closed)"

Expected results:
Working novnc session
Comment 1 Pádraig Brady 2015-04-28 22:28:10 UTC 
I'll probably do something with:

if sys.byteorder = 'big':
  swap

Note socket.ntohl() would be a noop on big endian I think,
which is opposite to what we want.
Comment 2 Pádraig Brady 2015-04-29 12:56:37 UTC 
Going with this from upstream

https://github.com/kanaka/websockify/commit/af10f458
Comment 3 Fedora Update System 2015-04-29 13:27:08 UTC 
python-websockify-0.6.0-2.fc22 has been submitted as an update for Fedora 22.
https://admin.fedoraproject.org/updates/python-websockify-0.6.0-2.fc22
Comment 4 Fedora Update System 2015-04-29 13:31:24 UTC 
python-websockify-0.6.0-2.el7 has been submitted as an update for Fedora EPEL 7.
https://admin.fedoraproject.org/updates/python-websockify-0.6.0-2.el7
Comment 5 Jaromír Cápík 2015-04-29 15:29:07 UTC 
Thanks Pádraig.

That was really quick.

I can confirm python-websockify-0.6.0-2.fc22 fixes the issue.
Comment 6 Fedora Update System 2015-04-30 11:48:42 UTC 
python-websockify-0.6.0-2.fc22 has been pushed to the Fedora 22 stable repository.  If problems still persist, please make note of it in this bug report.
Comment 7 Fedora Update System 2015-05-14 20:09:28 UTC 
python-websockify-0.6.0-2.el7 has been pushed to the Fedora EPEL 7 stable repository.  If problems still persist, please make note of it in this bug report.
Comment 8 Fedora Update System 2015-05-23 18:07:49 UTC 
python-websockify-0.6.0-2.fc21 has been submitted as an update for Fedora 21.
https://admin.fedoraproject.org/updates/python-websockify-0.6.0-2.fc21
Note You need to log in before you can comment on or make changes to this bug.