[Spacewalk-list] Spacewalk proxy connection problem after upgrade?to 2.0

[Michael Mraka]

sebastien foucou wrote:
% Speigner Thomas <Thomas.Speigner at ...> writes:
% 
% > 
% > 
% > 
% > Hi everyone,
% >  
% > We upgraded Spacewalk to version 2.0. the master instance looks fine and 
% everything looks fine.
% > But we have problems after upgrade of the proxy instance in our 
% datacenter.
% > Now all servers behind the proxy are not able to query the master instance 
% with error: “XMLRPC ProtocolError: <ProtocolError for ***proxy*** /XMLRPC: 
% 500 Internal Server Error>”
% >  
% >[snip]  
% >  
% > We uninstalled/installed/configured the Spacewalk proxy again without 
% luck.
% > Anyone with the same error?
% >  
% > kind regards ...Thomas SPEIGNER
% > 
% 
% Hello
% 
% I updated my platefome spacewalk in version 2.0:
% the main server and the proxy.
% 
% I have the following problem on the proxy, so that everything works on the 
% primary server:
% 
% in /var/log/httpd/error_log
% 
% Exception reported from spwkf1p1
% Time: Tue Aug 27 15:28:17 2013
% Exception type exceptions.KeyError
% Exception while handling function handler
% Request object information:
% URI: /XMLRPC
% Remote Host: 172.xxx.xxx.xxx
% Server Name: spwkf1p1:443
% Headers passed in:
%         Accept-Encoding: identity
%         Content-Length: 2718
%         Host: spwkf1p1:443
%         content-type: text/xml
%         user-agent: rhn.rpclib.py/$Revision$
%         x-client-version: 1
%         x-info: RPC Processor (C) Red Hat, Inc (version $Revision$)
%         x-rhn-client-capability: 
% packages.verify(1)=1,packages.extended_profile(2)=1,configfiles.base64_enc(1
% )=1,reboot.reboot(1)=1,configfiles.deploy(1)=1,caneatCheese(1)=1,packages.ro
% llBack(1)=1,configfiles.mtime_upload(1)=1,configfiles.diff(1)=1,packages.upd
% ate(2)=2,script.run(1)=1,configfiles.upload(1)=1,packages.runTransaction(1)=
% 1,packages.verify(1)=1,packages.extended_profile(2)=1,configfiles.base64_enc
% (1)=1,reboot.reboot(1)=1,configfiles.deploy(1)=1,caneatCheese(1)=1,packages.
% rollBack(1)=1,configfiles.mtime_upload(1)=1,configfiles.diff(1)=1,packages.u
% pdate(2)=2,script.run(1)=1,configfiles.upload(1)=1,packages.runTransaction(1
% )=1
%         x-rhn-transport-capability: follow-redirects=3
%         x-transport-info: Extended Capabilities Transport (C) Red Hat, Inc 
% (version $Revision$)
%         x-up2date-version: 1.7.14-1.el5
% Extra information about this error:
% Unhandled exception type
% 
% Exception Handler Information
% Traceback (most recent call last):
%   File "/usr/share/rhn/proxy/apacheServer.py", line 62, in __call__
%     ret = f(req)
%   File "/usr/share/rhn/proxy/apacheHandler.py", line 355, in handler
%     ret = handlerObj.handler()
%   File "/usr/share/rhn/proxy/broker/rhnBroker.py", line 215, in handler
%     status = self._serverCommo()       # part 2
%   File "/usr/share/rhn/proxy/rhnShared.py", line 201, in _serverCommo
%     status, headers, bodyFd = self._proxy2server()
%   File "/usr/share/rhn/proxy/rhnShared.py", line 366, in _proxy2server
%     http_connection.send(self.req.headers_in['wsgi.input'])
% KeyError: 'wsgi.input'
% 
% Exception reported from spwkf1p1
% Time: Tue Aug 27 15:36:20 2013
% 
% My server is "CentOS release 5.9 (Final)"
% 

Hello Tomas and Sebastien,

this is definitely a bug (caused by commit 62bfef31c610c85c2e2adb06781b17c21cdf7dff).
The issue is we still use mod_python not mod_wsgi on proxy/EL5.

You can watch the progress in https://bugzilla.redhat.com/show_bug.cgi?id=1002007

Thanks for reporting,
Regards

--
Michael Mráka
Satellite Engineering, Red Hat