[Synaptic-devel] [support #102356] failures w/ RedHat 9

[Michael Vogt]

This mail is an automated notification from the support tracker
 of the project: Synaptic.

/**************************************************************************/
[support #102356] Latest Modifications:

Changes by: 
                Michael Vogt <address@hidden>
'Date: 
                Tue 06/01/04 at 13:33 (GMT)

            What     | Removed                   | Added
---------------------------------------------------------------------------
          Resolution | None                      | Fixed
              Status | Open                      | Closed


------------------ Additional Follow-up Comments ----------------------------
problems should be fixed by now. I didn't got a reply to my requets about new 
RH packages, so I close this support request for now. If you still have 
problem, please reopen it. 






/**************************************************************************/
[support #102356] Full Item Snapshot:

URL: <http://savannah.nongnu.org/support/?func=detailitem&item_id=102356>
Project: Synaptic
Submitted by: 0
On: Fri 08/15/03 at 17:57

Category:  None
Priority:  5 - Normal
Severity:  3 - Ordinary
Resolution:  Fixed
Assigned to:  mvogt
Originator Email:  
Status:  Closed


Summary:  failures w/ RedHat 9

Original Submission:  
Synaptic is having some problems dealing with RH9.  
First, my config:

RedHat 9.0 with all current RH updates installed
synaptic_0.42-1.fr_i386.rpm installed
apt-0.5.5cnc6-fr1 installed
using WRQ Reflection 7.2 xterminal window to linux box
Reflection is running on Windows 2000/SP4 (not XDMCP)

OK, here is what I'm seeing...  

First, I start synaptic as a background process from 
the xterminal.  The process initially recieves an 
error "Xlib: extension 'RENDER' missing on display 
(myip:0.0)" seen in the xterminal window.  From some 
research I've done suggests that this error should not 
be a concern, but who knows.  

Next, whenever I make select any function in synaptic 
I receive a message back on my xwindow session 
saying, "strange things are going on in the world".  
Sure is strange!  My hope is that the author has coded 
this to be comical.

Now for the troublesome stuff...  When I close a 
window or subwindow I receive the 
error, "unregisterObserver() failed" back in the 
xterminal screen.  Not sure what this means and a 
Google search comes up with nothing.  

When I perform a list update things process as 
expected without error.  Then after I select some 
packages to install, synaptic starts processing the 
download.  Everything downloads fine.  Now when it 
attempts to install the rpm's we have some real 
problems.  The synaptic main and subwindow hang.  Any 
attempts to kill them short of a -9 do nothing.  The 
only way I can coax the process to do anything is to 
go back to the xsession terminal window and bring the 
synaptic process to the foreground.  When I do this a 
new child window appears with a double scrolling bar 
that tells me in its message that its trying to 
install the packages with the rpm process.  Nothing 
happens beyond the scrolling though.  Doing a ps -ef 
indicates that NO rpm process is running in the 
machine.  If I try to perform a rpm -qa command from a 
text based session (non x), I receive the message from 
rpm that, "warning: cannot get shared lock 
on /var/lib/rpm/Packages".  I have to kill off the 
xterminal window at this point to kill off synaptic.  
According to ps -ef the synaptic processes go away but 
the rpm lock remains.  I'm not familiar with the rpm 
locking scheme, but a quick machine reboot clears it.

So, what to do from here...  For the moment I'm 
telling synaptic to download the packages only and I'm 
later installing them manually with rpm.  When I later 
bring up synaptic again, it recognizes the additions.  

In one case I tried to let synaptic handle a 'broken' 
issue with rhs-printfilters vs redhat-config-printer.  
I saw the same symptoms as before but I noticed a 
curious thing afterward.  I performed an rpm erase 
with the -vv option and noticed that the previous 
iteration (using synaptic) had in fact removed the 
files but had not updated the rpm databases.

So now I'm wondering if this is really a synaptic 
problem or really something to do with the x session 
instead.  While I'm waiting, I will attempt to some 
work via the machine console and see if things are any 
different.

Thanks for your time,

Tim

(FYI, I was logged in but it appears the cookie 
dropped - savannah user "tgm")

Follow-up Comments
------------------


-------------------------------------------------------
Date: Tue 06/01/04 at 13:33         By: mvogt
problems should be fixed by now. I didn't got a reply to my requets about new 
RH packages, so I close this support request for now. If you still have 
problem, please reopen it. 

-------------------------------------------------------
Date: Tue 11/18/03 at 11:47         By: mvogt

We have a new RH9 RPM at
http://pimpldrive.pimpzkru.de/renate/synaptic. Can you
please check if it works better for you?

-------------------------------------------------------
Date: Fri 08/15/03 at 23:07         By: mvogt

thanks for your bugreport.

>First, I start synaptic as a background process from 
>the xterminal.  The process initially recieves an 
>error "Xlib: extension 'RENDER' missing on display 
>(myip:0.0)" seen in the xterminal window.  From some 
>research I've done suggests that this error should not 
>be a concern, but who knows.  

No, this is no problem.

>Next, whenever I make select any function in synaptic 
>I receive a message back on my xwindow session 
>saying, "strange things are going on in the world".  
>Sure is strange!  My hope is that the author has coded 
>this to be comical.

sorry for this. this was a internal debug statement that
made it into the release. It's (relativly) harmless, but
it is fixed in the upcoming 0.43 release.

>Now for the troublesome stuff...  When I close a 
>window or subwindow I receive the 
>error, "unregisterObserver() failed" back in the 
>xterminal screen.  Not sure what this means and a 
>Google search comes up with nothing.  

This is another debug statement. It warns that a internal
object couldn't be deregistered. It's harmsless, it just
shows that there is a internal (but harmless) problem.

the other stuff is more serious.

I'll see what I can do about it tomorrow. 

thanks, 
 Michael












For detailed info, follow this link:
<http://savannah.nongnu.org/support/?func=detailitem&item_id=102356>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.nongnu.org/