2.4.18-26 screen lockups and threading issues

Dave Johnson dave at wangtrading.com
Thu Mar 13 12:26:00 CST 2003


I meant to say "signal32" as opposed to "int32."  We
are still unsure why this all of sudden pops up in
applications under gdb with the new kernel upgrade...?

We fixed the threading problem.  It appears that one shouldn't 
define /lib in LD_LIBRARY_PATH, as there is a /lib/i686
directory that contains the pthreads.so and RedHat/Linux
automatically uses /lib/i686 if you don't have this defined 
(you can still have other libdirs listed...). 

Thanks to ezannoni at redhat.com for the quick response and the
pointers in the right direction.

----

Another possible issue with kernel...  Has anyone else seen 
Gconfd problems over NFS?  Looking at the bugzilla.gnome it 
shows up several times with no direct fix or workaround.  
2.4.18-26smp nfs locking somehow effects this, as we never 
had a problem before with user's homedirs nfs mounted, and 
then all of sudden they can't use evolution anymore (or at 
least any part of the evolution-mailer, which renders the 
application useless).  The NFS server is a Netapp.

- Dave

> -----Original Message-----
> From: Dave Johnson [mailto:dave at wangtrading.com] 
> Sent: Tuesday, March 11, 2003 3:19 PM
> To: 'linux-poweredge at dell.com'
> Subject: 2.4.18-26 screen lockups and threading issues
> 
> 
> 
> On a positive side, the kernel seems to noticeably speed things 
> up on the machine (bootups are several seconds faster).  Network 
> performance (and of course network stability) has increased.  We 
> ran some tests on a small farm of machines and it successfully 
> handled the load over the weekend.
> 
> Some concerns we have seen, which effects us rolling it out to 
> production are:
> 
> 1) On different Dell machines (1550s, 530s and 650s), we have 
>    seen lockups that last for several seconds (keyboard caps 
>    lock doesn't even register).  Most of these machines are 
>    running X at the time (RedHat 7.3).  Not sure what is 
>    causing this... ?  
> 
> 2) Is there something changed in the kernel threading model
>    for the default RH build for i686 and i686-smp of 2.4.18-26?
> 	- On machines with this kernel installed, GDB starts 
>         seeing int32 on debugged apps, which didn't exist in 
>         2.4.18-5smp.  From one of our developers here, even
>         sleep() behaves differently.
> 	
> 3) The latest NVIDIA driver seems to not be 100% with -26.  Has
>    anyone seen machine lockups with the NVIDIA display driver-4191.
>    (http://www.nvidia.com/view.asp?IO=linux_display_ia32_1.0-4191)
>    
> 
> Was wondering if anyone else has seen any of these issues... 
> ? We are about to open some bugs with RH...
> 
> Dave Johnson
> dave at wangtrading.com
> 
> 


--------------------------------------------------------
Note:
This message is for the named person's use only.  It may 
contain confidential, proprietary or legally privileged 
information.  No confidentiality or privilege is waived 
or lost by any mistransmission.  If you receive this 
message in error, please immediately delete it and all 
copies of it from your system, destroy any hard copies 
of it and notify the sender.  You must not, directly or 
indirectly, use, disclose, distribute, print, or copy 
any part of this message if you are not the intended 
recipient. Wang Trading LLC and any of its subsidiaries 
each reserve the right to monitor all e-mail 
communications through its networks.

Any views expressed in this message are those of the 
individual sender, except where the message states 
otherwise and the sender is authorized to state them 
to be the views of any such entity.
---------------------------------------------------------




More information about the Linux-PowerEdge mailing list