Red Hat Bugzilla – Bug 53547
Installer stops after format
Last modified: 2007-04-18 12:37:02 EDT
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt)
Description of problem:
No matter what install options I choose, my compaq deskpro 2000 (5166MMX)
hangs during install just after formatting the drive. It returns the
HeaderLoad: assertion "rdlen == dl" failed.
This problem only occurs when i use a maxtor 30,7 GB drive (5T030H3) or a
maxtor 20 GB drive (I don't have any other drives). The original 2.1 GB
western digital drive that came with the deskpro 2000 doen't have this
problem. I used the same disks on a clone PC and it works fine.
- The BIOS large disk translations option is turned off.
- I tried all the compaq firmware upgrades that check for older maxtor
firmwares (didn't apply to any of my drives)
- Sometimes there is no error but the system just hangs.
- Doing a custom install with the LILO option turned off works OK
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Take a compaq deskpro 2000
2. replace the original disk with a brand new one (Maxtor)
3. install linux
4. wait for the error
Actual Results: The error "HeaderLoad: assertion "rdlen == dl" failed. "
Expected Results: Installer should start transferring the images to the
This sounds like an install code bug. Reassigning to the component anaconda and
the owner of that component.
Can you attach the complete debug error message?
Do you mean the complete contents of the screen after linux stops responding or
do I have to do anything special to obtain a "complete debug error message"?
Well, when you see the message "HeaderLoad: assertion "rdlen == dl" failed.", is
there any more information on the screen with it? Usually when a crash occurs,
a bunch of information is written to the screen that gives us a clue as to where
the crash occured. If there's more information on the screen, tell us what it says.
I can't really say what the problem is at this point, and I haven't seen this
problem before. I'm a little confused as to why the computer works with the old
hard drive, but not the new one...but the new hard drive works in a different
computer. This makes it seem like a hardware problem to me, but I can't say for
sure. Does the BIOS identify the new drive?
Closing due to inactivity. Please reopen if you have more information.
Sorry to keep you waiting. Just back from vacation. I will have a replacement
server in 2 weeks. After that i'll be able to do a new install on the compaq to
reproduce the message.
I'll be back.
In case you're wondering. I have completed the install and the machine is being
used as a mailserver. It boots from floppy and that seems to work. Although
sometimes I can see CRC errors on the screen which isn't very comforting.
- the maxtor drives passed all the tests on the maxtor diagnostic diskettes.
- I ran windows NT server 4.0 SP 6a on it for 90 days without any problems
although I used the WD 2.1 GB drive to boot during that period.
Anyway. I'll have more info in a couple of weeks
Ok, I'm reopening the bug report and will put in Needinfo state until your new
server comes in.
Have you made any progress?
The replacement server will hopefully arrive on monday. I'll have more on
Ok here it is:
#Top of the screen show the following message a number of times (+/- 7 times)
Gtk-CRITICAL **: file gtkaccelgroup.c: line 188 (gtk_accel_group_attach):
assertion 'g_slist_find (accel_group->attach_objects, object) == NULL' failed
#then the following text
python: header.c:511: headerload: Assertion 'rdlen == dl' failed
install exited abnormally
sending termination signals...done
sending kill signals...done
/mnt/source umount failed ()
you may safely reboot your system
Can you try doing the install in text mode? Just boot the installer by typing
'text'. Does that help?
#During install it says:
WARNING: Unknown EDD version 0x1600 supported
In dont know if this message shows up during a graphical install.
The rest is the same.
Python: headerload asserttion failed etc. etc.
This looks like a dupe of bug #37280
*** This bug has been marked as a duplicate of 37280 ***