Red Hat Bugzilla – Bug 491348
Traceback in partitioning - unable to satisfy constraints
Last modified: 2009-04-14 16:59:06 EDT
The following was filed automatically by anaconda:
anaconda 220.127.116.11 exception report
Traceback (most recent call first):
File "/usr/lib64/python2.6/site-packages/parted/disk.py", line 181, in addPartition
File "/usr/lib/anaconda/storage/partitioning.py", line 754, in allocatePartitions
File "/usr/lib/anaconda/storage/partitioning.py", line 556, in doPartitioning
File "/usr/lib/anaconda/storage/partitioning.py", line 191, in doAutoPartition
File "/usr/lib/anaconda/dispatch.py", line 205, in moveStep
rc = stepFunc(self.anaconda)
File "/usr/lib/anaconda/dispatch.py", line 128, in gotoNext
File "/usr/lib/anaconda/dispatch.py", line 227, in currentStep
File "/usr/lib/anaconda/gui.py", line 1350, in setScreen
(step, anaconda) = self.anaconda.dispatch.currentStep()
File "/usr/lib/anaconda/gui.py", line 1513, in setup_window
File "/usr/lib/anaconda/gui.py", line 1523, in run
File "/usr/lib/anaconda/gui.py", line 1270, in run
File "/usr/bin/anaconda", line 1017, in <module>
PartitionException: Unable to satisfy all constraints on the partition.
Created attachment 336057 [details]
Attached traceback automatically from anaconda.
*** Bug 491493 has been marked as a duplicate of this bug. ***
*** Bug 491910 has been marked as a duplicate of this bug. ***
Still present with 18.104.22.168 today.
My bug#491910 was found during a RAID kickstart install.
Orion and Ronald: Are you seeing this during a kickstart installation, or a manual install on using dmraid hardware?
kickstart, but so far only on a machine with 2 existing non-linux partitions that are preserved. Installs on pure linux machines on which all partitions are first removed by kickstart are fine.
Orion, thanks for the feedback. The anaconda team will be focusing on improving the kickstart experience post-beta. The focus for beta is around manual installations.
I've added this issue to the known issues list for F11 Beta (https://fedoraproject.org/wiki/Fedora_11_Beta_release_notes#Anaconda_installer_issues).
- manual install !
- no RAID
00:12.0 SATA controller: ATI Technologies Inc SB600 Non-Raid-5 SATA (prog-if 01 [AHCI 1.0])
is this dmraid hardware ?
*** Bug 492471 has been marked as a duplicate of this bug. ***
*** Bug 493140 has been marked as a duplicate of this bug. ***
Been following this bug for some time and think I have a hack for it. Would appreciate people testing with one of the following update images:
The following updates are the same solution but less hackish. Would appreciate a test for each comment image, if you have time :).
but no luck.
installer stops with a popup message "Filesystem error detected, cannot continue"
with an OK "button"
attached all file under /tmp.
add. blkid and fdisk -l /dev/sd?
hint: sda2 is Solaris ZFS
Created attachment 337616 [details]
Created attachment 337617 [details]
Created attachment 337618 [details]
Created attachment 337619 [details]
Created attachment 337620 [details]
Created attachment 337621 [details]
Testing with the x86_64 updates image posted in comment#13 results in ...
Running anaconda 22.214.171.124, the Fedora system installer - please wait...
Traceback (most recent call last):
File "/tmp/updates/anaconda", line 549, in <module>
import signal, string, isys, iutil, time
File "/tmp/updates/isys.py", line 38, in <module>
File "/tmp/updates/block/__init__.py", line 20, in <module>
ImportError: /tmp/updates/block/dmmodule.so: wrong ELF class: ELFCLASS32
This was tested by performing a raid0 kickstart install of rawhide while booting with updates=http://jgranado.fedorapeople.org/storage/testing/01-04-2009-2010-x86_64.img
I hit this (see bug 491493) when attempting to format a partition as xfs in the custom partitioning screen, in the x86 F11 Beta installer.
(In reply to comment #14)
> Thanks !
> but no luck.
> installer stops with a popup message "Filesystem error detected, cannot
> with an OK "button"
This means your hitting another bug before this one. thx for testing!
The images I created for comment 12 and 13 were a little off. These new images should correct the problem and hopefully take care of the bug :)
*** Bug 493498 has been marked as a duplicate of this bug. ***
*** Bug 493502 has been marked as a duplicate of this bug. ***
Created attachment 337821 [details]
Traceback while testing 02-04-2009-1351-x86_64.img
Testing with rawhide + updates=http://jgranado.fedorapeople.org/storage/testing/02-04-2009-1351-x86_64.img yields the attached traceback.
Couple of points from Bug 493498.
- no RAID
- Encrypted /home (though installing into free space)
- Boot.iso (rawhide-20090331)
Also couple of points from https://bugzilla.redhat.com/show_bug.cgi?id=493498
320 GB HD
Booted from Boot.iso (rawhide-20090402)
Created attachment 338089 [details]
Testing with rawhide 20090403 +
yields the attached traceback.
Different than the other one posted.
*** Bug 494029 has been marked as a duplicate of this bug. ***
*** Bug 494184 has been marked as a duplicate of this bug. ***
Created attachment 338526 [details]
I got a very similar error today when I tried to install Fedora 11 Beta.
The disk layout is as follows:
Disk /dev/sda: 60801 cylinders, 255 heads, 63 sectors/track
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0
Device Boot Start End #cyls #blocks Id System
/dev/sda1 0+ 4145 4146- 33302713+ 7 HPFS/NTFS
/dev/sda2 4146 4158- 13- 97657 83 Linux
/dev/sda3 4158+ 52720 48563- 390081030+ 5 Extended
/dev/sda4 * 52721 60801- 8081- 64905151+ bf Solaris
/dev/sda5 4158+ 6589- 2432- 19531250+ 83 Linux
/dev/sda6 6589+ 7075- 487- 3906250+ 82 Linux swap / Solaris
/dev/sda7 7076+ 9687 2612- 20980862+ 83 Linux
There is plenty of room in the extended partition. I used the graphical installer, no RAID, nothing unusual. I just pressed the "New" button and tried to create a new root (/) partition, format as ext4, sized 20000MB, and immediately got the traceback when I pressed the "Ok" button (or whatever its name is).
This is addressed in current anaconda.
*** Bug 494807 has been marked as a duplicate of this bug. ***
Getting different traceback today (bug #495076)- no idea if I got farther or not.
I have a similar experience as Orion in comment#36. I'm not seeing this issue anymore, but am now seeing bug#495077.
No more partitioning tracebacks for me with 126.96.36.199.