GNOME Bugzilla – Bug 614220
Unable to satisfy all constraints on the partition
Last modified: 2014-01-05 19:56:32 UTC
Gparted refuses to display the partition table exhausting this message. fdisk is able to display the partition table. Since Gparted leaves it unclear what constrains are not satisfied I expect this to be a bug --- at least Gparted should tell what wasn't satisfied. Gparted 0.5.1 libparted 2.2 ##### root@aesculap:~# fdisk -l Platte /dev/sda: 250.1 GByte, 250059350016 Byte 255 K��pfe, 63 Sektoren/Spur, 30401 Zylinder Einheiten = Zylinder von 16065 �� 512 = 8225280 Bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x3bf9e81b Ger��t boot. Anfang Ende Bl��cke Id System /dev/sda1 1 7 56196 83 Linux /dev/sda2 * 8 10319 82831140 7 HPFS/NTFS /dev/sda3 10320 30401 161308665 f W95 Erw. (LBA) /dev/sda5 10320 15673 43005942 83 Linux /dev/sda6 15674 15934 2096451 82 Linux Swap / Solaris /dev/sda7 15935 24289 67111537+ b W95 FAT32 ##### AFAICS there is nothing I would expect problematic here ... PS: same for Gparted 0.5.2; GIT HEAD I could not test, doesn't compile :(
GParted uses the libparted library from the parted project to detect and manipulate partition tables. As such the message "Unable to satisfy all constraints on the partition" most likely originates from the libparted library. To have this changed requires a change in the parted project source code. As such would you be able to provide the output from the following command? parted /dev/sda unit s print I anticipate that the output from this command will produce the same message. If so then I suggest that your raise this problem on the bug-parted@gnu.org mailing list. This will bring your request to the attention of the parted team.
Yes. Same error message: root@sysresccd /root % parted /dev/sda unit s print Error: Unable to satisfy all constraints on the partition.
OK. Wrote mail to bug-parted@gnu.org
The link to the bug as reported on the bug-parted mailing list is as follows: http://lists.gnu.org/archive/html/bug-parted/2010-04/msg00007.html
A similar problem is reported in bug #624078 - No partitions detected
If this still happens with a modern version of parted, please issue a new report to bug-parted@gnu.org and I'll try to track it down. Since it isn't a bug in gparted, this should be closed as NOTGNOME.
Hi Phillip, When I have seen this error message in the past the problem seemed to arise due to having only one unallocated sector in front of a logical partition. By this I mean there was only 1 unallocated sector between: - the start of the extended partition and the start of a logical partition. - the end of one logical partition and the start of the next logical partition. I agree that the problem originates in the libparted library, which is why the problem was reported to the bug-parted mailing list in comment #3. Curtis
If that is the problem it should be working correctly in parted git head. It's now been almost two years since I committed a fix for that, but the problem at the time was not an unresolved constraints error, but a failure to actually inform the kernel of the new partition table. The patch is also in debian's parted, so this should be working fine on debian/ubuntu at least.
Hi Phillip, I tested on kubuntu 12.04 with parted-2.3-8ubuntu5.1 and any prior problems when there was only one unallocated sector in front of a logical partition do indeed appear to be fixed. Following are the output from my testing with both fdisk and parted # fdisk -l -u /dev/loop0 Disk /dev/loop0: 16 MB, 16777216 bytes 255 heads, 63 sectors/track, 2 cylinders, total 32768 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x0003c065 Device Boot Start End Blocks Id System /dev/loop0p1 63 32129 16033+ 5 Extended /dev/loop0p5 64 32067 16002 83 Linux # parted /dev/loop0 unit s print Model: Loopback device (loop) Disk /dev/loop0: 32768s Sector size (logical/physical): 512B/512B Partition Table: msdos Number Start End Size Type File system Flags 1 63s 32129s 32067s extended 5 64s 32067s 32004s logical I also tested on ubuntu 10.04 with parted-2.2-5ubuntu5.2 and it too worked as expected. When I look at the bug report posted to the bug-parted mailing list in comment #3, I can see that there is additional information and an extra partition 8. I will investigate to see if I can recreate the original partition layout as detailed in the email and then report back here. Curtis
Using the fdisk expert output from the mailing list post in comment #3, I tried recreating a sample sfdisk file which looks like the following: ---- begin sfdisk-layout.txt ----- # partition table of /dev/loop0 unit: sectors /dev/loop0p1 : start= 63, size= 112392, Id=83 /dev/loop0p2 : start= 112455, size= 165662280, Id= 0, bootable /dev/loop0p3 : start= 165774735, size= 322617330, Id=0f /dev/loop0p4 : start= 0, size= 0, Id= 0 /dev/loop0p5 : start= 165774798, size= 86011884, Id=83 /dev/loop0p6 : start= 251786744, size= 4192902, Id=82 /dev/loop0p7 : start= 169967637, size= 134223075, Id=0b /dev/loop0p8 : start= 165774735, size= 98189217, Id=83 ---- end sfdisk-layout.txt ----- If I have performed the recreation correctly, then there is an obvious partition table overlap between partitions 6 and 7 and a clear problem with logical partition 8 beginning at exactly the same sector as extended partition 3. Having said this, this recreated output does not match with the non-expert mode fdisk output from the mailing list post in comment #3. ;-( Hence my attempt at recreating the original partition table is suspect, and I am unable to recreate this problem. As such I am closing this report as NOTGNOME as suggested by Phillip in comment #6. If anyone encounters this problem, please do report it to the bug-parted@gnu.org mailing list as Phillip suggested.