Opened 8 years ago

Last modified 13 months ago

#6452 new bug

[DriveSetup] Corrupts MBR on some disks over 250GB.

Reported by: davidsaunders Owned by: stippi
Priority: normal Milestone: Unscheduled
Component: Applications/DriveSetup Version: R1/Development
Keywords: MBR SATA Cc:
Blocked By: Blocking:
Has a Patch: no Platform: All

Description

Often times after updating the partation table using DriveSetup, the last byte of the MBR (Root sector) is changed, causing it to become non-bootable on systems in witch the BIOS checks for the boot flag (Last two bytes in a Boot or Root sector). I am only able to reproduce this on some systems with a SATA HDD greater than 524288000-Sectors = 262144000KB = 250GB in size. The change seems to be random in nature, and this has been tested with multiple builds over the last 3 months.

Change History (12)

comment:1 Changed 8 years ago by davidsaunders

To use my system, when I encounter this error, I boot from a Haiku CD and use DD to make a copy of the Root sector from the raw device, copy the partition table into a known good root sector and use DD to write it back into the root sector.

comment:2 Changed 8 years ago by stargatefan

this however does not fix the problem.

comment:15 Changed 7 years ago by Premislaus

The problem still occurs. Partitions can be easily repaired with Hiren Boot CD. You need to use program DiskGenius, looking to options check\repair partition tables, give the OK, and then click save all and that's all :-).

How do you boot are the CD, choose the first option to load Windows from the hard disk.

Sorry for my english, i am not a native speaker.

Best regards!

comment:16 Changed 7 years ago by Premislaus

Still occur on hrev44061 gcc4h.

comment:17 Changed 4 years ago by jfjhh

This is still occurring on the nighties after hrev48449 gcc2

comment:18 Changed 4 years ago by pulkomandy

Summary: [DriveSetup] Currupts MBR on some disks over 250GB.[DriveSetup] Corrupts MBR on some disks over 250GB.

comment:19 Changed 4 years ago by waddlesplash

Milestone: R1R1/beta1
Platform: x86All
Priority: normalcritical
Version: R1/alpha2R1/Development

Changing milestone/version/priority according to reports in this ticket and on IRC. Corrupting end-users' hard drives if they have more than just Haiku installed is not something we want to have happen, and >250GB SATA drives are extremely common these days.

comment:20 Changed 3 years ago by pulkomandy

can somone provide an example of a corrupted bootrecord and a copy of the fixed version (after running hiren or some similar tool)?

comment:21 Changed 3 years ago by axeld

Would also be nice to have a detailed description of how to reproduce the issue.

comment:22 Changed 3 years ago by kallisti5

I wonder if this was related to the "GPT leftovers at end of disk confusing our partitioning systems" issue?

I haven't seen it on clean drives.

comment:23 Changed 2 years ago by kallisti5

Priority: criticalhigh

I've haven't seen any issues MBR partitioning disks 1TiB or more under Haiku in the last few years.

We really need more information to troubleshoot this one. Could an impacted user re-test under the latest nightlies and let us know their results?

hrev50577 was a recent change that could solve this issue if it was still ongoing.

comment:24 Changed 13 months ago by waddlesplash

Milestone: R1/beta1Unscheduled
Priority: highnormal

Nobody's seen this one in some time; moving out of B1 and downing prio.

Note: See TracTickets for help on using tickets.