Recovering from read errors in Linux

[le_mulot]

For posterity, here is what finally worked for me

1. clean-up the disc with isopropyl alcohol
2. run ddrescue but with the "best recovery settings" from DDRescue GUI which in CLI are,

Code: Select all

ddrescue --no-scrape --retry-passes=2 --sector-size=32  /dev/sr0 brimage mapfile

3. Follow with the disc backup instructions

[kaysee]

In the source code, the "best recovery settings" sets the cluster size (number of sectors per block for large reads), not the sector size (number of bytes per sector, which is 2048 for optical discs).

Line 3163 in the section you indicated is:

Code: Select all

self.cluster_size_choice.SetSelection(3)

cluster_size_choice is described on line 3097:

Code: Select all

#ClusterSize (No. of sectors to copy at a time) option.

The selection array is defined in lines 2969-2970:

Code: Select all

self.cluster_size_choice = wx.Choice(self.panel, -1,                                    choices=['256', 'Default (128)', '64', '32'])

So the SetSelection(3) function points to '32' -- that is, 32 sectors per cluster for large reads during the first phase of ddrescue. That is the default setting in the ddrescue commandline:

Code: Select all

-c sectors
--cluster-size=sectors
    Number of sectors to copy at a time. Defaults to 64 KiB / sector_size. Try smaller values for slow drives. The number of sectors per track (18 or 9) is a good value for floppies.