Looks like my information was old. Version 4 couldn't do it (even after they claimed it would); version 5 apparently handles it well. Guess I stopped trying it after version 4 yelled at me.
Another source <https://www.finetunedmac.com/forums/ubbthreads.php?ubb=showflat&Number=25692&Searchpage=1&Main=3008&Words=%22DiskWarrior+and+Time+Machine%22&Search=true#Post25692> says: > SuperDuper will make a file-level copy, handling multi-linked files > correctly, and will produce a copy without catalog damage even if the source > has catalog damage. (But if the source catalog is damaged, the copy may not > finish, and even if it finishes the new copy may have other non-catalog > inconsistencies that make it unusable.) You may want to try that, especially since I think it has a free trial. > On Jan 23, 2020, at 4:21 PM, Matt Penna <[email protected]> wrote: > > I’m curious about this because DiskWarrior works fine for me on Time Machine > volumes (but I’ve never tried it on a Time Machine sparsebundle). > > The only trouble I had with DiskWarrior on Time Machine drives was when it > was a 32-bit app and could not allocate enough memory to hold all the file > system structures—a showstopper with my Time Machine drive that at the time > had a 17GB B-tree and that definitely would not fit into the 4GB 32-bit RAM > limit. > > Since going 64-bit in late 2014, DiskWarrior should work on all Time Machine > volumes. > > In an ironic reversal with modern Macs, DiskWarrior can now ONLY work on Time > Machine drives; APFS-formatted drives drives are not supported with current > versions of DiskWarrior and Time Machine drives are still HFS+. (Alsoft says > DiskWarrior APFS support is coming soon, now that technical specs for the > file system have been finalized, though APFS is ostensibly robust enough to > not need much fixing. I know, I know…) > > Matt > >> On Jan 23, 2020, at 1:39 PM, Macs R We <[email protected]> wrote: >> >> Although all of this is true, as it turns out, I know of only one disk >> management tool that will even attempt to repair a Time Machine volume, >> because the volume architecture is so baroque -- and that is Disk Utility. >> If you try to repair the volume with Disk Warrior, it will flat out tell >> you, "this is a Time Machine volume, and I don't do those." You have to use >> only Disk Utility, not fsck. If DU can't solve the problem, it can't be >> solved. >> >>> On Jan 23, 2020, at 11:31 AM, Matt Penna <[email protected]> wrote: >>> >>> On Jan 23, 2020, at 1:25 PM, Michael <[email protected]> wrote: >>>> >>>>> On 2020-01-23, at 10:23 AM, Matt Penna <[email protected]> wrote: >>>>> >>>>> If you have DiskWarrior, I would give that a try. I believe it works on >>>>> disk images and sparsebundles. >>>>> >>>>> Matt >>>> >>>> I do not have disk warrior. >>> >>> Unfortunately, whenever a drive or image is not reparable like this, I’ve >>> always had to resort to a 3rd-party tool to fix the file system; the >>> built-in tools are not very robust. Perhaps someone else will have another >>> suggestion. >>> >>> It’s always been baffling to me that 3rd parties write better fix-it tools >>> than the people who write the OSes. This has been a problem for over 30 >>> years and it’s still as unsolved as ever, even if the file systems are a >>> lot less fragile than they used to be. >>> >>> Matt >>> _______________________________________________ >>> MacOSX-talk mailing list >>> [email protected] >>> https://www.omnigroup.com/mailman/listinfo/macosx-talk >> >> _______________________________________________ >> MacOSX-talk mailing list >> [email protected] >> https://www.omnigroup.com/mailman/listinfo/macosx-talk > > _______________________________________________ > MacOSX-talk mailing list > [email protected] > https://www.omnigroup.com/mailman/listinfo/macosx-talk -- Macs R We -- Personal Macintosh Service and Support in the Wickenburg and far Northwest Valley Areas. http://macsrwe.com
_______________________________________________ MacOSX-talk mailing list [email protected] https://www.omnigroup.com/mailman/listinfo/macosx-talk
