-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On 03/27/2014 06:48 PM, Brian C. Lane wrote: > Tried that (with my modified code) and it works fine. The header > munge is changing the PE count to 9 from 128 and re-calculating the > checksums. > > I went through the code -- when we rewrite the headers all the > sizes and references are updated, in the pmbr and in both headers.
Ok, now I'm not seeing that odd 2g problem. Maybe before I was running it on a 32 bit machine? Anyhow, the problem is that gpt-header-munge shrinks the table but doesn't update LastUsableLba to agree. It should do that, but I suppose we could put the two tests together. If the backup is *either* at the last sector, *or* the right location relative to LastUsableLba, then it's in the right place. Following up with a patch to do this. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.14 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBCgAGBQJTNLyMAAoJEI5FoCIzSKrw7DAH/j8v8CxDVlRNGsOVYEs7v9hp LfXsiKEvV/shIa89IOjOZCRYiBESNxNBoRqhlgujrlCKhlB0GCHTsZwdWxQy3nst /goTdl4gFwKZ/oCZ44SGM9jNpy2eTpwodgPZOIejerRq0a+mzbc5g5PL7xdpreai jZMKNWTHzxm6ruQt75Img8YyZO7rrxPyOljAB8L6PV2Rf7M4DE6hTkWzZ7sM1PFu GpGAwHfk7jFBmGEPylg9uC78eP60/swp+aXeSHmOcnylT/ZaPQEkvi1nK2uEdWa0 r5wRkIu976djJ2QTAfgC81SdMJjf4ZFIu6kL6zLPxlAS7xkkgrD6pNEuk+A91VE= =FfJR -----END PGP SIGNATURE-----
