Hi Thomas, The problem is obviously that there is no real correct partition type for an ISO 9660 image inside a partition. The “0xCD” type was a creative idea given the CD-ROM heritage, but not standard. We haven't seen problems with the ISO image from pre-UEFI FAI versions that used this ‘cd’ partition type. But with the new format image, it looks like setting just *any* partition type on the first partition causes Windows to become confused and wanting to wipe the disk entirely (it prompts to do this when the device is inserted). With the 0 type, at least the risk is a bit smaller that the USB drive will be wiped when someone plugs it in a Windows PC.
So, leaving the type at 0 is probably the best way to go for the FAI installer in general, despite the risk that the partition is lost when trying to manipulate the table in Windows and maybe other tools. For our specific case where we want to have an extra writable partition on the same stick, it looks like the best solution is to already append a small NTFS partition to the image we distribute. This partition can then be grown using the Windows tool if desired. Unlike when adding a partition, the grow operation does not destroy the first partition (at least not in Win 7, I only tested it there). To be really sure, we can still set some partition type on the first partition. This no longer causes problems in Windows if there already is a third partition. Regards, Alexander On Wed, 27 Jan 2021 at 19:23, Thomas Lange <la...@cs.uni-koeln.de> wrote: > > Hi Alexander, > > I've checked several other linux ISO images (ubuntu,grml, Centos, > Debian, archlinux) and only archlinux does not use type 0 for the > first partition on their ISOs. But archlinux set a flag called > hidden. Fedora use type 0 but with parted I canot set the partition, > cfdisk shows the partition. > > So, there's some magic in all this. In fai-cd we call xorriso with > -iso_mbr_part_type 00. I do not like to change this without knowing > it will not cause other problems. Can you find some information that > convice me that a different type is correct? > -- > regards Thomas
