xals
commented
7 years ago Hi @starchy,
Actually, this repository is a fork of http://git.ganeti.org/?p=instance-debootstrap.git;a=summary. I should have mentioned it. I created it to be able to have a separate partition for /var, and swap. And it is quite outdated, I forked it on version 0.14 and upstream is now 0.16.
And I never tested the export and import mechanism so I can't help you on this issue.
I think the best way to have information on this issue is to send an email to the Ganeti Google Group here: https://groups.google.com/forum/#!forum/ganeti
I'm working on using gnt-backup to export some plain LVM instances from a single-node cluster, copy the files to a brand new single-node plain LVM (for now) cluster, and import them. On either the source or destination node, the import operation results in the same error partway through importing the disk. After setting MAX_RECENT_OUTPUT_LINES to 200, I can see that sfdisk is bombing out:
Fri Jan 6 10:30:35 2017 disk/0 failed to receive data: Exited with status 1 (recent output: sfdisk: Checking that no-one is using this disk right now ... sfdisk: BLKRRPART: Invalid argument sfdisk: OK sfdisk: /dev/rootvg/fe8efe28-1d3b-4cc7-8184-05d550062c62.disk0: unrecognized partition table type sfdisk: No partitions found sfdisk: BLKRRPART: Invalid argument sfdisk: If you created or changed a DOS partition, /dev/foo7, say, then use dd(1) to zero the first 512 bytes: dd if=/dev/zero of=/dev/foo7 bs=512 count=1 (See fdisk(8).) restore: cannot open /dev/tty: No such device or address restore: ./lost+found: File exists buffer overflow detected : restore terminated
I'm not sure if this is the right place to report the issue, but I haven't had any luck figuring out how gnt-backup calls ganeti-instance-debootstrap or using its import script, which doesn't seem to be documented for manual use.