I imaged both drives and after much fiddling got a vm of Ubuntu Server up and running. I would recommend using the first .VDI as the boot drive. Otherwise you run into issues with booting because virtualbox will try to boot off the first drive created. There is a way to change the boot order of the drives in virtualbox if you don't do this, but easier to plan ahead.
If you installed server without the raid drives attached like I did, mdadm will still try to read the array. In my case, showed that my array was located at /dev/md127.
In my case mdadm said that 1 of two drives was functional, so it would seem that it should be possible to image the minimum number of drives for a functional array and still be able to access the array.
Made a directory called salvage in /home/username and mounted /dev/md127 there. This let me browse the RAID 1 array contents.
Installed subversion and apache 2 for good measure.
Code:
mkdir /home/username/USB_Storage
Mounted a flash drive to /home/username/USB_Storage.
I have run an
Code:
sudo svnadmin dump /repo/path > dumpfile.dump
successfully.
I used
Code:
sudo svnadmin create /var/lib/svn/repo
sudo chown -R www-data:www-data /var/lib/svn/repo
sudo chmod -R 775 /var/lib/svn/repo
to create a repository on the new host machine.
Using
Code:
sudo svnadmin load /var/lib/svn/repo < dumpfile.dump
fails at revision 75 out of 586 due to a checksum error.
Does anyone know how to fix this either in the dumpfile or in the original repo before I generate the dumpfile?
So far I've found recommendations to use something like
Code:
sed -i '/Text-copy-source-md5/d' your.dmp
but with no explanations about what sed is or how to use it.
Bookmarks