Hi all: I've got an amanda install where I have a dedicated amanda server w/ holding disk and tape drive, and a FreeNAS storage appliance with capacity of 145TB (about 50TB on shares that get amanda backups applied). To carry out the backup on the FreeNAS device, I have a "FreeNAS Jail" running Amanda client with the FreeNAS storage tree mounted inside the jail via the FreeNAS Jail system (eg, it is NOT mounted from a device; in fact, inside the jail it doesn't even look like its mounted, its just "there").
Everything was working fine. Then recently I had to restart the Amanda client jail on the FreeNAS. Since then, all level 1's have been coming out the same size as a level 0, which totally overwhelms the amanda system, and I end up with failed backups and a big fat mess. I've read the website on the amanda site about changing disk IDs, but I did a fairly thorough checking of that, and I don't think that's it. First, there is literally NO device on the mount point: root@AmandaBackup:/usr/local # mount cys-bkup/iocage/jails/AmandaBackup/root on / (zfs, local, nfsv4acls) root@AmandaBackup:/usr/local # And that stays exactly the same. I've found and ran the tar-snapshot-edit perl script in read mode to view the device IDs. I am seeing a few different device IDs show up, but the level 0, level 1 and leve1.new for any given share always have the same device ID. Here's a snippit of the output: File: amclient-tdriveCYS-2018-Session-A-B_0 Detected snapshot file version: 2 Device 0x2900ff0b occurs 6305 times. File: amclient-tdriveCYS-2018-Session-A-B_1 Detected snapshot file version: 2 Device 0x2900ff0b occurs 6306 times. File: amclient-tdriveCYS-2018-Session-A-B_1.new Detected snapshot file version: 2 Device 0x2900ff0b occurs 6305 times. File: amclient-tdriveCYS-2018-Session-C_0 Detected snapshot file version: 2 Device 0x2900ff0b occurs 3737 times. File: amclient-tdriveCYS-2018-Session-C_1 Detected snapshot file version: 2 Device 0x2900ff0b occurs 3737 times. File: amclient-tdriveCYS-2018-Session-C_1.new Detected snapshot file version: 2 Device 0x2900ff0b occurs 3737 times. .... File: amclient-tdriveCYS-2020-Session-S_2 Detected snapshot file version: 2 Device 0x2900ff06 occurs 2424 times. File: amclient-tdriveCYS-2020-Session-T-Z_0 Detected snapshot file version: 2 Device 0x2900ff06 occurs 3717 times. File: amclient-tdriveCYS-2020-Session-T-Z_1 Detected snapshot file version: 2 Device 0x2900ff06 occurs 3717 times. File: amclient-tdriveCYS-2020-Session-T-Z_1.new Detected snapshot file version: 2 Device 0x2900ff06 occurs 3717 times. File: amclient-tdriveCYS-2020-Session-T-Z_2 Detected snapshot file version: 2 Device 0x2900ff06 occurs 3717 times. File: amclient-tdriveCYS-2020-Session-rest_0 Detected snapshot file version: 2 Device 0x2900ff06 occurs 1 times. File: amclient-tdriveCYS-2020-Session-rest_1 Detected snapshot file version: 2 Device 0x2900ff06 occurs 1 times. File: amclient-tdriveCYS-General_0 Detected snapshot file version: 2 Device 0x2900ff06 occurs 9390 times. File: amclient-tdriveCYS-General_1 Detected snapshot file version: 2 Device 0x2900ff06 occurs 9390 times. File: amclient-tdriveCYS-General_2 Detected snapshot file version: 2 Device 0x2900ff06 occurs 9095 times. File: amclient-tdriveCreativeImage_0 Detected snapshot file version: 2 Device 0x2900ff06 occurs 14114 times. File: amclient-tdriveCreativeImage_1 Detected snapshot file version: 2 Device 0x2900ff06 occurs 14114 times. I've noticed that all affected devices will end up with a .new. I think the last time this happened, my recovery path was to move/rename the files on the client, force all DLEs to full on next run, then spend the next two weeks getting the fulls out of the way before I can get back to some semblance of a schedule. Due to the volume of data, this is more than a minor inconvieninece. Before I go forward with that strategy this time, can anyone provide more insight into what is going wrong and if there's any "clean" way to fix it without having to force new level 0's for everything. Thanks! --Jim
