Hi List :) At this time, I have a few hairs left on my head...(I recall a few e-mails claiming this was "easy to do" and "ran fine")...evidently my bacula-fu is weak...
I'm (still) running Bacula Version: 1.38.11 (28 June 2006) with a single, external HP DLT IV VS80 40/80G drive. Platform is FreeBSD6.1 on a P4/256 with *lots* of disk space. I'm trying to get the mtx-changer script (http://www.bacula.org/rel-manual/mtx-changer.txt) to work to provide auto-eject, auto-notify and auto-resume functionality. There may be other, better ways to do this (feel free to suggest), but the mtx-changer seemed to fit the bill. I've been following the instructions on http://www.bacula.org/rel-manual/Autochanger_Support.html. According to the above link - "That each Volume (tape) to be used must be defined in the Catalog and have a Slot number assigned to it so that Bacula knows where the Volume is in the autochanger. This is generally done with the label command. See below for more details. You must pre-label the tapes manually before using them." Q1 - Do I need to do this *before* bacula has been told about the auto-changer? So I label all of my tapes and then "update" the volume manually assigning a slot number (in relation to how they have been listed in the mtx-changer script)? Or, do I make bacula autochanger-aware and *then* label my tapes? (so it prompts me for what slot to use?)...or does it not matter on a well-functioning system? My Storage Daemon's configuration is: # Our pseudo-autochanger #Autochanger { # Name = "DLT-PSEUDO-CHANGER" # Device = HP-DLT-VS80-01 # Changer Device = /dev/nsa0 # Changer Command = "/usr/local/bin/mtx-changer %o %a %S" #} # HP tape drive configuration Device { Name = HP-DLT-VS80-01 Description = "HP DLT VS80 EXTERNAL" Media Type = HPDLT-IV Archive Device = /dev/nsa0 RemovableMedia = yes AutomaticMount = yes AlwaysOpen = yes Label Media = yes Device Type = tape Offline On Unmount = no Hardware End of Medium = no BSF at EOM = yes Backward Space Record = no Fast Forward Space File = no TWO EOF = yes # Autochanger = yes } Elements have been commented out because I keep reverting back to a single-drive operation...Offline On Unmount was commented out because I had to keep walking across the building to re-insert the tape....does the above config look sane? Also, the Director is configured using: Storage { Name = nasbox-sd Address = nasbox1 SDPort = 9103 Password = "m00c0wTheGreat" # Device = DLT-PSEUDO-CHANGER Device = HP-DLT-VS80-01 Media Type = HPDLT-IV # Autochanger = yes } Again, comments applied to revert to single-drive operation. Permissions on /dev/nsa0 are: lrwxr-xr-x 1 root wheel 6 Jan 1 1970 nsa0 -> nsa0.0 crw-rw-rw- 1 root bacula 0, 94 Feb 26 19:55 nsa0.0 As per Bacula documentation, /etc/devfs.conf: own nsa0.0 root:bacula perm nsa0.0 0666 Here's a second problem...when I undo the comments on the above configurations and stop/start fd/sd/dir, tail -f /var/log/mtx.log shows: Tue Feb 27 11:26:59 EAT 2007 : loaded /dev/nsa0 0 mtx-changer: Request loaded, dev /dev/nsa0 This would seem to be a problem (at least for me)...the mtx-changer script does a "stat stor" to implement the "loaded" command - this causes a "chicken-before-the-egg" lock up. Looking at ps waux we see: root 4559 0.0 1.0 3868 1844 ?? Ss 11:26AM 0:00.01 /usr/local/sbin/bacula-fd -u root -g wheel -v -c /usr/local/etc/bacula-fd.conf bacula 4573 0.0 1.2 4540 2176 ?? Ss 11:26AM 0:00.02 /usr/local/sbin/bacula-sd -u bacula -g operator -v -c /usr/local/etc/bacula-sd.conf bacula 4575 0.0 0.6 1652 1092 ?? I 11:26AM 0:00.01 /bin/sh /usr/local/bin/mtx-changer loaded /dev/nsa0 0 bacula 4580 0.0 0.6 1652 1092 ?? I 11:26AM 0:00.00 /bin/sh /usr/local/bin/mtx-changer loaded /dev/nsa0 0 bacula 4582 0.0 1.2 4148 2172 ?? I 11:26AM 0:00.02 /usr/local/sbin/bconsole -c /usr/local/etc/bconsole.conf bacula 4583 0.0 0.5 1512 864 ?? I 11:26AM 0:00.00 /usr/bin/grep /dev/nsa0 bacula 4584 0.0 0.5 1512 860 ?? I 11:26AM 0:00.00 grep ^Device bacula 4585 0.0 0.5 1512 860 ?? I 11:26AM 0:00.00 grep -v not open. bacula 4586 0.0 0.5 1512 860 ?? I 11:26AM 0:00.00 grep -v ERR= bacula 4587 0.0 0.5 1516 864 ?? I 11:26AM 0:00.00 grep -v no Bacula volume is mounted bacula 4588 0.0 0.3 1232 636 ?? I 11:26AM 0:00.00 sed -e s/^.*Volume // bacula 4589 0.0 0.3 1192 596 ?? I 11:26AM 0:00.00 cut -d" -f2 bacula 4597 0.0 1.3 5188 2456 ?? Ss 11:27AM 0:00.05 /usr/local/sbin/bacula-dir -u bacula -g bacula -v -c /usr/local/etc/bacula-dir.conf root 4600 0.0 1.4 5084 2544 p2 RV 11:28AM 0:00.00 grep bacula (csh) Which won't go away...the sd startup is calling mtx-changer which is in turn calling "stat stor" in the console...the storage daemon wont(?) reply to the mtx-changer command until it's got a response to its "loaded" request...?! Currently all volumes (from "list volumes") have an assigned slot of "0" - I realise this means bacula won't consider them. Having manually changed a single volume slot to 1, it didn't seem to make any difference to the above problem. Perhaps I am missing an important directive somewhere? Help! :) Thank you! Steve. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
