Date: Tue, 19 Sep 1995 11:50:41 -0500 (CDT) From: Bob Willcox <bob@luke.pmr.com> To: scott@statsci.com Cc: freebsd-ports@freefall.freebsd.org Subject: Re: Using amanda from ports Message-ID: <199509191650.LAA21751@luke.pmr.com> In-Reply-To: <m0suoca-000r3vC@main.statsci.com> from "Scott Blachowicz" at Sep 18, 95 03:25:02 pm
next in thread | previous in thread | raw e-mail | index | archive | help
Scott Blachowicz wrote: > > Bob Willcox <bob@luke.pmr.com> wrote: > > > I am having trouble getting amanda configured and running here. I have > > an Exabyte 8200 tape drive connected to an NCR 53C810 based adapter and > > I was doing exactly the same thing last night... > > > am running -stable as of 9/9. The problem I'm having is getting amanda > > ...but I'm using 2.0.5R. > > > to label a tape (via amlabel). It always complains thusly: > > > #amlabel pmr VOL001 > > rewinding, writing label VOL001amlabel: rewinding tape: Input/output error > > I think I did this > > mt blocksize 32768 > > to avoid that. Whether or not that's the correct thing to do is another > question. Well, I have gotten amlabel to work, though it took a hack to tapefd_rewind() to retry failed rewind attempts. It seems that on my system the first ioctl to the /dev/nrstx (no-rewind) device will fail and the second one works. I put a temp hack in tapefd_rewind() to simply sleep a second then retry the ioctl and that seems to be working (it still fails if I leave out the sleep). Perhaps some day when I have more time I can take a look into why the first one always fails :-( > > > an amcheck evokes the following response: > > > > #amcheck pmr > > Amanda Tape Server Host Check > > ----------------------------- > > /usr8/amanda/work: 906638 KB disk space available, that's plenty. > > ERROR: /dev/nrst1: not an amanda tape. > > (expecting a new tape) > > NOTE: skipping tape-writeable test. > > Server check took 37.242 seconds. > > > > Amanda Backup Client Hosts Check > > -------------------------------- > > WARNING: obiwan: selfcheck request timed out. Host down? > > I got that too and did some poking around. Seems it needs an entry in the > /etc/inetd.conf to fire off the /usr/local/libexec/amanda/amandad daemon. I > think I also added an entry to the /etc/services file for it. I found hints > about the services file entry in the log file that got created from just > running amdump to see what happened. Hmmm...shouldn't the pkg_add process take > care of updating system config files like those? Ok, I have setup /etc/inetd.conf, /etc/services, and the necessary .rhosts file to let this get a bit farther. Now, unfortunately, when I try the amcheck command I get sig 11's in the amandad daemon. If I simply execute the program from the command line it works ok so I suspect that I am still having configuration problems. What userid are others using for the dumpuser? I am currently trying operator. Should I setup bin so that I use it (apparently the amanda default)? > > That said, I haven't quite gotten it to work yet, but I'm further along. I'm > running this on a home system, so there's one system with a few discs to back > up and no space for a holding disk. I haven't figured out if this is > reasonable or even possible yet, but I wanted to see what amanda did. I'll > probably end up punting back to dump or tar or cpio or whatever given the > small scale of things here. Though I have quite a bit more I'm trying to backup, I too may simply stay with dump/restore if things don't get better soon. :-( > > Hope this helps a little... Thanks, it did... > Scott Blachowicz Ph: 206/283-8802x240 StatSci, a div of MathSoft, Inc. > 1700 Westlake Ave N #500 > scott@statsci.com Seattle, WA USA 98109 > Scott.Blachowicz@seaslug.org > > -- Bob Willcox bob@luke.pmr.com (or obiwan%bob@uunet.uu.net) Austin, TX
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199509191650.LAA21751>