Hi,
On Fri, Sep 07, 2001 at 01:53:29PM -0800, Anthony Valentine wrote:
Hello everybody!
I've got Amanda up and running and she's doing great! The only thing that I
have left to figure out is how to use the cleaning tape.
Can anyone tell me what files and/or scripts control how and when
John R. Jackson wrote:
FAILURE AND STRANGE DUMP SUMMARY:
office.pss /home/robert lev 0 FAILED [disk /home/robert offline on
office.pss.local?]
Waht does this mean?
Did you look through the FAQ at www.amanda.org?
Yes, I did. There was nothing about such message.
What version of
On Mon, 10 Sep 2001 at 12:58pm, Robert Vetter wrote
John R. Jackson wrote:
Did you look through the FAQ at www.amanda.org?
Yes, I did. There was nothing about such message.
http://amanda.sourceforge.net/fom-serve/cache/10.html
What version of Amanda?
1:2.4.1p1-12
That looks like an
Question for both of you (Gene and Thomas) ???
What about my scsitapedev if i don't have On my system the tape is /dev/nst0,
and the changer is either /dev/sg1 ???
Here is the result of my dev :
ls -al n* |more
lrwxrwxrwx 1 root root 8 jan 26 2000 nrst12 - rmt/0lbn
I am also having the same problem, but it is spread across all of my
machines. It does appear to be concentrated on the larger partition, but not
always. I am using dump universally on all machines, and have compiled
versions installed on all clients except for the two 7.1 machines where I
used
I am running Solaris 8, with Amanda 2.4.2p2
I have a question, when I take my drive offline mt -f /dev/rmt/0lbn
offline I'm unable to bring it back online, unless I physically press the
tape load on the changer.
I've tried using amtape, and mt to load a tape from the cartridge. But there
must
I had this same problem, particularly with the RPMs that come with 7.1. The
solution is either to apply the advfs.patch from
http://www.amanda.org/patches.html, or alternatively, use CVS and get the
latest greatest. Seems to fix many things, this one of them...
HTH,
Sean Noonan.
Hi there.
I'm having a problem getting Amanda to backup one of our servers. When
running amcheck I get the following error:
ERROR: nushtel: [access as operator not allowed from operator@backupservername]
I found the faq-o-matic entry for this issue and checked all of my
settings against the
Hi,
On Mon, Sep 10, 2001 at 08:46:52AM -0500, Gene Heskett wrote:
That was my next question Thomas. What do I set 'scsitapedev' to?
On my system the tape is /dev/nst0, and the changer is either /dev/sg1
or /dev/sg2 (I've got an ATAPI cd-writer too)
Hmm i can't give an clear answer, but if
Hi,
On Mon, Sep 10, 2001 at 11:12:18AM -0400, Eric Chamberland wrote:
[..]
I'm on Solaris 7 and i have a Storedge L20 (robot) with DLT-8000 (Dlttape IV)
So i have one tape on /dev/rmt/1...
and the other tape on /dev/rmt/2...
OK, on solaris you don't have the sg device as on linux. First
thank's for your help Thomas !!!
How can i know if chg-scsi is running on my system ???
wish chg- do i need to use (chg-manual, chg-multi, chg-scsi)
i think it will be chg-scsi because my robot have a medium changer scsi id 0
and the drive #1 is scsi id 1 and drive #2 is scsi id 2
i'm i
Hi,
On Mon, Sep 10, 2001 at 02:58:58PM -0400, Eric Chamberland wrote:
thank's for your help Thomas !!!
How can i know if chg-scsi is running on my system ???
You should get the amanda-2.4.2p2 version. Check www.sourceforge.net
for amanda. After unpacking you will see an docs directory, and
Thanks Sean,
Looks like it is time to bite the bullet and get rid of the RPM versions. I
will let you know if that fixes the problem.
markh
-Original Message-
From: Noonan, Mr. Sean P. [mailto:[EMAIL PROTECTED]]
Sent: Monday, September 10, 2001 10:25 AM
To: 'Mark Holm'; 'Joshua
Looking at /tmp/amanda/amandad.debug on the server I'm trying to backup I
see somethings that concern me, namely:
the BUILT_MACH line lists the wrong hostname ...
That's just documentation about where Amanda was built. It's not used
for anything.
... The other bit that bugs me is that it
I'm getting a weird error on one of my amanda clients (2.4.1p1 on
freebsd 4.2 using gnu tar v 1.13) ...
I hope you don't really mean 1.13 for tar. If it's not 1.13.19 or later
(from alpha.gnu.org), you're not getting good index files.
? sendbackup: index tee cannot write [Broken pipe]
This is
When I attempt to label tapes, I get:
$ amlabel Nodewarrior JZ1
rewinding, reading labelSegmentation fault - core dumped
Have you done a cvs update recently and rebuilt/reinstalled?
There is indeed a corefile in /tmp/amanda (which I can send to anyone
who wants to look at it). ...
The core
I am using amanda 4.2P2 and would like to be able to influence the order
files are backed up to tape. I would prefer largest files first. ...
... Is there some way of configuring the planner to do this?
You need to change driver, not planner. The entries are currently sorted
by estimated
My sources are a day old ... is that new enough?
Yes. Nothing has changed for several days.
#0 0x8056d75 in tapefd_rdlabel (fd=3, datestamp=0xbfbfdcb8,
label=0xbfbfdcbc)
at tapeio.c:775
Everything looks OK to me, and I'm not sure why it's saying it died at
the very start of the routine.
John R. Jackson writes:
Have you done a cvs update recently and rebuilt/reinstalled?
My sources are a day old ... is that new enough?
I'll update now and rebuild.
gdb /path/to/amlabel /path/to/core
where
quit
Sure.
Program terminated with signal 11, Segmentation fault.
On Mon, 10 Sep 2001, John R. Jackson wrote:
I am using amanda 4.2P2 and would like to be able to influence the order
files are backed up to tape. I would prefer largest files first. ...
Note that such a change is likely to significantly reduce the amount of
parallelism built into Amanda
On Mon, Sep 10, 2001 at 03:38:56PM -0500, John R. Jackson wrote:
I am using amanda 4.2P2 and would like to be able to influence the order
files are backed up to tape. I would prefer largest files first. ...
... Is there some way of configuring the planner to do this?
You need to change
I said 40 wasn't a big disklist, but 40 is probably big in terms
of number of entries for a single client. Could this be where the
problem is coming from?
Possibly, especially if there are other options involved, such as a
lengthy exclusion file name. The only way to know for sure is to look
On Mon, 10 Sep 2001, John R. Jackson wrote:
What's in ~operator/.amandahosts? It should be:
backupservername operator
Here's what's in .amandahosts:
console.corecom.net operator
console is the name of the backupserver, here is the exact verbage of the
amcheck error:
ERROR: nushtel:
I have a question, when I take my drive offline mt -f /dev/rmt/0lbn
offline I'm unable to bring it back online, unless I physically press the
tape load on the changer.
That's perfectly normal. It (usually) takes physical movement (as in,
inserting the tape in the drive) to get the drive to go
Here's what's in .amandahosts: ...
OK, that looks right.
We are using version 2.4.1p1. ...
OK, that version does not report anything else useful.
What does the following say:
ls -ld ~operator/.amandahosts
ls -ld ~operator
ls -ld ~operator/..
ls -ld ~operator/../..
John R. Jackson,
I keep seeing recommendations to use GNUTar rather than the native dump
programs. ...
Not from me you don't :-).
Can somebody give a quick pluses and minuses of going either way.
System dump/restore is highly OS (actually, file system format) dependent.
There is very little chance you could
OK, then here are a couple of other things to try. Is the .amandahosts
file being accessed when you do amcheck? For instance, does ls -lu
show a time change?
Note that ls only shows things to the minute, so if you have some
other utility that shows more accurate atime output, that would be
On Mon, 10 Sep 2001, John R. Jackson wrote:
What does the following say:
ls -ld ~operator/.amandahosts
-r 1 operator operator 28 Sep 10 14:14 /usr/local/amanda/.amandahosts
ls -ld ~operator
drwxrwxr-x 2 operator wheel 512 Sep 10 14:14 /usr/local/amanda
ls -ld
28 matches
Mail list logo
