the change time to be altered.
Our workaround was to use a GNU tar wrapper (2000+ lines of finely crafted
Perl) for LVM volumes that created a snapshot and backed that up. We just
ignored the access time problem for file systems not under LVM control.
Toomas Aas
John R. Jackson, Senior Systems
.
tape P02008/RCAC0 or a new tape.
Scott...
John R. Jackson, Senior Systems Analyst, Engineering Solutions, Inc
.
What kind of OS is being dumped and what kind of dump (e.g. ufsdump,
vxdump, etc)? Some versions of dump just don't do ACL's. Vendors want
you to buy their spiffy proprietary backup software that actually works
and leave the normal stuff crippled. Wonderful folks, those vendors.
Dave
John R
) and it will try very hard to reduce the amount
dumped to less than runtapes * tapelength. So reducing runtapes from 2
to 1 does not just affect what happens if you hit EOT, it also affects
the schedule planner puts together.
Guy Dallaire
Jon H. LaBadie
John R. Jackson, Senior Systems Analyst, Engineering
from amrecover and/or GNU tar?
What version of GNU tar are you using (tar --version)?
The amrecover.*.debug file in the Amanda temp directory might also help.
Chuck Amadi
John R. Jackson, Senior Systems Analyst, Engineering Solutions, Inc
did.
Just FYI, I don't find it worth the trouble to build just a client.
You're not going to save much space or time, and a couple of the server
tools are useful on the clients at times (which probably means we need
to shift them to a common area).
David
John R. Jackson, Senior Systems Analyst
Looks like some temporary compressed index files from the current run
should be there but are not. ...
Do you see similar behavior ?
As Joshua said, this is to be expected. Amanda is creating those files
and removing them during the time GNU tar is also backing them up.
You *might* (haven't
In amandad.debug on the client I get:
...
amandad: time 2951.811: sending REP packet:
Amanda 2.4 REP HANDLE 00F-80930508 SEQ 1126652514
OPTIONS features=feff9ffe7f;
ar0s1a 0 SIZE 30567356
ar0s1a 1 SIZE 17959269
amandad: time 2961.819: dgram_recv: timeout after 10 seconds
amandad:
mt -f /dev/nst0 compression off
[by the way, the argument off is worth a small fortune in
consulting fees, since it has to be guessed]
Yeah, I've been down that path, too. In my case, I wanted to turn
compression *on* and absolutely nothing I tried worked. And you're
absolutely right that the
...
how about:
The next 2 tapes Amanda expects to use are: ncti16, ncti17.
Fixed. Thanks.
Scott...
JJ
with whatever
you're running.
Pete
John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
up a
very good point about the cron PATH possibly being different than your
interactive PATH.
Michael Bretterklieber
John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
list entries
need to be relative to the thing being backed up. Since you are backing
up /home/max, the entries should be:
./o/mprj/PDF
./o/mprj/NMT
and so on.
Konstantin.
John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
after one hour.
Is any suggestion?
We would need to see the amdump.NN file that goes along with the
failed runs. There is detailed information in there about what each of
the process is doing.
It would also help to know what version of Amanda you're using and on
what OS.
David
John R. Jackson
, can I specify directories
with spaces? So I can exclude /Program Files?
Yes, you should be able to put whatever you need to in an exclusion file.
But, just like with include, you can't use exclude file because of
the protocol issue.
Ricky
John R. Jackson, Technical Software Specialist, ITaP/RCS
potential
restore that much easier.
Frank
John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
(*) Yes, I know some of you see odd balancing going on. No, I don't
know why it does that (without getting my PhD in Amanda balancing
algorithm analysis :-).
full+incremental on the
holding disk).
Adam
John R. Jackson, Technical Software Specialist, ITaP/RCS, [EMAIL PROTECTED]
amtapetype
to think those extra file marks in pass 2 actually took up space.
It also explains why amtapetype used to sometimes report a negative file
mark size if the math happened to end up that way. When that happens
now we just report it as zero.
Gene
John R. Jackson, Technical Software Specialist
the script decided the drive slot was
1 (I'm guessing here a bit, but think this is correct). But (I think)
that's wrong. It should be 0. So try putting:
driveslot=0
in your changer config file and see if that helps.
Brian
John R. Jackson
all the space currently available minus 10 MBytes
(for slop). That way, if the amount available changes (you grow or
shrink the partition, some dumps are left over from a previous run,
etc), you don't have to change the holdingdisk setting in amanda.conf.
Geoff
Frank
John R. Jackson, Technical
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
to that point.
Not to mention not knowing which block to make the parity block if you
fsf and then start writing more data.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
. If you're using
tapes that hold 20 GBytes, they will only be half utilized since an
error on any device during RAIT processing closes all of the devices.
Does that help?
Milos
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
(*) During debugging of the RAIT code (actually
to the various
drivers (and the corresponding *aioread() and *aiowait() routines).
That's not a terribly big deal, but it's not the trivial case of just
adding a few lines in output-rait.c that one might think.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
reference more than one in a real dumptype -- the entries will be
merged into a single large list.
As I said, I haven't even faintly tried this. It's just an idea based
on how I'm pretty sure this all works.
Leonid Mamtchenkov
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
for incrementals).
Or it could simply be the estimates exceed the available holding disk
space, even when it's empty. If you tell Amanda it can use 100 MBytes and
the estimate is 1 GByte, it's going to go direct to tape no matter what.
Jon H. LaBadie
John R. Jackson, Technical Software Specialist
, but if you
can't, just post the details and someone here will help you through it.
Make sure you mention what version of Amanda you're using, what platform
you're running it on, and the **exact** text of the error message(s)
you're seeing.
John R. Jackson, Technical Software Specialist, [EMAIL
Also, the alternate port patch I was talking about allows you to pick
the port in the amanda.conf file -- no rebuild is needed (except once
to apply the patch).
Brian
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
directory tree names. You almost had it right
with the last one, except you used lvol19 and I'm betting that should
have been lvol/19.
Put a different way, use the same thing you have in your disklist.
That's where the name comes from.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
the amrecover one.
* Make sure the VXDUMP value matches the path you see in amverify
**exactly**.
* If you run strings /path/to/amrecover | grep vxrestore, what do
you get back?
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
it in isolation, they may as
well not be using it.
I guess we'll have to agree to disagree about this :-).
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
configuration files.
Just the program binaries.
Dalton
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
would
automatically advance to EOF. This can be dealt with, but, as I said,
it will take some relatively advanced shell programming (although not
difficult in something like Perl).
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
).
Martin Öhler
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
checks and it still does not work, post back here again with
what you tried and the **exact** results you saw. I'm sure we can get
it working.
David
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
tape *drives*
and want to emulate a changer by having Amanda move from drive to drive.
As others have said, chg-zd-mtx or chg-scsi are probably what you want.
But I've been wrong before :((
I wasn't going to bring that up (just kidding :-).
Jon H. LaBadie [EMAIL PROTECTED]
John R
a full
cycle again. For instance, if your dumpcycle is 7 days and DailySet1-04
is within 7 days of the top of the tapelist file, putting it at the end
(and overwriting it on the next run) may make it impossible to do a full
restore if it contained the latest level.
dk
John R. Jackson, Technical
change that should be easy to tweak to fit.
Frank
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
if
things go wrong and be prepared to help us diagnose what happened.
Thanks.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
got, you need:
backup amanda
Note that the backup host name is not qualified. That, in turn, implies
some other things (see below), but maybe this will get you going.
Simon.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
And now for the more than you ever wanted to know
if amanda uses more than one tape in a single run? Is there a
nice way to have it eject only the tapes it used?
See above.
Simon.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
backup get archived
- back up the control file
The files are written to a large separate partition that is what Amanda
is actually told to back up.
The scripts, etc, are here:
ftp://gandalf.cc.purdue.edu/pub/amanda/dbbackup.*
Aline
John R. Jackson, Technical Software Specialist, [EMAIL
-22.html
VU#370308 - http://www.kb.cert.org/vuls/id/370308
http://www.microsoft.com/technet/treeview/default.asp?url=/technet/security/bulletin/MS02-039.asp
Cheers, Gene
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
and at the very end things
will be truncated, but it could also be some other silliness going on.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
:-) the problem. Give the following
patch a try.
--david
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
tapeio.diff
Description: tapeio.diff
.
Note, however, my *strong* suggestion that you make a copy of the file
first. If I'm wrong, I can still say I told you so :-).
Jon H. LaBadie
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
clients
Probably :-).
Oh, you want details. In theory the feature information capability
will limit such conflicts. I don't know of any others, and even this
one makes a best attempt at being benign.
Toralf Lund
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
parts.
David
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
amandad.20030108134258.debug
...
You shouldn't need the old --with-pid-debug-files flag any more.
Right?
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
this server:
%server% /var lev 0 FAILED [[parse of reply message failed]]
So you're running some variant of 2.4.2? And are you certain your *inetd
line is firing off the matching amandad?
What's in amandad*debug? How about sendbackup*debug?
John R. Jackson, Technical Software Specialist
limits on the size of a command line
or number of args, you could bang into that, and the symptoms may not be
obvious. Always check your backup system after a significant change (and
once in a while just for the heck of it :-) with some test restores, etc.
Will
Jon H. LaBadie
John R. Jackson
is (it will now be logged in the same message)?
A quick glance at the 2.4.2 code seems to indicate it's doing the exact
same thing as 2.4.3, so I'm not sure why there would be a problem (not
saying there isn't -- just don't get it yet).
--david
John R. Jackson, Technical Software Specialist, [EMAIL
** tape :-):
dd if=/dev/zero bs=32k count=10 of=/dev/whatever
If this works (and I'm betting it does not with your current setup),
then Amanda will probably be happy.
Martin Hepworth
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
estimated size helps that (the default
is 1 GByte, so the default file size for the first pass is 10 MBytes).
But it is perfectly OK to just ignore that parameter.
Axel
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
that the selfcheck portion of amcheck that runs on the
client is *not* running as root (neither is amcheck except right at
the very beginning). So it could be something like what you've said,
but it's hard to know for sure.
David
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
-no-bump HOST DISK
That will cause it to repeat the same level it did last time (or it
might jump back to a full dump, which is OK, too).
Alex
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
?
What's in it?
Mozzi
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
of the moon, etc.
So the little C programs are a better check when trying to figure
out Amanda lookup problems because they do exactly what Amanda does.
That's why I wrote them :-).
John Oliver
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
.
They should be in your Amanda debug directory, e.g. /tmp/amanda. You
should be able to find out with:
amadmin xx version | egrep 'AMANDA_DBGDIR|AMANDA_TMPDIR'
We really need to find them and run a debugger to get a traceback to
figure this out.
David
John R. Jackson, Technical Software
tape and just add it to the run, right?
Yup. If the old tape has valid data on it, you might run amrmtape
to clean out that information. Or just keep it around until the tapes
cycle and get reused.
-Josh
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Amflush has stopped working recently. ...
Not sure if you saw it, but a patch was posted recently for this.
Look for amflush hangs in the archive.
Bill Hults
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
in their respective drives, or can amanda figure out
data/parity just based upon labels/configuration.)
You'll have to give the drives in the proper order. Amanda cannot
tell from the labels or otherwise which is which.
--Adrian
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
ERROR: localhost: [could not access /mnt/sglab/monica/DPO
(/mnt/sglab/monica/DPO): Permission denied]
...
Looks like nsf mounts. ...
What does the National Science Foundation have to do with this??? :-)
Sorry, couldn't resist. I've flipped those characters many times myself.
In addition
who upgrade.
Frank
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
to guess_disk.
Toralf Lund [EMAIL PROTECTED]
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
here. If you ran ./configure yourself, it should
have hunted around and found this version of tar (as Jean-Louis said).
If, however, you're running a pre-built version of Amanda, all bets are
off about how the person who put it together set things up.
-Josh
John R. Jackson, Technical Software
kill the amrestore process on the server.
That usually puts the drive into high speed search mode looking for the
next tape mark. It certainly stops shipping all the data across the wire.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
not already using the 'h' flag.
Jacek Drobiecki
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Yeah, still have a problem. ...
Nuts! I was hoping your problem would just go away on its own :-).
Am I doing this on the troublesome client or on the tape host?
The client.
[EMAIL PROTECTED]
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
I wonder how hard it would be to train amanda to rotate holding
areas (partitions) on a raid instead of tapes? ...
That's already part of 2.4.3 (or the amanda-242-tapeio CVS branch).
Gene
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
... Can Amanda backup the 4 filesystems to the 4 tape-drives
simultaneously from one config file? ...
No. That will take a major overhaul (although it's definitely needed).
Aaron Rainwater
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Prior to the run a netstat -a | grep amanda shows
udp0 0 *:amanda*:*
...
That's normal.
After the failure a netstat -a | grep amanda doesn't return anything.
Then amandad is probably dieing. The FAQ talks about how to debug this.
Shawn
John R. Jackson
tape, skip out a bunch of files then
do an mt offline on the no-rewind device. Look at the tape when it
comes out. All the media will be to one side.
stephen.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
'
...
As long as you don't need that changer (which you probably don't)
then I'd just go into config/config.h and #undef'ing HAVE_CHIO_H and
HAVE_SYS_CHIO_H. That should effectively disable that part of the build.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
/*
I can never figure this stuff out in my head, so use this test script:
ftp://gandalf.cc.purdue.edu/pub/amanda/gtartest-exclude
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
with Amanda does a very good job of this -- if it's happy, you're almost
certain to be able to do restores.
If you're using dump, amverify will tell you the tape itself is readable
but it cannot tell you whether the internal (dump) structure is sane.
Only a real restore can do that.
John R
this again and make sure you remove the /tmp files first.
Also, try it with bs=64k.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
, could you get a debugger traceback
and post it?
Brian
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
some more entries in the NOTES section.
Jon H. LaBadie
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
server? ...
Set maxdumps in either the global area of amanda.conf or in the specific
dumptypes (see amanda(8)).
Trevor.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
(that's what Amanda is using) is
version 1.13.25? What does /bin/tar --version say?
Every time this has come up in the past it's been because tar was version
1.13, which is just plain evil.
Stefan
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
). My guess is you don't have that set up properly
for hda2 but it is correct for hda1.
The easy answer to your problem is to use the mount point name in your
disklist instead of the disk name. In other words, change hda1 in
disklist to /mnt/w2k.
[EMAIL PROTECTED]
John R. Jackson, Technical Software
by
the device name. The specific value depends on your OS, by usually is
a name with an 'n' in it, which I think for your OS would be /dev/nst0
(or maybe /dev/st0n).
Doug
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
for a third estimate (level 2 in this case)
if it's time (bumpdays) that a bump to the next level might be possible.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
96890112 fm 22 writing file: No space left on device
That no space left on device usually translates to you just hit end
of tape. Note that you wrote just under 100 GBytes. Does that roughly
match the tape size you expect?
Dick Kreutzer
John R. Jackson, Technical Software Specialist, [EMAIL
and the partition I wish to restore from. This
works fine I can do an ls and add files to be extracted and etc
This doesn't match with it cant find the index files, so I'm a bit
confused about what error you're seeing.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
and not do a whole lot
of other things.
Eric Zylstra
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
have to read that error message *very* carefully. It says the request
came in from cain, not cain.usace.army.mil. For whatever reason,
you're not using fully qualified domain names (*). Adding this line
should fix (get around) your problem:
cain root
- db
John R. Jackson, Technical Software
want
at least a weekend worth of logs left around.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
.
What do you see if you do df -k /bkp1?
Steve
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
I forgot one thing. Is /bkp1 listed in /etc/fstab (or the equivalent
for whatever OS you're using)? It has to be in there so Amanda can
convert the logical (mount point) name to a disk name to hand to dump.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
.
Bjoern v. Benckendorff
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
planner: Adding new disk localhost:/tmp.
FYI, you should not use localhost in your disklist file. Trust me.
It will eventually bite you. Use the fully qualified name of your host.
Steve
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
they don't take up any
(data) space, hence the appearance that they are zero length.
Brandon D. Valentine
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
.
Why the number of tapes is wrong is left as an exercise for the
reader :-).
J
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
specific tapes as no-reuse.
You might do this, for instance, if they have been taken off site for
protected storage and you want Amanda to skip over them if they happen
to come up in the cycle again.
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
) will be small enough to leave room for full dumps
of some other new disks.
After everything gets a full dump once, Amanda will begin shifting the
schedule around to balance things out.
Peter Normann
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
take a shot at it in between). I'm sure
it's just some header file magic that has to be dealt with.
Marc Davignon
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
I think the --with-smbclient and --with-samba-user options will be most
helpful.
As you say, --with-smbclient is the real problem. However the
--with-samba-user option is deprecated in favor of putting the user name
in the amandapass file.
John R. Jackson, Technical Software Specialist, [EMAIL
(e.g. in your
/etc/services file) or account for their potential use by making the
portrange larger.
You should also look at:
http://www.amanda.org/patches.html
for a related portrange patch to 2.4.2p2.
--david
John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
1 - 100 of 1745 matches
Mail list logo
