Re: problems with autoconf?
On Thu, 16 Aug 2001, John R. Jackson wrote:
> >i'm using GNU autoconf, version 2.52 to attempt (heh) to get the configure
> >script to work on the CVS version of amanda 2.5.0.. and i keep running in
> >to errors.
>
> Wow. Three problems in only one sentence :-).
compression technology limits me to only 3 problems per sentence.. soon i
hope to have it up to at least 5.
maybe more ;)
> First, you don't run autoconf yourself to build Amanda. As described in
> the FAQ (www.amanda.org), you should run ./autogen instead. Also, right
> after a "cvs checkout" you have to run it twice to get one of the files
> built properly. After that you just run it once to deal with any changes.
hrm. ok, i'm testing this right now, before i head home. so far, it's
spewed the same errors.. and failed exactly the same way. i'll run back to
autoconf 2.13 as you seem to suggest below..
> Second, the rumor is that new versions of autoconf are seriously broken.
> I use 2.13. I also use the automake from CVS. As I recall, the released
> version no longer works with something or other Amanda does (don't blame
> me for this -- I didn't break it).
meh, you've been one of the most helpful people on this list so far. gone
out of your way in many ways for me, when you didn't need to. it's not
unapreciated..
from the "new" autoconf (2.13_1), and autogen:
[jan@hi amanda] {835} ./autogen
ls: aclocal.m4: No such file or directory
aclocal: /usr/local/share/aclocal/sdl11.m4: 11: duplicated macro
`AM_PATH_SDL'
configure.in: 8: required file `config/config.h.in' not found
configure.in:1734: warning: AC_TRY_RUN called without default to allow
cross compiling
[jan@hi amanda] {836} ./autogen
ls: aclocal.m4: No such file or directory
aclocal: /usr/local/share/aclocal/sdl11.m4: 11: duplicated macro
`AM_PATH_SDL'
configure.in: 8: required file `config/config.h.in' not found
configure.in:1734: warning: AC_TRY_RUN called without default to allow
cross compiling
that's the fresh source for 2.5.0...
> Third, you must be planning on doing some **serious** C debugging if
> you're thinking of running the Amanda 2.5 branch. That's for developers
> only and is unstable and unsupported. When I was last in there up-porting
> some 2.4.2 changes, I found all sorts of problems.
actually, i'm having just enough problems with chg-chio on freebsd that
i'd like to get chg-scsi compiled. i read (oosp, i shouldn't do that) at
least one post about getting 2.5.0's chg-scsi working with 2.4.2-p2
binaries. this seems to be a bit of a solution for my problems.
i've not been able to dig up HOW anyone's gotten chg-scsi to compile on
FreeBSD 4.x, which is rather annoying.
chg-chio is basically, not calling chio in a way that i can debug and get
it to release the current tape.
i'm working on a bit of a time limit here, as i'd like to have this
working, if not in production, before monday.
thanks again,
-- jan
---/ f. johan beisser /--+
http://caustic.org/~jan [EMAIL PROTECTED]
"if my thought-dreams could be seen..
"they'd probably put my head in a gillotine"
-- Bob Dylan
Re: guh. new problem?
On Thu, 16 Aug 2001, John R. Jackson wrote: > Before you try working with amdump, you should use amcheck. It does > a better job of diagnosing problems. Also, if it's not happy, you can > bet amdump will be even less so. it seems happy.. gump# su amanda -c "amcheck daily" Amanda Tape Server Host Check - Holding disk /usr/dumps: 930872 KB disk space available, that's plenty amcheck-server: slot 1: date 20010816 label DailySet1-001 (active tape) amcheck-server: fatal slot chio:: /dev/ch0: CHIOMOVE: Invalid argument ERROR: new tape not found in rack (expecting a new tape) NOTE: skipping tape-writable test NOTE: info dir /var/log/amanda/DailySet1/curinfo: does not exist NOTE: it will be created on the next run Server check took 56.479 seconds Amanda Backup Client Hosts Check Client check: 1 host checked in 0.891 seconds, 0 problems found (brought to you by Amanda 2.4.2p2) again, the error is just wierd, the problem comes when the dumper on the client machine is called. i'm not sure if there's something i'm doing wrong or not. > In your case, use the "-c" option to just do the client checks, since > that's what's failing. You don't need to waste time doing the server > checks, in particular the tape ones. I usually use "-cl" which does > everything but the tape checks. so far, this has been a slight problem on the host side aswell, since chg-chio doesn't seem to get how to unmount a tape after it's spooled.. on the client side of things, the debugging output is fairly minor. i'm wondering if there's something i've really screwed up. i think, after all this is over, i'm going to write a metric buttload of documentation. > Here are the FAQ items of interest to what amcheck will probably say: > > http://amanda.sourceforge.net/fom-serve/cache/16.html > http://amanda.sourceforge.net/fom-serve/cache/140.html i've run through each of those. if you want, i'll provide the output from "User Amanda", that is mailed to me after each failure. > Have you gone through all the items they mention? and then some. i wish this were an obvious problem. > >should i have an actual device for dumping this? > > Huh? in this case: instead of /home/ using the /dev/ad0s2f. that, or the raw device itself. i'm still trying to figure out what's going wrong.. -- jan ---/ f. johan beisser /--+ http://caustic.org/~jan [EMAIL PROTECTED] "if my thought-dreams could be seen.. "they'd probably put my head in a gillotine" -- Bob Dylan
Re: guh. new problem?
>but, the output from AMANDAs server seems to be ok.. > >FAILURE AND STRANGE DUMP SUMMARY: > hi /usr/home lev 0 FAILED [disk /usr/home offline on hi.hostname?] Before you try working with amdump, you should use amcheck. It does a better job of diagnosing problems. Also, if it's not happy, you can bet amdump will be even less so. In your case, use the "-c" option to just do the client checks, since that's what's failing. You don't need to waste time doing the server checks, in particular the tape ones. I usually use "-cl" which does everything but the tape checks. Here are the FAQ items of interest to what amcheck will probably say: http://amanda.sourceforge.net/fom-serve/cache/16.html http://amanda.sourceforge.net/fom-serve/cache/140.html Have you gone through all the items they mention? >should i have an actual device for dumping this? Huh? >-- jan John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: problems with autoconf?
>i'm using GNU autoconf, version 2.52 to attempt (heh) to get the configure >script to work on the CVS version of amanda 2.5.0.. and i keep running in >to errors. Wow. Three problems in only one sentence :-). First, you don't run autoconf yourself to build Amanda. As described in the FAQ (www.amanda.org), you should run ./autogen instead. Also, right after a "cvs checkout" you have to run it twice to get one of the files built properly. After that you just run it once to deal with any changes. Second, the rumor is that new versions of autoconf are seriously broken. I use 2.13. I also use the automake from CVS. As I recall, the released version no longer works with something or other Amanda does (don't blame me for this -- I didn't break it). Third, you must be planning on doing some **serious** C debugging if you're thinking of running the Amanda 2.5 branch. That's for developers only and is unstable and unsupported. When I was last in there up-porting some 2.4.2 changes, I found all sorts of problems. >---/ f. johan beisser /--+ John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Need some sample MTX output
I'm working on cleanup and enhancements to the chg-zd-mtx script (actually, all of the sh driven scripts) based on trouble some folks have had recently. But mtx on my system is problematic (I'm not sure they ever considered a robot with ~800 slots :-). So, I'm looking for some sample output from other people who run mtx that I can do simulated testing with. Here's what I'd like: * The output of "mtx --version". * The output of "mtx status" with no tape drives loaded. * The output of "mtx load ..." (and the command line you used). * The output of "mtx status" with one (or more) drives loaded. * The output of "mtx unload ..." (and the command line you used). Please remove the Cc from any replies and send the results just to me to avoid the noise on the mailing list. Thanks. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: starttime weirdness
Thanks for the info. . . . > You said you didn't give Amanda a holding disk (in the initial setup), > but I suspect what you meant was you gave it a tiny one. My guess is > Amanda saw at least one estimate for your delayed host that would fit > and so everything that would not fit (which was the bulk of the entries) > sat until that got done, which did not happen until after starttime. In my usual style for "archive" backups, I go straight to tape by specifying no holding disk at all. Given your explanation in (the deleted part of) your message, I would think the entries without a starttime would go ahead and write to tape as there is no holding disk for which to wait. Yet, things still didn't appear to work this way when I attempted things first in this fashion. > By giving it a holding disk as large as the largest image, you guarantee > this won't happen since everything can go through the holding disk and > nothing goes direct to tape (which has speed and tape error handling > advantages as well). > > You could probably get by with less holding disk, but if it gets too > small you'll be back in the same situation. > > Actually, I think it only needs to be as large as the largest image that > is not delayed by starttime. Of course, that might not be any smaller > than what you've already set up. > > Amanda could probably be taught to flip back and forth between through the > holding disk and direct to tape if taper has gone idle and everthing left > to do through the holding disk is delayed (and maybe the estimated time > does not cross into the nearest starttime, or at least by very much). > Or, harder but maybe better, would be to allow FILE-DUMP (into the > holding disk) at the same time as PORT-DUMP (direct to tape). > > Volunteers? Yeah, that's what I thought. :-) I'll add it to the > TODO list. :) I don't ever want to underestimate what you do, John. Thanks for your never ending help on the mailing list and for your ongoing and continued efforts towards the improvement of Amanda. I'm not a programmer or I may be interested in helping out. Thus, I'm all the more thankful to those of you who do. Needless to say, your name comes up time and time again. Thanks!!! > > Tape overflow may also help this as you'll then only need enough holding > disk for one tape chunk instead of the largest image. > > >Paul Yeatman > > John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED] -- Paul Yeatman (858) 534-9896[EMAIL PROTECTED] == ==Proudly brought to you by Mutt== ==
problems with autoconf?
i'm using GNU autoconf, version 2.52 to attempt (heh) to get the configure
script to work on the CVS version of amanda 2.5.0.. and i keep running in
to errors.
[jan@hi amanda] {726} autoconf --version
autoconf (GNU Autoconf) 2.52
[jan@hi amanda] {727} autoconf
configure.in:7: error: possibly undefined macro: AM_INIT_AUTOMAKE
configure.in:8: error: possibly undefined macro: AM_CONFIG_HEADER
configure.in:1131: error: possibly undefined macro: AM_DISABLE_SHARED
configure.in:1141: error: possibly undefined macro: AM_CONDITIONAL
configure.in:1631: error: possibly undefined macro: AM_PROG_LIBTOOL
configure.in:1660: error: possibly undefined macro: AM_PROG_LEX
is there something i just Don't Get?
help anyone?
this is running on FreeBSD 4.3, amanda from the ports compiles just fine
(version 2.4.2-p2). this has been a slight point of frustration for me
(heh, slight is relative). is there a patch, or something i don't know
about?
i've not been able to find much in the archives, except that most people
seem to not have this problem, at all.
any advice? beauller? hello?
---/ f. johan beisser /--+
http://caustic.org/~jan [EMAIL PROTECTED]
"if my thought-dreams could be seen..
"they'd probably put my head in a gillotine"
-- Bob Dylan
RE: Waiting for ack
<-- Snip -->> What RPMs did you update? Did you setup some firewalling by default? <-- Snip -->> Boy I wish I could answer that one. RedHat has made it so simple to keep your system updated using the RedHat Network that I no longer have to manually update the system. Hence I'm kind of oblivious of what is being updated. Maybe that can be a bad thing... No firewalling is installed. Amanda starts up on both systems, server and client. The client completes sendsize then hangs. Thanks for the helping hand! I really appreciate your effort and time! James
Re: Waiting for ack
> I am having difficulties in running amanda. I am unable to get a backup (but > once was) and I think it was after updating some RPM on my system. It takes > 30 minutes to fail and the only info I can find in in the amandad.debug file > on the client. I get 4 lines of > amandad: waiting for ack: Connection refused, retrying > followed by one line of > amandad: waiting for ack: Connection refused, giving up! > Any hints as to what went wrong? What RPMs did you update? Did you setup some firewalling by default?
Re: tapeio branch: using smaller files?
This particular one is not a FAQ, yet. Because tapeio is not in common use, yet. The 2 GB file limit question is a FAQ when asked in regard to holding disk files and in regard to amrestore, and each situation has a somewhat different answer. There is a config option that works around the problem nicely with holding disk files. In your case it's a matter of your "tapes" being limited to 2 GB in size. Thus your options would appear to be using bigger "tapes", splitting your dumps up into less than 2 GB each, or somehow splitting your dumps across multiple tapes. Using bigger tapes in your case means putting your dumps on a filesystem that supports large files. Splitting your dumps up into smaller pieces is a commonly used solution with amanda to workaround this problem. There is a tar wrapper script at: ftp://gandalf.cc.purdue.edu/pub/amanda/gtar-wrapper.* ... that can help with this. Splitting a single dump across multiple tapes is not yet supported in amanda, though there is an ongoing project to add this feature. -Mitch On Thu, 16 Aug 2001, Dan Debertin wrote: > This probably falls under the "why can't I split my backups across > multiple tapes?" FAQ, but maybe not. > > I'm using the file: driver from the tapeio branch to back up a large > filesystem to another disk. The system that I'm using (Linux) doesn't > support files larger than 2GB, which is causing backups to fail. > > I'm thinking that my only options are upgrading to kernel 2.4 and > recompiling glibc, or switching to a different OS, but if there's any way > around this from the amanda-configuration standpoint, that would be > simplerideas? > > > Dan Debertin > -- > ++ Dan Debertin > ++ Senior Systems Administrator > ++ Bitstream Underground, LLC > ++ [EMAIL PROTECTED] > ++ (612)321-9290 x108 > ++ GPG Fingerprint: 0BC5 F4D6 649F D0C8 D1A7 CAE4 BEF4 0A5C 300D 2387 > > >
more on the lil client problem..
forgot to include some output from amadmin.. gump# su amanda -c "amadmin daily disklist hi.hostname /usr/home" line 12: host hi.hostname: interface default disk /usr/home: program "DUMP" priority 2 dumpcycle 0 maxdumps 1 strategy STANDARD compress NONE auth BSD kencrypt NO holdingdisk YES record YES index NO skip-incr NO skip-full NO ---/ f. johan beisser /--+ http://caustic.org/~jan [EMAIL PROTECTED] "if my thought-dreams could be seen.. "they'd probably put my head in a gillotine" -- Bob Dylan
guh. new problem?
thanks to the many helpful answers i've had on this list.. but sadly, i need more help. this time, it seems the problem is client side.. so far, i've gotten most things working, the server side is now able to create and handle the dumps, but it doesn't seem that i'm able to get the client side working.. nor, it seems, am i getting anything from the logs.. but, the output from AMANDAs server seems to be ok.. FAILURE AND STRANGE DUMP SUMMARY: hi /usr/home lev 0 FAILED [disk /usr/home offline on hi.hostname?] i've gone through what documentation i've been albe to dig up, but haven't found enough info for me to really see what is going on. should i have an actual device for dumping this? trying this next.. -- jan ---/ f. johan beisser /--+ http://caustic.org/~jan [EMAIL PROTECTED] "if my thought-dreams could be seen.. "they'd probably put my head in a gillotine" -- Bob Dylan
Re: [GNUTAR program not available]
At Thu, 16 Aug 2001 15:28:27 EDT, [EMAIL PROTECTED] wrote... : Was gnutar installed *before* amanda? Look in /tmp/amanda/amandad*debug : -- there should be a line that says "GNUTAR=/bin/tar" or some such. If : not, configure didn't find it and you need to recompile/reinstall amanda. Indeed. Thanks all. -fil ___ Fil Krohnengold| Network Systems Administrator [EMAIL PROTECTED] | The American Museum of Natural History 212/769-5294 | CPW @ 79thSt / NY, New York / 10024
RE: Waiting for ack
Thanks for the reply! All help is greatly appreciated!
I have checked both the server and client machines for the services. They
are auto starting. I have even restarted xinetd.
As far as the RPM updates, Amanda was not one of them so the amanda.rpmnew
file did not exist.
I don't think it is tcp wrappers. It was working once for a few months. I
did check the files and they were empty with the exception of the notes.
I did an nslookup from both machines and was able to find by IP and by name.
Thanks again for pitching in!
James
<-- Snip -->
try running 'ntsysv' and double checking that 'amanda' and 'xinetd' are
both selected, restart xinetd '/sbin/service xinetd restart'
check for /etc/xinetd/amanda.rpmnew files
IIRC, xinetd also uses tcpwrappers. check /etc/hosts.{allow,deny}
double check forward and reverse DNS match up ( this one caught me once)
<-- Snip -->
tapeio branch: using smaller files?
This probably falls under the "why can't I split my backups across multiple tapes?" FAQ, but maybe not. I'm using the file: driver from the tapeio branch to back up a large filesystem to another disk. The system that I'm using (Linux) doesn't support files larger than 2GB, which is causing backups to fail. I'm thinking that my only options are upgrading to kernel 2.4 and recompiling glibc, or switching to a different OS, but if there's any way around this from the amanda-configuration standpoint, that would be simplerideas? Dan Debertin -- ++ Dan Debertin ++ Senior Systems Administrator ++ Bitstream Underground, LLC ++ [EMAIL PROTECTED] ++ (612)321-9290 x108 ++ GPG Fingerprint: 0BC5 F4D6 649F D0C8 D1A7 CAE4 BEF4 0A5C 300D 2387
Re: [GNUTAR program not available]
On Thu, 16 Aug 2001 at 2:25pm, fil krohnengold wrote
> Amanda Backup Client Hosts Check
>
> ERROR: diglib1: [GNUTAR program not available]
> Client check: 16 hosts checked in 1.274 seconds, 1 problem found
>
> (brought to you by Amanda 2.4.2p2)
> {amanda}hammond:~>
>
> gnutar, readline, termcap and amanda are all fresh installs.
Was gnutar installed *before* amanda? Look in /tmp/amanda/amandad*debug
-- there should be a line that says "GNUTAR=/bin/tar" or some such. If
not, configure didn't find it and you need to recompile/reinstall amanda.
> bash-2.02$ tar --version
> tar (GNU tar) 1.13
I'm not sure, buy you may want to grab a more recent version, like
1.13.19, from alpha.gnu.org.
--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University
Re: Waiting for ack
Hello...
James Stevens wrote:
> I am having difficulties in running amanda. I am unable to get a backup (but
> once was) and I think it was after updating some RPM on my system. It takes
> 30 minutes to fail and the only info I can find in in the amandad.debug file
> on the client. I get 4 lines of
> amandad: waiting for ack: Connection refused, retrying
> followed by one line of
> amandad: waiting for ack: Connection refused, giving up!
> Any hints as to what went wrong?
>
> James
>
try running 'ntsysv' and double checking that 'amanda' and 'xinetd' are
both selected, restart xinetd '/sbin/service xinetd restart'
check for /etc/xinetd/amanda.rpmnew files
IIRC, xinetd also uses tcpwrappers. check /etc/hosts.{allow,deny}
double check forward and reverse DNS match up ( this one caught me once)
--
Christopher McCrory
"The guy that keeps the servers running"
[EMAIL PROTECTED]
http://www.pricegrabber.com
I don't make jokes in base 13. Anyone who does should get help.
--Douglas Adams
Re: [GNUTAR program not available]
>ERROR: diglib1: [GNUTAR program not available] This message means GNU tar could not be found (maybe it was not installed yet) when ./configure was run for Amanda. Now that you have GNU tar available, remove config.cache (or "make distclean") and start the Amanda build over again. >bash-2.02$ tar --version >tar (GNU tar) 1.13 NO NO NO NO NO !! That version is evil, evil, evil. Don't use it! Get 1.13.19 from alpha.gnu.org. >-fil John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
[GNUTAR program not available]
Howdy - I've been running v2.4.2p2 with great results for Solaris
2.6 and 7 machines. I've added several client machines along
with way without a hitch - until now.
I just added a new machine - a Sun E250 runing Sol2.6. I need to
use tar because the folks who run that machine refused to follow
my advice concerning fs sizes. And, well:
{amanda}hammond:~> amcheck -c daily
Amanda Backup Client Hosts Check
ERROR: diglib1: [GNUTAR program not available]
Client check: 16 hosts checked in 1.274 seconds, 1 problem found
(brought to you by Amanda 2.4.2p2)
{amanda}hammond:~>
gnutar, readline, termcap and amanda are all fresh installs.
I've tried trussing every pid I could find related to this on
both client and server but that's turned up nothing. And:
diglib1:~> sudo su - amanda
Password:
bash-2.02$ tar --version
tar (GNU tar) 1.13
Copyright (C) 1988, 92,93,94,95,96,97,98, 1999 Free Software
Foundation, Inc.
[...]
I'm stumped.
-fil
___
Fil Krohnengold| Network Systems Administrator
[EMAIL PROTECTED] | The American Museum of Natural History
212/769-5294 | CPW @ 79thSt / NY, New York / 10024
Re: Trouble using amrecover. . .(Long Message)
In your dd tests you always rewind the tape first. Did you rewind the tape before doing the amrecover? Amidxtaped does not rewind on the off chance you've pre-positioned the tape to the appropriate image. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Waiting for ack
I am having difficulties in running amanda. I am unable to get a backup (but once was) and I think it was after updating some RPM on my system. It takes 30 minutes to fail and the only info I can find in in the amandad.debug file on the client. I get 4 lines of amandad: waiting for ack: Connection refused, retrying followed by one line of amandad: waiting for ack: Connection refused, giving up! Any hints as to what went wrong? James
Trouble using amrecover. . .(Long Message)
Hello everyone!
I am having a bit of trouble using amrecover and am hoping that someone can
see what is going wrong.
As root, I run amrecover, then select a few files to test restoring. This
is what I get:
sbs|amvr: /home/amv > amrecover Gemini
AMRECOVER Version 2.4.2p2. Contacting server on sbs ...
220 sbs AMANDA index server (2.4.2p2) ready.
200 Access OK
Setting restore date to today (2001-08-16)
200 Working date set to 2001-08-16.
200 Config set to Gemini.
200 Dump host set to sbs.
$CWD '/home/amv' is on disk '/home' mounted at '/home'.
200 Disk set to /home.
/home/amv
WARNING: not on root of selected filesystem, check man-page!
amrecover> add HDDUpgrade.doc aixupdates.tgz
Added /amv/HDDUpgrade.doc
Added /amv/aixupdates.tgz
amrecover> extract
Extracting files using tape drive /dev/rmt0.1 on host sbs.
The following tapes are needed: Gemini-4
Restoring files into directory /home/amv
Continue? [Y/n]: y
Load tape Gemini-4 now
Continue? [Y/n]: y
EOF, check amidxtaped.debug file on sbs.
amrecover: short block 0 bytes
UNKNOWN file
amrecover: Can't read file header
extract_list - child returned non-zero status: 1
Continue? [Y/n]: y
amrecover> exit
200 Good bye.
sbs|amvr: /home/amv >
Here is the debug file:
sbs|amvr: /tmp/amanda > cat amidxtaped.20010816083444.debug
amidxtaped: debug 1 pid 197734 ruid 1006 euid 1006 start time Thu Aug 16
08:34:44 2001
amidxtaped: version 2.4.2p2
> SECURITY USER root
bsd security: remote host sbs.sbs.com user root local user amanda
amandahosts security check passed
> 6
amrestore_nargs=6
> -h
> -p
> /dev/rmt0.1
> sbs
> ^/home$
> 20010816
Ready to execv amrestore with:
path = /usr/local/sbin/amrestore
argv[0] = "amrestore"
argv[1] = "-h"
argv[2] = "-p"
argv[3] = "/dev/rmt0.1"
argv[4] = "sbs"
argv[5] = "^/home$"
argv[6] = "20010816"
amrestore: error reading file header: I/O error
amidxtaped: amrestore terminated normally with status: 2
Rewinding tape: done
amidxtaped: pid 197734 finish time Thu Aug 16 08:35:40 2001
If I access the tape using mt and dd, it reads fine:
sbs|amvr: /home/amv > mt -f /dev/rmt0.1 rewind
sbs|amvr: /home/amv > dd if=/dev/rmt0.1 bs=32k count=1
AMANDA: TAPESTART DATE 20010816 TAPE Gemini-4
1+0 records in.
1+0 records out.
sbs|amvr: /home/amv > mt -f /dev/rmt0.1 rewind
sbs|amvr: /home/amv > mt -f /dev/rmt0.1 fsf 1
sbs|amvr: /home/amv > dd if=/dev/rmt0.1 bs=32k count=1
AMANDA: FILE 20010816 sbs /home lev 0 comp .gz program /usr/bin/gtar
To restore, position tape at start of file and run:
dd if= bs=32k skip=1 | /usr/bin/gzip -dc | usr/bin/gtar -f...
-
1+0 records in.
1+0 records out.
sbs|amvr: /home/amv > mt -f /dev/rmt0.1 rewind
sbs|amvr: /home/amv > mt -f /dev/rmt0.1 fsf 1
sbs|amvr: /home/amv > dd if=/dev/rmt0.1 bs=32k skip=1
of=/gemini/amanda/home.tgz
15586+0 records in.
15586+0 records out.
sbs|amvr: /home/amv > ls -l /gemini/amanda/home.tgz
-rw-rw 1 root system 510722048 Aug 16 08:57
/gemini/amanda/home.tgz
sbs|amvr: /home/amv > gtar -tvzf /gemini/amanda/home.tgz |pg
drwxr-xr-x root/usr167 2001-06-28 15:24:00 ./
drwxr-xr-x root/system 187 2000-10-16 08:08:48 ./TT_DB/
drwxrwx--- amanda/backup 174 2001-08-15 19:24:26 ./amanda/
drwxr-xr-x amanda/backup 377 2001-07-03 08:39:29 ./amanda/amanda-2.4.2p2/
drwxr-xr-x amanda/backup 128 2001-06-29 11:24:46
./amanda/amanda-2.4.2p2/amplot/
drwxr-xr-x amanda/backup 666 2001-06-29 11:28:43
./amanda/amanda-2.4.2p2/changer-src/
drwxrwx--- amanda/backup 1 2001-06-29 11:24:31
./amanda/amanda-2.4.2p2/changer-src/.libs/
drwxr-xr-x amanda/backup 743 2001-06-29 11:28:07
./amanda/amanda-2.4.2p2/client-src/
drwxrwx--- amanda/backup49 2001-06-29 11:22:46
./amanda/amanda-2.4.2p2/client-src/.libs/
drwxr-xr-x amanda/backup 1104 2001-06-29 11:27:58
./amanda/amanda-2.4.2p2/common-src/
.
.
.
And so on.
Here is my system info in case it's needed:
Hardware: IBM RS/6000 H80
OS:AIX 4.3.3.0
Amanda: 2.4.2p2
My disklist entries:
sbs /homegemini-store
sbs /u02/OracleBackup oracle-data-full
sbs /u01 oracle-app-full
sbs /gemini/disk4/stbkup gemini-store
My dumptypes:
define dumptype global {
comment "Global definitions"
index yes
record yes
program "GNUTAR"
}
define dumptype oracle-data-full {
global
comment "Full dump of this filesystem always"
compress none
priority high
holdingdisk no
dumpcycle 0
}
define dumptype oracle-app-full {
global
comment "Full dump of this filesystem always"
compress client
priority high
exclude "./oradata/oracle"
dumpcycle 0
}
define dumptype gemini-store {
global
comment "Backup of Gemini EOD Store files"
compress client
priority high
dumpcycle 0
}
Sorry for the long message, I just don't want to omit any relevant
information.
Any suggestions would be very appreciated!!
Thanks!
Anthony Valentine
Spenard Builders Supply
AIX Sys Admin/Oracle DBA
[EMAIL PROTECTED]
Re: tapetype running for 8 days !?
>Do you have an idea, whether the long time for tapetype can be ok? Is there >any possibility to check, whether the program runs correctly? If you're using 2.4.2p2, then it's not running properly, and I'd be very suspicious of the drive or the kernel driver. >Thomas. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
RE: tapetype running for 8 days !?
I´m using the latest release, so the newest tapetype is used. Unfortunately the mailing list archive is temporarily unavailable, and the ftp archive is out of date. Generally, the drive is a rather cheap drive, but as far as I read it shouldn´t be so bad. Do you have an idea, whether the long time for tapetype can be ok? Is there any possibility to check, whether the program runs correctly? Best regards, Thomas. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of John R. Jackson Sent: Thursday, August 16, 2001 2:43 PM To: Thomas Eisenzopf Cc: Amanda user's group Subject: Re: tapetype running for 8 days !? >I?m preparing to use amanda ... What version? The tapetype program with 2.4.2* is much better than the one with 2.4.1*. And if you're just starting with Amanda, you should be starting with the latest release (2.4.2p2). >... therefore I use the tapetype command because >my tapedrive (OnStream DI30, an IDE ADR tape drive, 15GB uncompressed, no >hardware compression) is not preconfigured in amanda. A word of caution. That drive has had serious problems reported here. I don't use either it or the OS that's usually involved, so don't remember much more about it, but you might search the mailing list archives for advice. I suspect you'll also find a tapetype for it and can avoid running that yourself. >Thomas. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: regenerating an amanda report
On Wed, 15 Aug 2001, David Carter wrote: > Hello all, > > I didn't receive a mail report this morning from amanda, although I can see > that the backups completed successfully, and the tape contents look good. > > Is there a way I can re-generate the amdump report that I usually get via > email? > Look at the amreport command (man amreport as well). You can point it at a log file and it will regenerate the report. Very handy. Later, Bill Carlson -- Systems Programmer[EMAIL PROTECTED] | Anything is possible, Virtual Hospital http://www.vh.org/ | given time and money. University of Iowa Hospitals and Clinics | Opinions are mine, not my employer's. |
Re: Configuration
>I have not formatted the tape or ever succesfully mounted the tape drive to >/backup (which is where I want it). Ummm, I'm guessing you've probably also not used tapes with Unix much, either :-). Tapes do not need to be formatted like disks, i.e. there is nothing like mkfs/newfs involved. Also, they are not mounted like disks. You always just access them via their /dev entry. >I have set the following parameters in amanda.conf > tapedev "/dev/st0" As you noted, this must be what's called a "no-rewind" device name. What that means is when the program does a close, the tape stays put. The alternative, "rewind", will rewind the drive whenever the program does a close (which is one of the more useless things ever invented). The usual OS convention for "no-rewind" is to have an 'n' in the name, so I suggest you look around for something like "/dev/nst0" (or the 'n' may be someplace else in the name). Once you have that, you need to make sure the drive is working at all. Load a tape in it and do: mt -f /dev/whatever rewind Once you get to the point that this succeeds, you can start trying to get Amanda working, i.e. try the amlabel again. > localhost sdb1 nocomp-root Please don't use "localhost". That will bite you eventually. Always use fully qualified host names. >And since I am sure someone has the answer to this one and as soon as I get >past it I will run into something else. what is a good resource to help walk >me through the setup process? Have you read "the chapter": htpp://www.backupcentral.com/amanda.html >Michael Blood John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: "could not access" using tar
>... My only guess >is that amanda is denied access because this home directory only gives >permissions to the user ... That's the right guess. Amcheck (actually, selfcheck) does not run as root so cannot get there. The actual backup will run GNU tar as root (under runtar) and so should be OK. I don't know of any good solution to this problem. >... I've come to believe that as long as the >amanda user is in the group with access to the device, this shouldn't >matter. ... That only applies to using "dump". Device access permissions are irrelevant when using GNU tar. >Paul Yeatman John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: tapetype running for 8 days !?
>I?m preparing to use amanda ... What version? The tapetype program with 2.4.2* is much better than the one with 2.4.1*. And if you're just starting with Amanda, you should be starting with the latest release (2.4.2p2). >... therefore I use the tapetype command because >my tapedrive (OnStream DI30, an IDE ADR tape drive, 15GB uncompressed, no >hardware compression) is not preconfigured in amanda. A word of caution. That drive has had serious problems reported here. I don't use either it or the OS that's usually involved, so don't remember much more about it, but you might search the mailing list archives for advice. I suspect you'll also find a tapetype for it and can avoid running that yourself. >Thomas. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
The Endless Estimate
I have been having some problems with Amanda lately. After doing some tweaking with it yesterday, I ran it from the command prompt, using the same command I have in the cron file for the user amanda. I have it backing up itself and another server. All servers are running RH Linux, and Amanda had been working in the past. After kicking it off yesterday, it got estimates for itself right away. It then went to the other server and tried to get an estimate. When I left yesterday, it was still getting the estimate. I got in this morning and it is still getting the estimate. What's going on here? The logs don't tell me anything about a problem. Thanks for any assistance.
Re: Amanda + Solaris 8 + StorEdge L9
Hi, Thank's for pointing me to mtx, now it works. And yes I was using the sgen from Sol 8 but wanted to go with amanda stuff only ( chg-scsi ). Thank's again servus cd > > Quoting Christian Doppler ([EMAIL PROTECTED]): > > > > I'm having problems getting that combination running. > > > > Amanda 2.4.2p2, Solaris 8 04/01 on E250, Sun StorEdge L9 > > > > Someone may has a System like that running and can give me hints. > > > > I have a System running (Amanda 2.4.2p2, Solaris 2.6 on E450, Sun > > DLT-4700 Changer and using the stctl 1.2.5 driver) since 3 Years > > whitout any problems but now i have to change. > > > > Any Hints would be nice > > Hi... > > I have the same setup here (mostly, anyway)... -- Solaris 8 (01/01) on > an E220R, with a StorEdge L9. I'm using mtx (http://mtx.sourceforge.net) > to control the L9, along with an mtx-changer script from Tim Skirvin > (http://www.killfile.org/~tskirvin/software/mtx-changer/mtx-changer-1.1). > The chg-scsi driver segfaulted and died on me whenever I tried to use it, > so the solution became 'mtx' shortly after (I love impossible deadlines). > :-) > > There's a script in the contrib directory of the mtx source tree that > configures the Solaris 8 'sgen' driver to handle the tapechanger as > well... it may be the piece you're missing. > > Overall, the whole system works great. I'm very happy with it. > > Hope this helps... > > Ted > > -- > +--- + > Ted Coady, B.Sc. ++ Finger for PGP ++ > Unix Systems Administrator, Group Telecom NF ++ public key ++ > --, <[EMAIL PROTECTED]>+--- + >\__ > "Excess on occasion is exhilarating. It prevents moderation from > acquiring the deadening effect of a habit." > -- W. Somerset Maugham -- # Christian Doppler # Unix Systemadmin # # If you really think that you can run a secure web server on Windows you've not understood security # # Unix is user-friendly... #...it's just selective about who it's friends are
