Re: amdump question
According to Joshua Baker-LePain: On Tue, 5 Feb 2002 at 7:36pm, Mary N Koroleva wrote Dear Joshua, Thank you very much for your help! I use BSDI 4.2. AMANDA is already here. As an aside, pre-compiled amanda is generally bad. You learn a lot going through the building and installation process. I have created the user amanda, have added it to group operator. amanda.conf has copied in /etc/amanda. Then: dmps:/etc/amanda | 25 amdump amanda.conf amdump: could not find directory /etc/amanda/amanda.conf dmps:/etc/amanda | 26 Would you be so kind to promt me how to compel amdump work? The amdump man page tells you the proper syntax. You're telling amanda that the configuration name is amanda.conf, and it's looking for an amanda.conf file in the configuration directory. The best ways to get started with amanda are (somewhat in this order): 1) The chapter on amanda at www.backupcentral.com 2) docs/INSTALL in the amanda source distribution 3) FAQ-O-Matic available via www.amanda.org 4) The man pages 5) This list -- Joshua Baker-LePain Department of Biomedical Engineering Duke University -- Mary N Koroleva Russian Institute E-mail: [EMAIL PROTECTED]for Phone: +7 095 737 0601Public Networks
Re: Holding disk
What was in your Amanda E-mail report for this run?
*** THE DUMPS DID NOT FINISH PROPERLY!
The next tape Amanda expects to use is: System1-001.
FAILURE AND STRANGE DUMP SUMMARY:
daemon.sys /etc RESULTS MISSING
daemon.sys /home RESULTS MISSING
daemon.sys /root RESULTS MISSING
snip
STATISTICS:
Total Full Daily
Estimate Time (hrs:min)0:00
Run Time (hrs:min) 0:00
Dump Time (hrs:min)0:00 0:00 0:00
Output Size (meg) 0.00.00.0
Original Size (meg) 0.00.00.0
Avg Compressed Size (%) -- -- --
Filesystems Dumped0 0 0
Avg Dump Rate (k/s) -- -- --
Tape Time (hrs:min)0:00 0:00 0:00
Tape Size (meg) 0.00.00.0
Tape Used (%) 0.00.00.0
Filesystems Taped 0 0 0
Avg Tp Write Rate (k/s) -- -- --
NOTES:
planner: Full dump of daemon.system:/var/qmail promoted from 23 days
ahead.
snip
DUMP SUMMARY:
DUMPER STATS TAPER
STATS
HOSTNAME DISK L ORIG-KB OUT-KB COMP% MMM:SS KB/s MMM:SS
KB/s
--- -- -
---
daemon.sys /etc
MISSING ---
daemon.sys /home
MISSING ---
daemon.sys /root
MISSING ---
snip
(brought to you by Amanda version 2.4.2p2)
What's in the corresponding amdump.NN file (on the server, in the same
directory as amanda.conf)?
amdump: start at Mon Feb 4 11:00:00 CET 2002
planner: pid 3641 executable /usr/local/libexec/planner version 2.4.2p2
planner: build: VERSION=Amanda-2.4.2p2
planner:BUILT_DATE=Fre Nov 30 18:43:15 CET 2001
planner:BUILT_MACH=Linux milestone.system 2.4.4 #3 SMP Son Sep 23
13:29:35 CEST 2001 i686 unknown
planner:CC=gcc
planner: paths: bindir=/usr/local/bin sbindir=/usr/local/sbin
planner:libexecdir=/usr/local/libexec mandir=/usr/local/man
planner:AMANDA_TMPDIR=/tmp/amanda AMANDA_DBGDIR=/tmp/amanda
planner:CONFIG_DIR=/etc/amanda DEV_PREFIX=/dev/
planner:RDEV_PREFIX=/dev/ DUMP=/sbin/dump
planner:RESTORE=/sbin/restore GNUTAR=/bin/tar
planner:COMPRESS_PATH=/bin/gzip UNCOMPRESS_PATH=/bin/gzip
planner:MAILER=/usr/bin/Mail
planner:listed_incr_dir=/usr/local/var/amanda/gnutar-lists
planner: defs: DEFAULT_SERVER=milestone.system DEFAULT_CONFIG=daily
planner:DEFAULT_TAPE_SERVER=milestone.system
planner:DEFAULT_TAPE_DEVICE=/dev/st0 HAVE_MMAP HAVE_SYSVSHM
planner:LOCKING=POSIX_FCNTL SETPGRP_VOID DEBUG_CODE
planner:AMANDA_DEBUG_DAYS=4 BSD_SECURITY USE_AMANDAHOSTS
planner:CLIENT_LOGIN=operator FORCE_USERID HAVE_GZIP
planner:COMPRESS_SUFFIX=.gz COMPRESS_FAST_OPT=--fast
planner:COMPRESS_BEST_OPT=--best UNCOMPRESS_OPT=-dc
planner: dgram_bind: socket bound to 0.0.0.0.985
READING CONF FILES...
startup took 0.003 secs
SETTING UP FOR ESTIMATES...
setting up estimates for daemon.system:/etc
setup_estimate: daemon.system:/etc: command 0, options:
last_level 1 next_level0 24 level_days 3
getting estimates 0 (2860) 1 (120) -1 (-1)
setting up estimates for daemon.system:/home
setup_estimate: daemon.system:/home: command 0, options:
last_level 1 next_level0 23 level_days 4
getting estimates 0 (1180) 1 (30) -1 (-1)
setting up estimates for daemon.system:/root
daemon.system:/root overdue 1 day for level 0
setup_estimate: daemon.system:/root: command 0, options:
last_level 0 next_level0 -1 level_days 0
getting estimates 0 (16953) 1 (0) -1 (-1)
setting up estimates for daemon.system:/share/prog/os
setup_estimate: daemon.system:/share/prog/os: command 0, options:
last_level 1 next_level0 23 level_days 4
getting estimates 0 (28250) 1 (70) -1 (-1)
setting up estimates for daemon.system:/share/prog/pub/photos
setup_estimate: daemon.system:/share/prog/pub/photos: command 0, options:
last_level 0 next_level0 27 level_days 0
getting estimates 0 (265880) 1 (0) -1 (-1)
setting up estimates for daemon.system:/share/prog/pub/transfer
setup_estimate: daemon.system:/share/prog/pub/transfer: command 0, options:
last_level 1 next_level0 24 level_days 3
getting estimates 0 (20400) 1 (10) -1 (-1)
setting up estimates for daemon.system:/var/qmail
setup_estimate: daemon.system:/var/qmail: command 0, options:
last_level 1 next_level0 23 level_days 4
getting estimates 0 (1400) 1 (210) -1 (-1)
setting up estimates for daemon.system:/var/log
setup_estimate: daemon.system:/var/log: command 0, options:
last_level 1 next_level0 25 level_days 2
getting estimates 0 (8920) 1 (2220) -1
Can you change /tmp/amanda to a more permanent directory?
Hi all, I found out, that a reboot of the amandaserver causes the night following backup to fail, because /tmp is emptied by the system after wakeup. Is there a configuration possible to change the defautl /tmp/amanda to a more permanent place? adTHANXvance Sascha
RE: Amanda install reality check
On Wed, 6 Feb 2002 at 4:26pm, Bort, Paul wrote 1. REBDA (Read Everything Before Doing Anything) Where Everything is at least the chapter at www.backupcentral.com and docs/INSTALL (plus anything else relevant in docs, like SAMBA). 1a. Read it again. ;) Really. Wrapping your head around AMANDA can take a bit, but it's well worth it. 2. Be prepared to run the configure/install process a few times until you get it the way you want. 2a. Start with a config named Testing or something like that. Set record no. Play with it. Break it. Fix it. Any other suggestions from the list? 9. Watch out for firewalls you may not expect to be there, e.g. ipchains in default RedHat installs. 10. Restart (x)inetd. Yes, again. ;) 11. If/when you ask the list for help, be as detailed as possible about the problem. We're a friendly (usually), helpful bunch, but we ain't mind readers. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: Can you change /tmp/amanda to a more permanent directory?
On Thu, 7 Feb 2002 at 10:19am, Sascha Wuestemann wrote I found out, that a reboot of the amandaserver causes the night following backup to fail, because /tmp is emptied by the system after wakeup. Is there a configuration possible to change the defautl /tmp/amanda to a more permanent place? [jlb@chaos amanda-2.4.2p2.2]$ ./configure --help Usage: configure [options] [host] Options: [defaults in brackets after descriptions] Configuration: *snip* --with-tmpdir[=/temp/dir] area Amanda can use for temp files [/tmp/amanda] -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Amanda and two servers linux
Hi: I have this configuration: 1. One RH 7.1 Linux server, with Oracle Amanda. Amanda copies into an HP SCSI streamer 24GB/48GB. Say it A. 2. Another RH 7.1 Linux server, who is file server (Samba) mailer. Say it B. 3. Amanda have to backup files from B. I have NFS partitions, but the amanda report marks FAILED. Only the nfs partitions fails. Other backups works fine. Please, help Thanks
Re: Amanda and two servers linux
On Thu, 7 Feb 2002 at 10:21am, Adolfo Manuel Pachón Rodríguez wrote 1. One RH 7.1 Linux server, with Oracle Amanda. Amanda copies into an HP SCSI streamer 24GB/48GB. Say it A. 2. Another RH 7.1 Linux server, who is file server (Samba) mailer. Say it B. 3. Amanda have to backup files from B. I have NFS partitions, but the amanda report marks FAILED. So, to clarify, you are: a) NFS exporting partitions from B b) Mounting them on A c) Adding them to the disklist as A /mnt/nfsFromB Is that correct? If so, that's not the way to do it. Only the nfs partitions fails. Other backups works fine. One of two things is going on: 1) You're using dump as your backup program, and it doesn't understand NFS (only ext2), so it's choking and dying. 2) You're using tar as your backup program, but you didn't set no_root_squash in your NFS export options from B, so tar can't see the stuff in B. As mentioned above, though, this is not the way you should be doing this. Is there any reason you can't install amanda on B and just back it up as a client? That's how amanda is designed to be used. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: amdump sending blank emails
tar release starting from 1.12 would do I think Joshua Baker-LePain wrote: On 6 Feb 2002 at 1:49pm, gene wrote I'm not sure what I have done, as I wasn't getting blank emails from amdump yesterday. I can't find anything wrong. amverify and/or amcheck send email with text ok. I don't get anything in the body of the email from amdump. 'man amrecover' will tell you how to generate a new email report from the log files. Any errors you encounter doing so may be instructive. doesnt find any files in the directory. I do see the index file being created and I can unzip it and see the files (with some long number appended in front of each). Oops. Bad version of tar. You need at least 1.13.19, which is available at alpha.gnu.org. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University -- Tom Van de Wiele System Administrator Eduline Colonel Bourgstraat 105a 1140 Brussel http://www.eduline.be
Re: amdump sending blank emails
On Thu, 7 Feb 2002 at 11:55am, [EMAIL PROTECTED] wrote tar release starting from 1.12 would do I think Only with the patches from www.amanda.org. 1.13.19 doesn't need any patches. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
arkc and tape names
Hi, has anyone been able to make arkc recognise tape names with spaces? e.g. 'INCR - 2' is a valid tape name in the gui and from the 'arkc -tape -list' command but: # arkc -tape -statistics -D name=INCR - 2 # arkc -tape -statistics -D name='INCR - 2' # arkc -tape -statistics -D name=INCR - 2 # arkc -tape -statistics -D name=INCR\ -\ 2 have all failed with: arkc error: Add '-moreinfo' option to get more information whereas with a bogus name I get: # arkc -tape -statistics -D name=bogus Error: Bad value for name arkc error: Add '-moreinfo' option to get more information any clues would be appreciated. TIA Tom -- Thomas Robinson Ehbas Ltd T: 01273 234 665 F: 01273 704 499
samba backup - sometimes yes, sometimes no
I'm doing 4 W2K partitions (single client) to a Solaris host using Samba shares. Generally all 4 work fine, but too frequently the largest partition, C, fails with smbclient received signal 13. A broken pipe, remote end terminated early I guess. This happens on both full and incremental. My configuration only allows one dump from this client at a time. Amanda 2.4.2. Any common, fixable :) reasons for such failures? -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Amanda install reality check
The reality check point of the poster who started this thread is very valuable for the amanda community. It is true: amanda *is* complicated. Given that it does not cost anything, there is no point in complaining of course and that is not what I intend to do. However in my opinion the awareness of this problem is too low. I find it hard to believe that most standard amanda using unix system operators suggest ways to install amanda this way (quote is meant as an example for the common sense of list members / amanda operators): 1. REBDA (Read Everything Before Doing Anything) This hint is so vague that it really does not help much. You should read some files from the amanda source distribution, I agree, but where is the starting point *which* files to read, given you have a certain amount of backup technology knowledge? And what will you learn? What will you have to learn? In fact, there are only hints for reading. There is no official guide how to install amanda in a standard environment although of course standard environments exist or at least can be described. 2. Be prepared to run the configure/install process a few times until you get it the way you want. Inacceptable. Why the hell should OS software installs work the try and error way only? After reading the right documents (1.) I should be able to run configure, make and make install the correct way, maybe with a few options to configure, but not more. [...] 8. Build your own. Whoever made the RPM or DEB didn't have your network in mind. Most of the knowledge put into RPM spec files by for example RedHat is exactly what is missing for amanda newbies. No you can call your amanda user amanda and the group amanda, or use group disk, but make sure the permissions for the block devices, or let the amanda user be operator, but this is not recommended, and ..., where you *will* get messages like must be run as amanda; instead decisions which are unnecessary for the end user done by the RPM packager (we use amanda/disk and it works), the correct list of chown and chmod commands in the install script and so on. It is all defined by the vendor (RedHat here), and there is basically no reason to ever change these definitions. If the RPMs do not work the way they should, fix them and provide the fix to RedHat, but saying do not use RPMs is the hard way. We use both the amanda server and client RPMs from RH 7.1 and are happy with them. Moritz
Re: Amanda install reality check
On Thu, 7 Feb 2002 22:13, Moritz Both spake thus: The reality check point of the poster who started this thread is very valuable for the amanda community. It is true: amanda *is* complicated. Given that it does not cost anything, there is no point [snip] [...] 8. Build your own. Whoever made the RPM or DEB didn't have your network in mind. Most of the knowledge put into RPM spec files by for example RedHat is exactly what is missing for amanda newbies. No you can call your [snip] them and provide the fix to RedHat, but saying do not use RPMs is the hard way. Generally .deb's have worked very well for me with a range of complicated software: Postfix, Apache+PHP, even tho' later I may've gone on to compile them for various reasons. Amanda deb's caused lots of pain, and no quick ramp up to compiling. The fact that packages are ignored/not specifically documented (10 lines for debian specific stuff, the rest the standard amanda doc's) made it overly difficult for me to get Amanda going, and I had to learn by bug-tracking and painful doco scouring to fix the packaged install because the standard documentation mislead me, and the package did not document adequately. The debian package installs with user=backup, group=backup. So far so good. The doco says to put .amandahosts in $HOME. I couldn't understand why my hosts file had no effect until I clicked that the $HOME for user backup is /var/backup, and that the debian package symlinks /var/backup.amandahosts - /etc/amandahosts. Also the user doco and the chapter on backup central says cut and paste these lines into your /etc/inetd.conf. I couldn't see why I was getting user amanda cannot connect as backup@hostname, where a client had a compiled version (it was a messy Mandrake box for which dependencies were broken and compiling seemed easier). Of course you have to put the user name that the inetd service will be run as in the inetd.conf line: amanda dgram udp wait backup /usr/sbin/tcpd /usr/lib/amanda/amandad If you're busy and you don't have huge amounts of experience with the inetd.conf format the significance of the word backup as being the user to run that amandad daemon under just doesn't jump out at you. Some interactive post install scripts for the debian packages which put sensible defaults in and explained why, would save hours of agony. -- --- Sarah Hollings [EMAIL PROTECTED] --- IT ManagerPh +61 7 3365 6080 Fax +61 7 3365 6171 Key Centre for Human Factors and Applied Cognitive Psychology The University of Queensland, Saint Lucia, QLD 4072 Australia
Re: arkc and tape names
has anyone been able to make arkc recognise tape names with spaces? U, what does arkc (whatever that is) have to do with the Amanda backup package (www.amanda.org)? Tom John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: samba backup - sometimes yes, sometimes no
... too frequently the largest partition, C, fails with smbclient received signal 13. A broken pipe, remote end terminated early I guess. Surely there is some information lurking here. For instance, my first guess is that you ran into a tape error (or end of tape), but that would have generated some more entries in the NOTES section. Jon H. LaBadie John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
recovering the information
Hello everybody: how can i recover the information of an amanda tape without the amanda software??
Re: Amanda install reality check
Moritz Both [EMAIL PROTECTED] wrote:
I find it hard to believe that most standard amanda using unix
system operators suggest ways to install amanda this way (quote is
meant as an example for the common sense of list members / amanda
operators):
1. REBDA (Read Everything Before Doing Anything)
This hint is so vague that it really does not help much. You should
read some files from the amanda source distribution, I agree, but
where is the starting point *which* files to read, given you have a
certain amount of backup technology knowledge? And what will you
learn? What will you have to learn?
A different paradigm.
When I did an evaluation of available backup solutions some time ago,
it stunned me how vendors clouded up their basic concepts in marketing
must sound different hogwash. While there is still not much more than
{full,incremental,differential} backups, it takes time and effort to
make out who is doing what.
Amanda, too, comes with her own special terminology. Watch the number of
beginners stumbling into amanda-users with How can I schedule full
backups for the weekends...?.
In fact, there are only hints for reading. There is no official
guide how to install amanda in a standard environment although of
course standard environments exist or at least can be described.
There are the UofM papers that started the whole thing, and give the
general idea. Then there is a set of how-to docs and man pages. Print
them all out, duplex, two-up, bind them as you will want them anyway
when your server is down.
2. Be prepared to run the configure/install process a few times until
you get it the way you want.
Inacceptable. Why the hell should OS software installs work the try
and error way only? After reading the right documents (1.) I should
be able to run configure, make and make install the correct way, maybe
with a few options to configure, but not more.
Inacceptable? It's about getting a feeling for what's going on.
Backup/Restore is essential, and time for trying out and getting
familiar with essential tools is never wasted.
When you need the tool most, you may find yourself with a server
half-down, having to manually reconfigure/fix permissions etc.
[...]
8. Build your own. Whoever made the RPM or DEB didn't have your network
in mind.
[...]
If the RPMs do not work the way they should, fix
them and provide the fix to RedHat, but saying do not use RPMs is
the hard way.
We use both the amanda server and client RPMs from RH 7.1 and are
happy with them.
Huh.
Amanda as she stands is very much a unix administrators' tool. Five
machines are fine, fifteen even better. That is a setup in which an
administrator who cannot work himself out of a paper bag without a set
of shiny rpms will eventually find himself in trouble.
My 0.02 EUR,
hauke
--
Hauke Fath /~\The ASCII
tangro software components GmbH \ / Ribbon Campaign
D-69115 Heidelberg X Against
Ruf +49-6221-13336-0, Fax -21 / \ HTML Email!
Re: recovering the information
On Thu, 7 Feb 2002 at 2:34pm, Monserrat Seisdedos Nuñez wrote
how can i recover the information of an amanda tape without the amanda
software??
As explained in docs/RESTORE:
mt rewind
mt fsf 1 (the first file is just a tape header)
dd if=/dev/tape of=image1 bs=32k skip=1
That will get you the first image off the tape. Repeat the dd for
subsequent images. You can then peruse the images with the appropriate
restore program (using gzip -d as appropriate).
You can also pipe the dd straight into {restore|tar x}.
--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University
Re: arkc and tape names
On Thursday 07 February 2002 08:23 am, John R. Jackson wrote: has anyone been able to make arkc recognise tape names with spaces? U, what does arkc (whatever that is) have to do with the Amanda backup package (www.amanda.org)? arkc is part of arkeia, John. Thats the high-priced spread. Nice, but legendarily expensive. -- Cheers, Gene AMD K6-III@500mhz 320M Athlon1600XP@1400mhz 512M 98.5+% setiathome rank, not too shabby for a hillbilly
Re: Cron and Amdump
John, Thanks for the great advice. Funny, just before I got your email, I set up the cron the following way: 45 0 * * 2-6/usr/sbin/amdump BIG1 /tmp/debug This morning, I have the following in /tmp/debug amdump: could not find directory /etc/amanda/BIG I am not sure why BIG1 is being truncated to BIG, and why did it just start to happen. Amanda home directory/login is valid (su - amanda -c pwd). I tested that last week. I am running last nights dump right now, but I will test the cron again by changing its time to a few minutes in the future for testing. Thanks. John R. Jackson wrote: For some strange reason, cron has stopped running Amdump for me at night. ... Here is Amanda's crontab: 0 16 * * 1-5/usr/sbin/amcheck -m BIG1 45 0 * * 2-6/usr/sbin/amdump BIG1 Here are some debugging things I would try: * Change the amdump line to something like this: -- Cheers, Karl Bellve, Ph.D. ICQ # 13956200 Biomedical Imaging Group TLCA# 7938 University of Massachusetts Email: [EMAIL PROTECTED] Phone: (508) 856-6514 Fax: (508) 856-1840 PGP Public key: finger [EMAIL PROTECTED]
Re: Amanda install reality check
On Thu, 7 Feb 2002, Hauke Fath wrote: Amanda as she stands is very much a unix administrators' tool. Five machines are fine, fifteen even better. That is a setup in which an administrator who cannot work himself out of a paper bag without a set of shiny rpms will eventually find himself in trouble. I think amanda falls under the qmail/smtpd category. If you have trouble installing it, maybe you should rethink whether you are qualified to implement the backup procedure. I'm not trying to be rude or take the elite attitude, but there are some systems that should only be admined by qualified individuals; to do otherwise is to create an environment for disaster. Amanda is a backup solution. Backup is a key element to any network setup, it is not something one should just implement and forget about. This is why so many are fed up with many of the commercial backup products, those products tend to hide as much of the details as possible. But knowing those details can at times mean the difference between huge data loss or saving the day yet again. Amanda may take a bit of effort to install, but in the process one learns quite a bit about how it works and what to expect. And once learned, further amanda installs are like falling off a log. $.02 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: recovering the information
That depends on how you backed it up. Look at the header on the tape: 1. Load the tape on your tape drive 2. Skip the first file set (amanda label) using: mt -f norewind device fsf 1 3. Run: dd if=norewind device bs=32k On one of my tapes that shows: AMANDA: FILE 20020131 host1 /fs2 lev 0 comp N program /usr/sbin/ufsdump To restore, position tape at start of file and run: dd if=tape bs=32k skip=1 | /usr/sbin/ufsrestore -f... - So to restore from that tape, I would: 1. Rewind the tape: mt -f /dev/rmt/2cbn rew (no rewind device is used) 2. mt -f /dev/rmt/2cbn fsf 1 (skip the amanda header; no rewind device is used) 3. dd if=/dev/rmt/2cbn bs=32k skip=1 | /usr/bin/ufsrestore -x -f - You will need to use the appropriate restore command. This one is on a Solaris system. This will only restore the first file set. If you wish to restore additional file sets, you must position the tape to the beginning of the appropriate set. For instance, if you wish to restore the 33 fileset you would change the mt command to read: mt -f no rewind device fsf 33 to skip the amanda header record and the first 32 file sets. Hello everybody: how can i recover the information of an amanda tape without the amanda software?? --- Wayne Richards e-mail: [EMAIL PROTECTED]
RE: Amanda install reality check
--On Wednesday, February 06, 2002 22:20:47 -0600 W. D. [EMAIL PROTECTED] wrote: At 16:46 2/6/2002, Frank Smith, wrote: Forget about all fulls on weekend, incrementals weekdays Does this mean do full backups each time? No. it just means you might need to change your mindset from 'traditional' backup methods, where you might run full backups on Friday night (or the 1st of the month or whatever) and incrementals on other nights, to letting Amanda intermingle full and incrementals on each run, thereby getting better use out of your tape capacity. We use amanda for our 'traditional' backups, via multiple configs and cron. It works just fine, but OTOH, we aren't hurting for tape space. I've toyed with the idea of letting amanda just do its thing, but if more efficient tape use is the only advantage then I can't really justify the time/effort/pain it would take to configure and tweak it. Are there any other reasons to move away from our traditional ways? --Chris
Re: Amanda install reality check
--On Thursday, February 07, 2002 13:13:37 +0100 Moritz Both [EMAIL PROTECTED] wrote: 2. Be prepared to run the configure/install process a few times until you get it the way you want. Inacceptable. Why the hell should OS software installs work the try and error way only? After reading the right documents (1.) I should be able to run configure, make and make install the correct way, maybe with a few options to configure, but not more. I agree with you, but Amanda has way too much config information compiled in that (in my opinion) is better suited for the config file. If the paths/users/portranges/etc were read from the config file then there would be little need for recompiling. Frank -- Frank Smith [EMAIL PROTECTED] Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
Re: Cron and Amdump
Funny, just before I got your email, I set up the cron the following way: 45 0 * * 2-6/usr/sbin/amdump BIG1 /tmp/debug This morning, I have the following in /tmp/debug amdump: could not find directory /etc/amanda/BIG I am not sure why BIG1 is being truncated to BIG, and why did it just start to happen. Two things. The output redirect you added, , is a csh-ism. Is your cron running the amanda entries with that shell, or is it running them with /bin/sh? One way to tell would be to add a temporary entry like this: ... ps -p $$ /tmp/cron.shell.test Second, it looks like the '1' on the end of BIG1 got mis-interpreted by the shell (or you have a typo). The '1' got put together with the output redirection as though you had done this: /usr/sbin/amdump BIG 1 /tmp/debug which means redirect file descriptor 1 (stdout) to this file. That also implies cron is using some sh variant (sh, ksh, bash) rather than csh, so you should really be using this syntax (which is the equivalent of ): /usr/sbin/amdump BIG1 /tmp/debug 21 Karl Bellve John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: Amanda install reality check
Bill Carlson [EMAIL PROTECTED] wrote: [...] I think amanda falls under the qmail/smtpd category. If you have trouble installing it, maybe you should rethink whether you are qualified to implement the backup procedure. Well... I think if you don't have trouble installing amanda for the first time, you are either using very good step by step instructions stating how it is done within your organization... or you are God :) I'm not trying to be rude or take the elite attitude, but there are some systems that should only be admined by qualified individuals; to do otherwise is to create an environment for disaster. Totally agreed. I also agree backups are a key element not only of networks, but of corporations. But this does not qualify as a reason why amanda installation must be complicated and can't be done using RPM. It is also true that the (approx.) month of work time that passes before your amanda backups are safe and in production in a let's say 10 client environment is valuable for the person who does it. You are getting very good knowledge of the internal processes of amanda almost automatically by studying the /tmp/amanda/* and amdump files. However, this will not help your organization when you are gone. Another guy will start learning it (installing, supervising, restoring) all over unless you have written the detailed step by step instructions mentioned above. But if you *have* written these instructions for all amanda tasks for your organization so you are replaceable, they will certainly contain shell scripts which install etc. - close to the same scripts that are in the RPM package. By the way, I do not believe in every network is different. Every network is similar. There are many more common things in all networks than differences. Greetings, Moritz
Re: Amanda install reality check
First I want to say thanks to the folks contributing to this thread. I'll be the first to admit the Amanda documentation could use more work (of course, I'd be the first to say that about almost *any* software :-). I'm watching the comments go by and will try to incorporate as much as possible. To save some time (which translates to getting things improved faster), I would ask that specific examples be cited where possible. Saying the whole thing sucks gets the point across :-) but doesn't help figure out how to make it better. It's much more helpful to suggest a specific paragraph or reference would be better if it said XXX. We use amanda for our 'traditional' backups, via multiple configs and cron. ... Are there any other reasons to move away from our traditional ways? I think you answered your own question :-). Look at the effort you had to put in (multiple configs, multiple cron entries) to bend Amanda into doing things the traditional way. Reduced effort is another reason for letting Amanda do its thing. That being said, though, comfort is a very important thing when it comes to backups. If you understand your setup and it works for you, that's by far the most important thing. --Chris John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Q re. tapeless amanda + windows
I'm fairly new to amanda - here's what I'm trying to accomplish and I'd like to know if Amanda can do this. I can't find too much about this online so hence my post. I've got the following situation: 1. A unix machine with disk space, but no tape backup capabilities. 2. Several windows clients whose IP is reasonably static, but does occassionally change. (DHCP with long leases) that need to be backed up. 3. I have to be able to extract amanda's backups to a directory on the server. (I.e. they have to be something readable and non-proprietary, i.e. a tar or something, or if not, at least some way to script amanda to extract) I'm trying to set up amanda to backup these machines, I'll also have to make a web management system for amanda. My question is: Can amanda do this kind of thing? Has anyone done this? If so, how? (configuration files are welcome if you don't mind sending). A FAQ or some sort of doc would be great. I've started to read through some of the docs but I'd like to get this up and running ASAP and tweak it later - so any help one can offer would be greatly appreciated. Thanks cosimo
Re: DLT2700xt Changer conf
Hi, On Tue, Feb 05, 2002 at 04:33:09PM -0700, John Gonzalez, Tularosa Communications wrote: I'm new to amanda and dont see anything specific to this drive/changer. I do see some generic DLT configurations for DLT drives, and see something that should work. However, I have no clue how to get this working with the 7 tape changer unit. Any help and any of the glue scripts that people are using would be appreciated. Any and all info appreciated, TIA. Hmm, could you give some more infos about the OS Thomas -- --- | Thomas Hepper[EMAIL PROTECTED] | | ( If the above address fail try ) | | ( [EMAIL PROTECTED])| ---
[data timeout]
I've been having problems getting a successful level 0 backup from one of our servers. The amreport shows the following: FAILURE AND STRANGE DUMP SUMMARY: lopt /dev/hda1 lev 0 FAILED [data timeout] lopt /dev/hda1 lev 0 FAILED [dump to tape failed] Here are some of the FAILED AND STRANGE DUMP DETAILS: DUMP: Date of this level 0 dump: Wed Feb 6 23:22:40 2002 ... DUMP:Volume 1 started with block 1 at: Wed Feb 6 23:25:16 2002 ... DUMP 2.42% done at 1455 kB/s, finished in 3:21 ... DUMP: 61.83% done at 1543 kB/s, finished in 1:14 This is were details ends. Incidently, it is about this far into the dump, when dump fails. Does anyone know what causes a data timeout ? I thought it might have been my dtimeout setting, so I increased that from the default setting of 1800 seconds, but this had no effect. Thanks, Ben
Re: Q re. tapeless amanda + windows
On Thu, 7 Feb 2002 at 11:34am, [EMAIL PROTECTED] wrote 1. A unix machine with disk space, but no tape backup capabilities. You'll want amanda with tapeio support. That's either the tapeio branch out of CVS or 2.4.3b2 (note that's a beta version). This treats files on disk just like they're tapes. 2. Several windows clients whose IP is reasonably static, but does occassionally change. (DHCP with long leases) that need to be backed up. If their WINS names don't change, I think you should be fine (but I don't know enough (thankfully) about Windows networking to be sure). Amanda backs up 'doze clients via smbclient, and the disklist entries look like this: $SAMBA_SERVER //$DOZECLIENT/$SHARE where $SAMBA_SERVER is a *nix box with Samba installed, $DOZECLIENT is the Windows box you want backed up, and $SHARE is the specific share you want to back up. 3. I have to be able to extract amanda's backups to a directory on the server. (I.e. they have to be something readable and non-proprietary, i.e. a tar or something, or if not, at least some way to script amanda to extract) And this is amanda's strength. Amanda only schedules and execute backup programs. It doesn't actually get the bits off the disks. For that it relies on either a vendor supplied dump program or GNUtar. For Samba clients, it'll be tar (via smbclient). I'm trying to set up amanda to backup these machines, I'll also have to make a web management system for amanda. My question is: Can amanda do this kind of thing? Has anyone done this? If so, how? (configuration files are welcome if you don't mind sending). A FAQ or some sort of doc would be great. I've started to read through some of the docs but I'd like to get this up and running ASAP and tweak it later - so any help one can offer would be greatly appreciated. docs/INSTALL and docs/SAMBA are your friend. Set up your *nix server (make sure you have Samba installed), and get it to back up itself. Once that's done, adding the win clients should be trivial. You're on your own (AFAIK) as far as the web management goes. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: [data timeout]
On Thu, 7 Feb 2002 at 11:47am, Benjamin Gross wrote I've been having problems getting a successful level 0 backup from one of our servers. The amreport shows the following: FAILURE AND STRANGE DUMP SUMMARY: lopt /dev/hda1 lev 0 FAILED [data timeout] lopt /dev/hda1 lev 0 FAILED [dump to tape failed] Here are some of the FAILED AND STRANGE DUMP DETAILS: DUMP: Date of this level 0 dump: Wed Feb 6 23:22:40 2002 ... DUMP:Volume 1 started with block 1 at: Wed Feb 6 23:25:16 2002 ... DUMP 2.42% done at 1455 kB/s, finished in 3:21 ... DUMP: 61.83% done at 1543 kB/s, finished in 1:14 This is were details ends. Incidently, it is about this far into the dump, when dump fails. Let me guess -- this is a Linux client with a 2.4 kernel? There have been numerous reports of date timeouts using dump on recent distros. Without starting that flamewar up again (sorry John!), there are two solutions, one of which may work and one of which will work: 1 (may work): Upgrade your dump/restore to the latest version available from http://dump.sourceforge.net. I haven't used dump with amanda in a while, so I have no idea if this will work. 2 (will work): Switch your dumptype to use tar instaed. Of course you need to decide if this solution will work for you, but it will get rid of the data timeouts. (Incidentally, this is what I did.) And, if I guessed wrong, then you need to look in your system logs for anything strange going on with that disk. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: Holding disk
I think I see what's happening. Driver (the program that runs things in Amanda) first tells taper (the tape management piece) to get started, then while that is going on, it does the planner (estimates) steps. It starts taper early to give it time to deal with robots, human operators, etc. When the estimates are done, however, driver has to wait for taper to say it is ready in case there is an error and driver has to drop into degraded mode right from the start. That's where your setup is hanging. Taper is never responding. Note the following right after taper is told to start: changer: opening pipe to: /usr/local/libexec/chg-manual -info changer: got exit: 0 str: 0 99 1 changer_query: changer return was 99 1 changer_query: searchable = 0 changer_find: looking for System1-001 changer is searchable = 0 changer: opening pipe to: /usr/local/libexec/chg-manual -slot current You have configured Amanda to use the manual tape changer. My guess is that it's sitting there waiting for you to mount the tape and that's why driver is not able to go any further. If you don't have a tape changer, the easiest thing to do is just point Amanda at that device with tapedev and not set any of the changer options (tpchanger in particular). If you want to use chg-manual and you are not running amdump from a tty session, you need to look at the comments in that script for how to alter it to notify you of mount requests (e.g. via E-mail or system log message). Tom John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: [data timeout]
On Thu, Feb 07, 2002 at 12:01:03PM -0500, Joshua Baker-LePain wrote: On Thu, 7 Feb 2002 at 11:47am, Benjamin Gross wrote Let me guess -- this is a Linux client with a 2.4 kernel? There have been numerous reports of date timeouts using dump on recent distros. Without starting that flamewar up again (sorry John!), there are two solutions, one of which may work and one of which will work: yes, this is a linux box that has been recently upgraded to 2.4 kernel. Thank you for the fast response and the suggestions. Regards, Ben
Re: before I recreate the wheel.....
never mind the nieve nature of the question...I have the chg-zd-mtx working (minus all of the req. for a cleaning tape which I removed) Don Don Potter wrote: Does anybody have a changer.conf for an Sun StorEdge L280...I'm using MTX (1.2.15) on a Solaris 8 box. I haven't seen one in the arcchive by chance. Thanks, Don
amcheck
When I run su amanda -c amcheck daily the backup server times out with the selfcheck request timed out error. Believe it or not, this just started happening, and I _haven't_ changed anything. The funny part is this: amdump works fine and runs nightly! It backs-up the server, even tho amcheck says it's down. I'm using the latest stable version with RH7.2 The system has been running fine for the last month (it's only a test setup so this isn't that big of a problem). I have rechecked .amandahosts; made sure xinetd was setup properly; emptied the hosts.deny file. Added ALL: ALL to the hosts.allow file, I doubled the etimeout from 300 to 600, but amcheck still times out. Has anyone else ever encountered a similar problem? Thanks, Brad
Re: amcheck
On Thursday 07 February 2002 15:19, R. Bradley Tilley wrote: When I run su amanda -c amcheck daily the backup server times out with the selfcheck request timed out error. Believe it or not, this just started happening, and I _haven't_ changed anything. The funny part is this: amdump works fine and runs nightly! It backs-up the server, even tho amcheck says it's down. I'm using the latest stable version with RH7.2 The system has been running fine for the last month (it's only a test setup so this isn't that big of a problem). I have rechecked .amandahosts; made sure xinetd was setup properly; emptied the hosts.deny file. Added ALL: ALL to the hosts.allow file, I doubled the etimeout from 300 to 600, but amcheck still times out. Has anyone else ever encountered a similar problem? Thanks, Brad I forgot to tell you guys that all the other AMANDA clients backup just fine. I have about 4 Linux servers, and 45 Windows PCs that AMANDA backs up. It's only the AMANDA backup server itself that times out. Thanks Again.
Re: LTO and changers
On Wed, 6 Feb 2002, John R. Jackson wrote: I know that amanda can not split a dump across multiple tapes, so if a dump was too large to fit on the tape, will amanda write the dump to the next available tape in the changer? Do you mean a bunch of stuff was written to the first tape and then one of the images hit EOT, but the image itself is smaller than a tape? If so, then yes, Amanda will go on to the next tape and start the image over again there (assuming you have configured it to do so). Yes that is what I was asking. I guess the question was a little ambiguous. Does amanda support more than one tape device, and if so can it write to more then one device at a time? Not yet. OK. Thanks again. This list is great! Drew
Re: [Amanda-users] Re: [data timeout]
I have the data timeout problem for a while and for me I've narrowed it down to a particular string causing checksum failures in the tcp stuck on a particular motherboard. I spent some time going back and forth with David Miller and have so far got no where. I can duplicate this with a 735 byte string If I stick the file with the string on any partitions they fail, and if I remove the file it works. The motherboard I'm seeing this on is an ASUS A7V133 I've built a kernel with the minimal requirements to make the machine boot. and tried every 2.4 kernel from 2.4.0 to 2.4.18-pre7. I attempted to try a 2.2 kernel but it didn't support the tulip card in the machine. also its not the network card as I've tried others. The last byte of the packet contating the 735 byte string is corrupted. The problem will go away if you use compression as the string is obviously changed before it travels over the wire. I have no idea where to go now. On Thu, Feb 07, 2002 at 12:01:03PM -0500, Joshua Baker-LePain wrote: On Thu, 7 Feb 2002 at 11:47am, Benjamin Gross wrote I've been having problems getting a successful level 0 backup from one of our servers. The amreport shows the following: FAILURE AND STRANGE DUMP SUMMARY: lopt /dev/hda1 lev 0 FAILED [data timeout] lopt /dev/hda1 lev 0 FAILED [dump to tape failed] Here are some of the FAILED AND STRANGE DUMP DETAILS: DUMP: Date of this level 0 dump: Wed Feb 6 23:22:40 2002 ... DUMP:Volume 1 started with block 1 at: Wed Feb 6 23:25:16 2002 ... DUMP 2.42% done at 1455 kB/s, finished in 3:21 ... DUMP: 61.83% done at 1543 kB/s, finished in 1:14 This is were details ends. Incidently, it is about this far into the dump, when dump fails. Let me guess -- this is a Linux client with a 2.4 kernel? There have been numerous reports of date timeouts using dump on recent distros. Without starting that flamewar up again (sorry John!), there are two solutions, one of which may work and one of which will work: 1 (may work): Upgrade your dump/restore to the latest version available from http://dump.sourceforge.net. I haven't used dump with amanda in a while, so I have no idea if this will work. 2 (will work): Switch your dumptype to use tar instaed. Of course you need to decide if this solution will work for you, but it will get rid of the data timeouts. (Incidentally, this is what I did.) And, if I guessed wrong, then you need to look in your system logs for anything strange going on with that disk.
backups still failing.
okay I've got to a point where I know its not amanda's fault but someone here should be able to help. where doing Daily backups to DDS3 tapes with a block size of 4096, which seem to work sometimes. more often than if I use a block size of 0. and where doing Monthly backups to DDS4 tapes with a block size of 4096, and they completely fail with the error messages below. any ideas. - Forwarded message from backup [EMAIL PROTECTED] - Date: Fri, 8 Feb 2002 00:16:54 +1100 From: backup [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: MonthlySet1 AMANDA VERIFY REPORT FOR MonthlySet103 MonthlySet104 MonthlySet105 MonthlySet101 MonthlySet102 Tapes: MonthlySet103 MonthlySet104 MonthlySet105 MonthlySet101 MonthlySet102 Errors found: MonthlySet103 (sobek.tsa._sobek.20020205.0): amrestore: WARNING: not at start of tape, file numbers will be offset amrestore: 0: restoring sobek.tsa._sobek.20020205.0 /bin/tar: Skipping to next header /bin/tar: Skipping to next header /bin/tar: Error exit delayed from previous errors 64+0 records in 64+0 records out
[no subject]
Sorry for the confusion. When I run the amrecover on the TRU64 client Iit looks like this. amrecover -C config -s backup host-t backup host-d /dev/nst0 This Produces AMRECOVER Version 2.4.2p2. Contacting server on backup host ... 220 backup host AMANDA index server (2.4.2p2) ready. 200 Access OK Setting restore date to today (2002-02-07) 200 Working date set to 2002-02-07. 200 Config set to config. 200 Dump host set to TRU64 server. $CWD '/' is on disk '/' mounted at '/'. Scanning /hdisk1... 200 Disk set to /. No index records for disk for specified date If date correct, notify system administrator Invalid directory - / If I do amrecover -C config -s backup host-t backup host-d /dev/nst0 on the backup host then use sethost Tru 64 server and setdisk / I get AMRECOVER Version 2.4.2p2. Contacting server on backup host ... 220 backup host AMANDA index server (2.4.2p2) ready. 200 Access OK Setting restore date to today (2002-02-08) 200 Working date set to 2002-02-08. 200 Config set to config 200 Dump host set to backup host $CWD '/' is on disk '/' mounted at '/'. 200 Disk set to /. / amrecover sethost TRU 64 server 200 Dump host set to TRU 64 server. amrecover setdisk / 200 Disk set to /. amrecover ls 2002-02-08 #.mrg...login 2002-02-08 (null)/ 2002-02-08 .TTauthority 2002-02-08 .Xauthority 2002-02-08 .bash_history 2002-02-08 .cshrc 2002-02-08 .dt/ 2002-02-08 .dtprofile 2002-02-08 .login 2002-02-08 .netscape/ 2002-02-08 .new...cshrc 2002-02-08 .new...login 2002-02-08 .new...profile 2002-02-08 .new..DXsession 2002-02-08 .profile 2002-02-08 .proto...cshrc 2002-02-08 .proto...login 2002-02-08 .proto...profile So any idea why I can access the inex from the backup host but not the remote server ?? :) Thank you for your time [EMAIL PROTECTED] 02/08/02 02:24 AM The TRU64 Client will backup successfully, but when i run an amrecover it says that it cant find the index files. Could you post the exact transcript of what you're seeing. According your notes: 3. I do an amrecover on the backup host then use sethost and setdisk to select the TRU64 server 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] __ The contents of this e-mail are privileged and/or confidential to the named recipient and are not to be used by any other person and/or organisation. If you have received this e-mail in error, please notify the sender and delete all material pertaining to this e-mail. __
Re: amrecover and index files
When I run the amrecover on the TRU64 client Iit looks like this. ... amrecover -C config -s backup host-t backup host-d /dev/nst0 200 Dump host set to TRU64 server. ... If I do amrecover -C config -s backup host-t backup host-d /dev/nst0 on the backup host then use sethost Tru 64 server and setdisk / I get ... amrecover sethost TRU 64 server 200 Dump host set to TRU 64 server. You've edited the output (which is OK, I understand why), but when you look at these two 200 lines, very, very, carefully, is the host name exactly the same? Amanda uses that name (plus the disk name, so that would be another thing to check, although I assume you didn't change / to something else) to find the index files and also to look up information in the log files. If things don't match up right (for instance, a short host name vs. a fully qualified name), it will lead to trouble. If that doesn't help, I'll need to see the un-edited amrecover*debug (from the machine you run amrecover on) and amindexd*debug (from the server) side. Feel free to send that to me offline of the mailing list. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
RE: recovering the information
OK, Question, if I may interlope here... You say that you can position the tape at the beginning of a fileset? If you wish to restore additional file sets, you must position the tape to the beginning of the appropriate set. For instance, if you wish to restore the 33 fileset you would change the mt command to read: mt -f no rewind device fsf 33 to skip the amanda header record and the first 32 file sets. How does one know where a particular fileset is on the tape? That is, if I want to restore a backup of ClientX, then I assume that of the many clients backed up to the tape, and there will be a fileset for each client? How then would I know which fileset is ClientX? Or am I misunderstanding? Thanks, B Moro -Original Message- From: Wayne Richards [mailto:[EMAIL PROTECTED]] Sent: Thursday, February 07, 2002 7:20 AM To: Monserrat Seisdedos Nuñez Cc: Amanda-Users (E-mail); [EMAIL PROTECTED] Subject: Re: recovering the information That depends on how you backed it up. Look at the header on the tape: 1. Load the tape on your tape drive 2. Skip the first file set (amanda label) using: mt -f norewind device fsf 1 3. Run: dd if=norewind device bs=32k On one of my tapes that shows: AMANDA: FILE 20020131 host1 /fs2 lev 0 comp N program /usr/sbin/ufsdump To restore, position tape at start of file and run: dd if=tape bs=32k skip=1 | /usr/sbin/ufsrestore -f... - So to restore from that tape, I would: 1. Rewind the tape: mt -f /dev/rmt/2cbn rew (no rewind device is used) 2. mt -f /dev/rmt/2cbn fsf 1 (skip the amanda header; no rewind device is used) 3. dd if=/dev/rmt/2cbn bs=32k skip=1 | /usr/bin/ufsrestore -x -f - You will need to use the appropriate restore command. This one is on a Solaris system. This will only restore the first file set. If you wish to restore additional file sets, you must position the tape to the beginning of the appropriate set. For instance, if you wish to restore the 33 fileset you would change the mt command to read: mt -f no rewind device fsf 33 to skip the amanda header record and the first 32 file sets. Hello everybody: how can i recover the information of an amanda tape without the amanda software?? --- Wayne Richards e-mail: [EMAIL PROTECTED]
