RE: how to make .amandahosts
On the server, you can usually just do this (as the amanda user!) echo `hostname` `whoami` ~/.amandahosts If you use the same username for AMANADA on clients and servers, you can do the same on clients, just put the server FQDN in place of `hostname`. -Original Message- From: Pablo Quinta Vidal [mailto:[EMAIL PROTECTED] Sent: Wednesday, March 31, 2004 10:38 AM To: [EMAIL PROTECTED] Subject: how to make .amandahosts Hello Due to all the problems I had I decided to reinstall Amanda. The problem is that I cant find the .amandahost file. Where is it? or where and how should I create it? Please send me an exaple file. Thanks. _ Horóscopo, tarot, numerología... Escucha lo que te dicen los astros. http://astrocentro.msn.es/
RE: Second config
Hey eveyrone, Now that I've got a set of tapes running, I'm trying to figure out how to include another set of tapes. What I would like to do is have an identical set of tapes same number, same tapecycle, etc., but named sligtly differently that I can rotate off-site. I haven't done this with two configurations, but I've done it with one, just by setting the *cycle numbers correctly in amanda.conf. If tapecycle is 2*dumpcycle, you will have two complete sets of backup tapes. In my case, each set was a magazine that loaded into the changer, so they were easy to keep separate. It just worked. If you want to be more explicit about it, you could write a couple of shell scripts: Odd weeks: amadmin DailySet1 reuse DailySet001 amadmin DailySet1 reuse DailySet002 amadmin DailySet1 reuse DailySet003 amadmin DailySet1 reuse DailySet004 amadmin DailySet1 reuse DailySet005 amadmin DailySet1 no-reuse DailySet006 amadmin DailySet1 no-reuse DailySet007 amadmin DailySet1 no-reuse DailySet008 amadmin DailySet1 no-reuse DailySet009 amadmin DailySet1 no-reuse DailySet010 Even weeks: amadmin DailySet1 reuse DailySet006 amadmin DailySet1 reuse DailySet007 amadmin DailySet1 reuse DailySet008 amadmin DailySet1 reuse DailySet009 amadmin DailySet1 reuse DailySet010 amadmin DailySet1 no-reuse DailySet001 amadmin DailySet1 no-reuse DailySet002 amadmin DailySet1 no-reuse DailySet003 amadmin DailySet1 no-reuse DailySet004 amadmin DailySet1 no-reuse DailySet005 You could even get creative with a status flag file, combine them into one, and cron it to run weekly. But I don't see the need for two configurations. If you really want two configurations, the way I did it was to copy the DailySet1/ directory to a different name in the same parent. Do this in as many places as it appears, update the config file with the new config name and directory, and wipe the tape list. (I suppose you could clean up the curinfo and logs, but I never bothered.)
RE: Amanda GUI
For configuration, testing and restore purposes for newbies like me. It's OK to be a newbie, but the best way to get over it is to dive in, find something you don't understand, and figure it out. If that leads you to something else you don't get, figure that out too. Take notes in your own words, so you can find your way back out of the jungle. You're completely right, the GUI shouldn't make things to simple. An administrator should understand how things work. IMHO, a GUI would be nice for the not-so-skilles _users_. I regularly hear about the nice GUI of Tivoli (the system formerly known as ADSM ;) ). To be honest, I never used that GUI and I know it's not nearly as capable in configuring your backup as writing your dsm.sys (that's one of the client configuration files) manually, but people are capable to do their backup with it. I'm working on at least part of a web-based GUI for AMANDA between now and the end of the year. I'd like to release it to the community after it's done if I can clear that with the company. Our reason for an AMANDA GUI? We're deploying AMANDA at our client sites as a backup solution for a document imaging system, and they're mostly Windows or AS/400 shops. They're doing command-line backups now, and they're not doing them very often. They would do their backups much more often if it was a matter of looking at a web page for the next tape to use (amadmin DailySet1 tape), loading that tape, and hitting a button to check configuration (amcheck DailySet1), then hitting another button to run the backup (amdump DailySet1). They already know that a restore is going to involve a support call, so I don't have to write a restore interface. Yet. Perhaps I find the time to implement something like this in the next weeks...but to be honest before having a GUI for amanda I would prefer seeing a windows client *duck* :) No need to duck! Lots of people would like this, just search for 'SAMBA' in the mailing list archives. I looked at this back in the day, and it breaks down to two parts: A tar clone that can backup and restore files under NT, and amandad. There are instructions for building amandad under Cygwin, and there was a SourceForge project for an AMANDA client that had (IIRC) 90% of a working NT tar, with ACLs. If you combine the two with a little elbow grease, you could have a Windows client in an afternoon. If you need to do it without a full Cygwin installation, you could probably just copy the right files and registry entries out of an existing one. (I know you can run Cygwin's ssh.exe with just one other DLL, I think it's cygwin1.dll.)
RE: Amanda GUI
Subject: Amanda GUI Does one exist? In our minds and hearts, yes. If you want to download it, you'll have to write it first.
RE: amcheck - why run it?
Lately, with the advent of some network viruses, I've had backups not occur when I run amcheck before running amdump. My script, which is run from crontab, checks to see if amcheck ran successfully before I allow amdump to run. I'm using samba to back up Windows servers across the network to my Sun server. Is there any good reason to have this type of procedure in place where amcheck must be successful before running amdump? Instead of making amdump run in lockstep to amcheck, you might want to just have amcheck send you an e-mail if there is a problem. (See the -m option.) Then amdump will at least try to run even if amcheck failed. Also, I don't run amverify either because I lack the time during the backup window for this. I'm not sure this is necessary either. What is amverify really checking? I don't think it checks what was dumped on tape vs what is on the disk. Is it necessary to make sure you really have a valid backup? I never really worried about amverify, because it only makes sure your backup ran. Instead, I would recommend doing a test restore at least once a month, which makes sure the backup worked, like amverify AND it makes sure that you can restore. (In our case, I was backing up SQL Server databases, and made sure that I could restore the database dump all the way from tape-amanda-samba-SQL Server.)
RE: Crashing machine
I have to second Paul on this one. If you would like, you could go through any number of processes to make sure the AMANDA code is the same as it was a month ago when everything was fine (MD5SUM against backups, Re-install, etc.) and I expect you will still have the problem, which leaves either (a) non-AMANDA software, or (b) hardware. As the hardware is subject to thermal stress, and the sofware is only subject to stray alpha particles and disk error, I'd lean towards hardware. As a test, if you have another server available, you could build AMANDA on another machine and move the tape subsystem over. If the problem doesn't follow... -Original Message- From: Paul Bijnens [mailto:[EMAIL PROTECTED] Sent: Thursday, September 18, 2003 4:52 PM To: Crashers, Bart -- MFG, Inc. Cc: Amanda Users (E-mail) Subject: Re: Crashing machine Crashers, Bart -- MFG, Inc. wrote: Any ideas here? Anyone heard of such a thing? Am I barking up the wrong tree thinking that amanda might be responsible for my crashes? It's a real Amanda just stresses your hardware a lot, uses a lot of cpu/memory (gzip), loads the disks (holdingdisk), network (scsi-)bus, and will much faster trigger the problem. I still suspect hardware.
RE: Amanda win32 Client
I also looked at the project on sourceforge for the Win32 client. The task that I found too daunting to contemplate when I tried this was interfacing to the Win32 file system. The win32 client project included a TAR that they claimed worked on NT and picked up ACLs. So really the remaining task is a TCP/IP daemon that can accept and execute commands from the AMANDA server. Maybe I'll dig up some of my old Perl scripts... -Original Message- From: Christophe Kalt [mailto:[EMAIL PROTECTED] Sent: Thursday, September 11, 2003 11:35 AM To: [EMAIL PROTECTED] Subject: Re: Amanda win32 Client On Sep 03, Kurt Yoder wrote: | Anyone on this list interested in being paid to re-do the Windows | client? Ideally, I'd want something with an easy-to-use installer. | Should work more consistently than the current Windows client, and | file restores should be simple. I don't care if it's a ground-up | reimplementation, or what language it's in. You should.. this shouldn't be a reimplementation, just a port. The old win32 client was kind of a port, but since it was done separately and not merged, it drifted off the unix code. Not fun to do, i started looking into it a while back, but the best approach by far.
RE: make a backup disappear
You have two separate problems you need to address: First, letting AMANDA know that that tape is not valid anymore. Second, adjusting your backup schedule to replace the lost backups. The first step is to remove the tape from the tape database with the 'amrmtape' command, which expects a configuration name and tape label as parameters. It will handle tapelist and the log files. IIRC, you won't need to do anything special for the second step, because when AMANDA examines the logs to determine which levels are needed, the backups done on the removed tape won't be counted. You can use "amadmin due" to see which disks need a level 0. -Original Message-From: Jeremy L. Mordkoff [mailto:[EMAIL PROTECTED]Sent: Friday, August 22, 2003 5:29 PMTo: [EMAIL PROTECTED]Subject: make a backup disappear Hi- I just overwrote a backup tape with a tar file. How do I tell Amanda that that backup no longer exists and that it should try to reschedule as many fulls as possible for tonight? Do I just find the right log in /var/log/Amanda/conf and delete it? Thanks, JLM Jeremy Mordkoff Tatara Systems 978-206-0808 (direct) 978-206-0888 (fax) injustice anywhere threatens justice everywhere -- Dr. Martin Luther King
RE: configure fails even with-group specified
Maybe because there are no spaces between the settings, so they don't parse? Try this: ./configure --with-user=vilbu\ --with-group=vilbu\ --with-configdir=/etc/amanda\ --with-config=daily\ --with-gnutar=/bin/tar\ --with-tapedevice=/dev/nst0\ --with-amandahosts\ --with-smbclient=/usr/bin/smbclient ^ note leading spaces... -Original Message- From: S. Keel [mailto:[EMAIL PROTECTED] Sent: Wednesday, July 30, 2003 6:06 PM To: Amanda Users Subject: configure fails even with-group specified Hi all, I'm a little confused as to why configure fails. Here's what I'm using when I run configure... ./configure --with-user=vilbu\ --with-group=vilbu\ --with-configdir=/etc/amanda\ --with-config=daily\ --with-gnutar=/bin/tar\ --with-tapedevice=/dev/nst0\ --with-amandahosts\ --with-smbclient=/usr/bin/smbclient ...but the response I get is this... checking build system type... i686-pc-linux-gnu checking host system type... i686-pc-linux-gnu checking target system type... i686-pc-linux-gnu checking for a BSD-compatible install... /usr/bin/ginstall -c checking whether build environment is sane... yes checking for gawk... gawk checking whether make sets $(MAKE)... yes configure: error: *** --with-group=GROUP is missing ...the user 'vilbu' is a member of three groups: 'disk', 'users', and 'vilbu', while it's primary group is 'vilbu'; and the group most certainly exists. Anyone have any thoughts as to why configure fails in this case? Thanks, Stefan
RE: amanda.hosts
Yep. It's working exactly as it should. You probably don't want it sending a backup to tape.hijack-your-data.net. The .amandahosts should be more like this: tape.control.att.com amanda cluster-adm.control.att.com amanda The exact machine name in the error message is what has to appear in the .amandahosts file. -Original Message- From: Bruntel, Mitchell L, SOLCM [mailto:[EMAIL PROTECTED] Sent: Monday, July 14, 2003 5:27 PM To: [EMAIL PROTECTED] Subject: amanda.hosts Ok: amandad runs, and In the /tmp/amanda/ debug file, I get: host failure: ... Amanda 2.4 ACK HANDLE 000-0003EE70 SEQ 1058213638 amandad: time 0.002: bsd security: remote host tape.control.att.com user amanda local user amanda amandad: time 0.016: check failed: [access as amanda not allowed from [EMAIL PROTECTED] amandahostsauth failed amandad: time 0.016: sending REP packet: Amanda 2.4 REP HANDLE 000-0003EE70 SEQ 1058213638 ERROR [access as amanda not allowed from [EMAIL PROTECTED] amandahostsauth failed BUT my $HOME/.amandahosts file shows: tape amanda cluster-adm amanda And I even tried copying my .amandahosts file to the same directory where amanda.conf lives. Still ng. (here is home directory) drwxr-xr-x 2 amanda operator 512 Jul 14 12:56 . drwxr-xr-x 16 root root 512 Jul 10 15:00 .. -rw-r--r-- 1 amanda operator 30 Jul 14 17:22 .amandahosts -rw-r--r-- 1 amanda operator 173 Jul 14 09:44 .profile -rw--- 1 amanda operator5190 Jul 14 17:22 .sh_history -rw-r--r-- 1 amanda operator 30 Jul 14 17:07 amanda.hosts i even thought i might have fat fingered file name, so I double checked spelling.
RE: manually amdump'ing one client only?
Assuming you backup every night at midnight, is an extra backup at noon really that important? If you just let AMANDA do her thing, that machine will get a normal backup tonight, and if it was supposed to get a level 0 last night, it will get it tonight. The previous backups are stlil (presumably) good. -Original Message- From: Martin, Jeremy [mailto:[EMAIL PROTECTED] Sent: Wednesday, July 09, 2003 10:21 AM To: [EMAIL PROTECTED] Subject: manually amdump'ing one client only? Sometimes I need to just back up one client, i.e. if amdump failed the previous night on just one client only. Is the easiest way to do this just to copy the disklist somewhere safe, then edit the disklist and remove all entries except for the one server I need to backup... or is there a better way? I'm backing up to the hard drive, so I'm not worried about wasting a tape on just one client if that is necessary. Thanks! Jeremy
RE: How to change the Mail From on Amanda.
Try setting the hostname on the backup server to something the mail server will accept. You will have to change disklist and .amandahosts (if you use it) to match. -Original Message- From: Elcio Mello [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 08, 2003 1:31 PM To: [EMAIL PROTECTED] Subject: How to change the Mail From on Amanda. Hi, I have had some problems with the mail server destination, it dont acept mail from a no qualified domain (Amanda use the local domain with its from). I would like to know how can I chage the completely mail from( user and domain) in amanda. -- Data -- SO: Conectiva Linux 8.2 Amanda: version 2.41 Thanks and regards, Elcio Mello.
RE: [bugzilla-daemon@bugs.gentoo.org: [Bug 19403] amanda-2.4.3.ebuild (New)]
And if you follow the link and read the bug text, it's 2.4.4, too! I know what my Gentoo box is getting this weekend! -Original Message- From: Anthony A. D. Talltree [mailto:[EMAIL PROTECTED] Sent: Thursday, June 26, 2003 5:49 PM To: [EMAIL PROTECTED] Subject: Re: [EMAIL PROTECTED]: [Bug 19403] amanda-2.4.3.ebuild (New)] An AMANDA ebuild for Gentoo will be available shortly in portage. Yay! You're one zassy frood who really knows where his towel's at.
RE: Need A sample Tapelist file
tapelist starts as an empty file. `touch tapelist` will get you ready for
amlabel.
-Original Message-From: Harry Mbang
[mailto:[EMAIL PROTECTED]Sent: Friday, June 13, 2003 11:25
AMTo: [EMAIL PROTECTED]Subject: Need A sample
Tapelist file
Hi,
Could someone
please send me an example of a tapelist file or point me to where I can find
such. I have installed Amanda several times on a Pentium 4 (HP VL 420)
running Suse 8.1 Professional, but each time the tapelist (what I understand
to be one of the three main configuration files along with amanda.conf and
disklist) is not created. I am using a PowerTape (PT-25) with the
following specs:
{
length 1206
mbytes
filemark
0 kbytes
speed 179
kps
}
and a DT-2400
Cartridge.
I believe if I know the format of
a tapelist I will be able to proceed with using amlabel which complains of a
missing tapelist file, so please give me a sample tapelist file.
However, if anyone knows why tapelist is not created please clue me in.
Cheers.
Harry.
RE: Questions about Amanda - part II
Hi, My questions is: 2. I need to get the backups of all my machines. Do I need to create a /etc/amanda/Directory/amanda.conf for each of them ? Usually, a single machine (the amanda server) has the tape drive, and all other machines (amanda clients) write their backups to the server. There is a single amanda.conf, on the server. Ok, but if I put a lot of clients on 'disklists', Will amanda use one tape for each client ? It will runs all backups on the same time, ok ? AMANDA will use up to the number of tapes specified for runtapes in amanda.conf. AMANDA will get dumps from as many clients as possible at the same time. Limiting factors usually include the inparallel setting, the net usage setting, and the amount of holding space you have made available. 3. If I put 'dump cycle' to 5, the partition that will be backuped wil be divided by 5 and on each day 1/5 of the partition will be backuped, ok ? No. The partition will get a level 0 (full) dump once every 5 days, and incremental dumps the other days. By the way, this is almost certainly not what you want. You probably want 'dumpcycle 1 week' and 'runspercycle 5'. Sorry, but it sounds so confuse for me. I read the chapter about Amanda on Essencial System Administration by Oreilly and the author talk something about: Amanda divide the partition that will be backup and each day one part of the partition will be backup. What I need is do a full backup in one day and incrementals on the next days. So, I put 'dump cycle 7' and 'runs per cycle 5' for 1 day of full backup and 4 days for incrementals, Is it correct ? Each disk list entry will follow its own cycle of full and incremental backups, within the bounds set by dumpcycle and runspercycle. They normally won't all do full backups on the same day, because that would use so much more tape than just doing the incrementals. Setting dumpcycle 5 and runspercycle 5 would tell AMANDA to make sure every disk list entry gets a full backup at least every five runs, but doesn't require that those full backups be on specific days of the week, or all at the same time. I have a problem, last night Amanda send me a email about a backup but, the emails didn't have nothink. Something wrong ? Possibly. Check the logs in /tmp/amanda for that backup run and see if they ended normally. Also send yourself a simple text message as the backup user and see if it gets through ok.
RE: Multiple backup groups
Offsite backups are easy: $ amadmin YourConfig SomeTape no-reuse This will mark the tape as being out of the rotation until you $ amadmin YourConfig SomeTape reuse You can see which tapes are marked by examining the tapelist file. As for GFS, why would you want it? AMANDA will pick the best backup levels for the rotation you've specified, based on the data that needs to be backed up. If you're paranoid and want to take tapes off site, you can set up the rotation to allow it. (The changer we used to use had a ten tape magazine, and we had three magazines. We configured AMANDA to make sure there was a complete backup every five tapes (every week), across thirty tapes. I always had one magazine in the changer, one in the safe, and one at the bank. Every magazine had two complete backup sets, and for critical data, every tape had a complete backup. No problem.) For your Oracle database, backing up a complete database dump every night will provide the most redundancy and fastest restore, but takes a lot of tape. If you don't have room for that, you could get closer by making the main dump one disk list entry, and the incremental logs another. AMANDA would probably bump the main database dump down to an incremental level where the unchanged dump file wouldn't get backed up, to make room for the log files that have changed. -Original Message- From: Derek Suzuki [mailto:[EMAIL PROTECTED] Sent: Monday, June 09, 2003 4:23 PM To: [EMAIL PROTECTED] Subject: RE: Multiple backup groups [snip] one rotation. And it would be nice to have multiple media sets per configuration so that offsite and GFS-style backups could be handled with minimal effort.
RE: Questions about Amanda
1. Always that I create a new configuration (ex: /etc/amanda/Daily5/amanda.conf), I need to create the directories on /var/amanda/Daily5/index/... and others. Could I do this automatic ? $ cp -R /etc/amanda/Daily5 /etc/amanda/NewConfig $ cp -R /var/amanda/Daily5 /var/amanda/NewConfig Then mark all the tapes no-reuse (or remove them) from the new config and alter the files to suit. I don't know of a more automatic way to manage confiurations. 2. I need to get the backups of all my machines. Do I need to create a /etc/amanda/Directory/amanda.conf for each of them ? No, you can just add them to the disklist file that amanda.conf points to. The example included in the distribution shows several machines being backed up by one server. 3. If I put 'dump cycle' to 5, the partition that will be backuped wil be divided by 5 and on each day 1/5 of the partition will be backuped, ok ? What will happen if the machine that is backuped, down on the 4 day ? Could I restore the 1,2,3 days before it down ? That's not how backups are divided. If dumpcycle is 5, and runspercycle is 5, then every disk will have a full backup at least every 5 runs. On the days when a disk doesn't get a full backup. You will be able to restore a machine up to the last time it was backed up, and if a machine was down on a day it was supposed to get a backup, the next incremental (or full) backup will still catch everything that changed, since it's based on the last time a backup was run on that machine. 4. It it possible to add a commando to eject the tape after the backyp is done ?? What ? A pair of ampersands between commands (under the bash shell) will run the second command if the first one succeeds. So this: $ amdump YourConfig eject /dev/nst0 Will eject the tape in /dev/nst0 (a common tape device under Linux) if amdump YourConfig ran ok.
RE: Multiple backup groups
The problem with the offsite is that I want it to include everything from both the daily and the weekly database backups. So the Right, so if you put both disklist entries in the same AMANDA configuration, and rotate a week's worth of tapes off-site, you're covered. You do have at least two week's worth of tapes, I hope. disklist is a composite of the other two jobs. If we tried to include the Oracle data files under the regular daily rotation, they would get backed up every day. Not necessarily. When the backup starts, the planner gets estimates on different levels of backup for each disk list entry, then picks which level is appropriate for each entry based on a whole raft of factors like which ones haven't had full backups the longest, how much tape space is available, and what the configuration files specify for priority. If one of those disk list entries is the main database dump, a level 0 (full) backup might be 4Gb, but if that was backed up yesterday, then today's level 1 (incremental) backup for that DLE might only be 10K. OTOH, the DLE with the log files would have changed much more since yesterday, and even a level 1 of that directory would catch the files that have changed since the last tape. So it would be okay.
RE: Multiple backup groups
Incremental backups are at the file level, like Gnutar (exactly like Gnutar, if that's the program you're using for backups.) AFAIK, Oracle, PostgreSQL, Informix, MSSQL, et al cannot restore from a straight backup of the live database files, and this is never the recommended method of backup. You should use the utilities provided by the database vendor to dump a copy of the database to a static file on disk, then backup that file. I was assuming you were already doing that, and that this was a database that was only dumped weekly, with incremental logs appearing daily. My understanding of Oracle (search the amanda-users archives for details) is that if you're not dumping the database to a file for your backups, you essentially don't have backups. (The exception is backup programs that use the Oracle API rather than the file system.) The no-reuse option can be used for permanent offsites, but you should add new tapes to the rotation to keep the number available at the right level. (Just curious: Does offsite permanently mean that they're in the same bin with the 9-track tapes? Magnetic media doesn't last permanently anywhere, especially on the big magnet we call planet Earth. Might want to plan other storage systems if you really need permanent. -Original Message- From: Derek Suzuki [mailto:[EMAIL PROTECTED] Sent: Monday, June 09, 2003 5:59 PM To: [EMAIL PROTECTED] Subject: RE: Multiple backup groups Hm. Are you saying that Amanda can do a block-level incremental backup? I had assumed that the main database files would have to be backed up on every run, since they get written to thousands of times a day. If it is feasible to incorporate everything into a daily rotation, then I'd still have to force the archival backups (which are meant to be taken offsite permanently) to include a level 0 dump of every target. -Original Message- From: Bort, Paul [mailto:[EMAIL PROTECTED] Sent: Monday, June 09, 2003 2:36 PM To: [EMAIL PROTECTED] Subject: RE: Multiple backup groups The problem with the offsite is that I want it to include everything from both the daily and the weekly database backups. So the Right, so if you put both disklist entries in the same AMANDA configuration, and rotate a week's worth of tapes off-site, you're covered. You do have at least two week's worth of tapes, I hope. disklist is a composite of the other two jobs. If we tried to include the Oracle data files under the regular daily rotation, they would get backed up every day. Not necessarily. When the backup starts, the planner gets estimates on different levels of backup for each disk list entry, then picks which level is appropriate for each entry based on a whole raft of factors like which ones haven't had full backups the longest, how much tape space is available, and what the configuration files specify for priority. If one of those disk list entries is the main database dump, a level 0 (full) backup might be 4Gb, but if that was backed up yesterday, then today's level 1 (incremental) backup for that DLE might only be 10K. OTOH, the DLE with the log files would have changed much more since yesterday, and even a level 1 of that directory would catch the files that have changed since the last tape. So it would be okay.
RE: Multiple backup groups (explained)
[snip] Now, is there a way to do something like this with Amanda, a way to define set groups like this via different configs and share the same pool of tapes? Is there a way where I can indeed tell Amanda '30G of people to backup, do only 10G a day, but make sure every one is backed up at least twice in one week'? I really don't think you need multiple configs, since you want to share tapes. I'd do it in one config. Tell the users that their machines will be backed up, at least in part, every night. An advantage of doing this is that if a user misses a night when they should have had a full backup, they'll get one the next night _automatically_. Full backups will be spread over the three days to keep the run time and tape usage consistent. Letting AMANDA do her job really is better. amanda.conf: # Change this to the number of tapes you want to cycle through: tapecycle 6 # Full backups at least every three runs: dumpcycle 3 runspercycle 3 # One tape per run: runtapes 1 If you only want to use 10Gb of a 12Gb tape, change the tapetype to reflect that. Or leave it at the tested capacity. Add a machine or two a night. Check the run times to make sure they're not colliding with normal use time. If the tape is slow, using a big holding disk and boosting the inparallel setting can let the dumps finish quickly, then take their time going to tape.
RE: Backing up MySQL tables
Dumping the database and then backing up the dump is the right way to go. Think of the dump as a snapshot of the database. By backing up the snapshot, you know you're getting a consistent and restorable backup. If you're worried about the disk space it takes, you can compress it. I've seen this method used for MySQL, PostgreSQL, MS SQL, and Oracle. -Original Message- From: Alex Thurlow [mailto:[EMAIL PROTECTED] Sent: Wednesday, April 02, 2003 1:28 PM To: [EMAIL PROTECTED] Subject: Backing up MySQL tables I need to start backing up MySQL tables on one of my machines, but I've heard that to you run the risk of table corruption if you just copy the mysql directory while MySQL is running. Stopping the database is not an option, and the only solution I've come up with is having another script do a mysql dump and then backing up that. It seems like a waste of time and resources to basically do 2 backups though. I was just wondering what anyone here does for MySQL backups on their own systems? Alex Thurlow
RE: Frontend , UI for amanda ?
Thomas, I'd like to disagree with you on a point-by-point basis: Friendly user interface (not necessary fancy GUI) is a measure of all good quality software products. Amanda is not beyond the scope of this view. So the Linux Kernel, MVS/OS, Sendmail, and PostgreSQL are not of good quality? I disagree. AMANDA is written to be as invisible to the user as any of the above. Amanda is client-server based product. This means, while its client might be in bare OS, amanda server is not. The server has to be fully functioning to provide the data recovering/restoring service. Not true. I can restore any of my files given only hardware that can read the tape and a machine either running Red Hat 6.2 or capable of supporting it. AMANDA is not needed for the restore, as she is only a backup manager. The backup utility ( tar/dump/xfsdump, gzip, etc.) with mt and dd is all that is needed to read the tapes. The possible loss of this functionality was of great concern when adding the ability to split a disk across tapes was being discussed on the AMANDA-Hackers mailing list. In terms of management of amanda, I appreciate all efforts that have been put into the product to make it less demanding for human intervention however, it is still a client/server architecture involving many resources and objects. Initial configuration, though one time, takes some time. The difficulty of getting used to it has I think that a GUI for initial configuration will make it easier for new users to get into trouble faster. The INSTALL file included with the package is a very good start, with step-by-step instructions. Since AMANDA should be compiled from scratch at least for each server, the new user isn't going to avoid the command line during install. been reflected in this mailing list. For a changing (amount of data, schedule, tape devices) environment, managing amanda does not seem to be a piece of cake. As far as I am concerned, Bernd's presented a good Changes in the amount of data being backed up will usually be handled correctly by AMANDA. Changing tape devices is also easy once you have the first one working, and given the cost, unlikely to be a common occurrence. And why would the schedule need to change frequently? Other than restore tests and tape changes, there should be a time of day when your users can tolerate the performance overhead of a backup running, in exchange for having backups. If you need to run extra backups spontaneously, I don't see how a GUI can be better than typing `amdump YourConfig ` at the shell prompt. question if we look at it as a bigger picture rather than as a stupid question from a lazy and less knowledgeable admin. It's certainly not a stupid question. There are good places and good reasons for a GUI. I'm probably going to write one in the next six months or so, as we deploy AMANDA to client sites. As always, it's open source. If the lack of a GUI bothers you enough, you can write one, or hire someone to write it for you. The software scratches the collective itch. That said, here's my case for a GUI: At remote sites where the user is unlikely to even log in to the console, they need some way to interact with AMANDA. They at least need to know what tape to put in next; need to be able to add new tapes; and mark existing tapes as NO-REUSE. (In my case, I can count on them calling support for a restore.) Once I have that in place and can release it, other people might want to add things like a restore interface, or updating the disklist.
RE: Cygwin client - advantages?
It has several advantages that make the installation overhead worthwhile: 1. Client-side compression 2. Better chance of backing up ACLs (using special tar) 3. Works on machines that have all shares disabled (I have seen disabling shared recommended for some servers, especially SQL.) 4. More straightforward restore 5. Doesn't require SAMBA on AMANDA Server Also, I haven't tested it, but I suspect sending a tar file using the AMANDA protocol is faster than carrying on an extended SMB conversation. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Sunday, March 23, 2003 5:35 AM To: [EMAIL PROTECTED] Subject: Cygwin client - advantages? Can anyone tell me... What, if any, are the advantages of running an amanda client (under Cygwin - Windows XP) over using samba? I understand amanda-2.4.4 will install under the Cygwin environment, but I would like to know what makes it worth going to the trouble. Thanks, Dick
RE: Pre-backup script ?
I just have cron on the amanda server call a script rather than amdump directly. That script and passwordless ssh do all the twiddles to the clients I need done, runs amdump, and then untwiddles everything. The disadvantage is that you shut down stuff for the entirety of the amdump run, not just while that client is being dumped. But it's a heck of a lot easier... You could mitigate this by splitting your backups into multiple configurations (easy if you have multiple drives or a changer), and making the mail server a separate backup. (If you're in the Undisclosed Summons of America, you might not want to keep your e-mail backups as long as you keep your other backups anyway.)
RE: AMANDA and the Windows world
So, I tried this, and you are correct -- the ACLs aren't saved by tar (I guess I was hoping that the cygwin tar would be able to work with this... perhaps in future versions? However, I also did do my test restore. It worked, although the permissions were, of course, incorrect. So, if you're like me, and are really only interested in backing up user data -- not programs, or system files, it'll probably work correctly. I'm still looking around to see if there are cygwin based (or, at least, command line based that can write to STDOUT, as I imagine that's all that's needed) backup programs with do handle acls... As a workaround, you could dump the ACLs with CACLS.EXE to a text file. This would have to be run on each directory you care about, and you'd have to write something to parse the output afterwards and apply it, but it's a start, and doesn't change Cygwin or AMANDA. The output looks something like this: U:\docsCACLS vssver.scc U:\docs\vssver.scc DOMAIN\Domain Admins:F DOMAIN\user:F To make that back into a CACLS command, it would have to look like this: U:\docsCACLS vssver.scc /P DOMAIN\Domain Admins:F DOMAIN\user:F The Perl transformation of one into the other is left as an exercise for the student. But this will do the job.
RE: 2.4.4b1 cygwin client and specifying DLEs with spaces...
If your system is also generating old 8.3 names (for backwards compatibility) you can mix them in: /cygdrive/c/docume~1/myfolder The command dir /x (under XP Pro, yours may vary) shows the short names for files that need them. -Original Message- From: Richard Morse [mailto:[EMAIL PROTECTED]] Sent: Monday, February 10, 2003 2:43 PM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: 2.4.4b1 cygwin client and specifying DLEs with spaces... Hi! This is probably a really easily answered question, but I'm not having any luck finding it (probably going blind)... I've got the 2.4.4b1 client set up on a cygwin machine. The backup is going well, but in order to create a DLE, I had to do some aliasing. Is there a way to specify a DLE like: X.. /cygdrive/c/Documents and Settings/myuserfolder comp-tar That is, include paths with spaces? Alternatively, in the .amanda-gtar-excludes file, can I specify directories with spaces? So I can exclude /Program Files? Thanks, Ricky
RE: Running AMANDA over the Internet
Amanda is a backup manager, not a security manager. There are no steps taken to ensure the security of the backups. Several solutions are available, though: - Use the Kerberos support built in to Amanda. I've never played with this. - Use tar with a wrapper script on the client that encrypts the backup before sending it. You might be able to find samples of this in the list archives. - Use an encrypted VPN (CIPE, FreeSWAN, SSHTunnel) between servers. This is the method I used, because I use the same tunnel for monitoring and file transfers. Search the list, think about what method fits your needs. Feel free to ask more questions. Good Luck. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 05, 2003 9:50 AM To: [EMAIL PROTECTED] Subject: Running AMANDA over the Internet Hello, Let's take the scenario where I have got an AMANDA server located at one central site and have a few other servers located at various places around the globe which of course all need to be backed up by the centraon site's AMANDA server. My question is now more related about security and how secure it is to run backups over the internet. Is AMANA secure by default to run over the internet or are there any optional compiling options or features which I should use to make the clients itself and the dump secure ? Many thanks for your opinion Regards
RE: Running AMANDA over the Internet
Actually, an SSH tunnel is one of the least easy VPNs for this because of the many different ports AMANDA can use. Unless you have limitations on what you can install on the boxes, a full VPN (like IPSec, as mentioned in another post) is probably your best bet. The VPN model of point-to-point connections suits well because AMANDA's traffic is also shaped like that, where one of the points is always the server, and the other is a client. If you're really more comfortable with SSH, you could schedule a tar on each client, and follow it with an scp to send the data to the backup server, where it can be written to tape. This would add an extra step to any restore, but doesn't require a VPN. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 05, 2003 11:34 AM To: Bort, Paul Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: Running AMANDA over the Internet I think the easiest way would be to use an SSH tunnel. Would this be easy to implement ? Any examples maybe or pointers on how to acheive that ? Thanks Regards Bort, Paul To: '[EMAIL PROTECTED]' [EMAIL PROTECTED], [EMAIL PROTECTED] [EMAIL PROTECTED] cc: om Subject: RE: Running AMANDA over the Internet Sent by: owner-amanda-users@ amanda.org 02/05/03 04:20 PM Amanda is a backup manager, not a security manager. There are no steps taken to ensure the security of the backups. Several solutions are available, though: - Use the Kerberos support built in to Amanda. I've never played with this. - Use tar with a wrapper script on the client that encrypts the backup before sending it. You might be able to find samples of this in the list archives. - Use an encrypted VPN (CIPE, FreeSWAN, SSHTunnel) between servers. This is the method I used, because I use the same tunnel for monitoring and file transfers. Search the list, think about what method fits your needs. Feel free to ask more questions. Good Luck. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 05, 2003 9:50 AM To: [EMAIL PROTECTED] Subject: Running AMANDA over the Internet Hello, Let's take the scenario where I have got an AMANDA server located at one central site and have a few other servers located at various places around the globe which of course all need to be backed up by the centraon site's AMANDA server. My question is now more related about security and how secure it is to run backups over the internet. Is AMANA secure by default to run over the internet or are there any optional compiling options or features which I should use to make the clients itself and the dump secure ? Many thanks for your opinion Regards
RE: Replacing a partially bad tape
Just wait until amcheck is calling for that tape number, then label it with the -f (force) option, and use it. -Original Message- From: Anthony Valentine [mailto:[EMAIL PROTECTED]] Sent: Tuesday, January 21, 2003 3:01 PM To: Amanda Users Subject: Replacing a partially bad tape Hello! I have a tape that has gone bad, however the damage is very near the end of the tape, so that I am still able to get data off of the beginning. I would like to replace this tape with a new one, however I don't want Amanda to forget about the old one yet. Aside from waiting until the day it's going to be overwritten (which isn't for 60 days, and I don't want to forget), can I label the new tape but still have Amanda keep the index for the old one, just in case I need it? Thanks in advance! -- Anthony Valentine [EMAIL PROTECTED]
RE: Should I backup /dev/sda7 which is - / on my system
Try being more explicit in the disk list, either '/dev/sda7' or '/' instead of 'sda7'. -Original Message- From: Kevin Passey [mailto:kpassey;kdpsoftware.co.uk] Sent: Tuesday, November 12, 2002 9:33 AM To: Amanda (E-mail) Subject: Should I backup /dev/sda7 which is - / on my system Hi, I have been backing this disk up - but since I've updated by RH7.2 install from RHN I get the following message. ERROR: dilmom: [could not access sda7 (sda7): No such file or directory] Should I be backing up this disk - I would have thought so - but now amanda will not recognise it. Can anybody point me in the right direction? Thanks Kevin
RE: How can I skip tapes from backups that were missed due to system down?
The best way to handle this would be to relabel all of your tapes and ditch the hard connection between tapes and days, which is just going to keep getting in your way. Just number them sequentially. That said, if you're really stuck on your current arrangement, you can use the 'no-reuse' option of amadmin to take tapes out of rotation. Just mark your calendar for when those tapes come around again, and remember to mark them 'reuse' when you're about to use each marked tape. You can also search the list archives for more 'no-reuse' info. -Original Message- From: Kevin Passey [mailto:kpassey;kdpsoftware.co.uk] Sent: Tuesday, November 12, 2002 9:37 AM To: Amanda (E-mail) Subject: How can I skip tapes from backups that were missed due to system down? Hi again, My tapes are labelled mon,tue,wed etc. I want to use tue tape but amanda is expecting fri - how can I skip fri and mon tapes. Thanks Kevin
RE: ERROR: /devb/nst1: no tape online
There's nothing in your `mt status` that says what tape drive it's talking to. My guess is that your SCSI devices got renumbered somehow. I think the nst* devices are issued in the order they're found, and if the device that was nst0 is gone, nst1 is now nst0. Try `mt -f /dev/nst0 status` and `mt -f /dev/nst1 status` and see which one looks like your tape drive. -Original Message- From: Paul Miller [mailto:paul;fxtech.com] Sent: Tuesday, November 12, 2002 11:03 AM To: [EMAIL PROTECTED] Subject: ERROR: /devb/nst1: no tape online Can anyone tell me what this means? I recently moved a linux system from a tower case to a rack case. Everything was wired up the way it was, and everything is working, except amanda. For some reason, it doesn't think there are tapes in the drive (a DLT4000). I'm getting this error: bash-2.04$ /usr/sbin/amcheck Daily Amanda Tape Server Host Check - Holding disk /amanda: 1714756 KB disk space available, using 666180 KB ERROR: /dev/nst1: no tape online (expecting tape Daily05 or a new tape) NOTE: skipping tape-writable test Server check took 0.002 seconds mt seems to work: bash-2.04$ mt status SCSI 2 tape drive: File number=0, block number=0, partition=0. Tape block size 0 bytes. Density code 0x1a (unknown to this mt). Soft error count since last status=0 General status bits on (4101): BOT ONLINE IM_REP_EN Could this be a messed up compression setting? I could really use some advice!
RE: Performance tuning for Linux
The only thing I've done here to improve backup performance (and performance in general) is using hdparm to maximize IDE throughput. (Yes, I know it's evil to run servers on IDE, but it's CHEAP.) -Original Message- From: Orion Poplawski [mailto:orion;colorado-research.com] Sent: Friday, November 01, 2002 12:54 PM To: [EMAIL PROTECTED] Subject: Performance tuning for Linux Just wondering if there is anything that should be done to improve backup performance under linux. I was thinking along the lines of tape block sizes (variabl, fixed), st driver buffer sizes, etc. The tape dump program does not appear to be configurable for block size at run time, what does it use? Thanks! - Orion Poplawski
RE: Thoughts on Win32
Don't know. I use it for DB dumps, so I want a full backup every time anyway. -Original Message- From: [EMAIL PROTECTED] [mailto:ahall;secureworks.net] Sent: Thursday, October 24, 2002 2:22 PM To: Galen Johnson Cc: [EMAIL PROTECTED] Subject: Re: Thoughts on Win32 On Thu, 24 Oct 2002, Galen Johnson wrote: [EMAIL PROTECTED] wrote: Greetings, I am about to attempt to archive my Win32 hosts with amanda. I was wondering what the list users experience has been with samba and/or the Win32 amanda client. Thank you, Andrew I've had great luck with Samba. I have avoided the win32 client mainly because it doesn't appear to be under development any longer. Others have reported success with cygwin and compiling amanda on each win client. =G= In reading the list archives I found a message that stated that due the the fact that the ctime is not perserved it causes amanda to run a full backup every run. Is this still the case? Andrew
RE: Thoughts on Win32
Andrew, Here are the methods currently available, that I know of, and my opinions on them: 1. SAMBA via smbclient: easiest to set up, but direct restores are tricky; better to restore to the *nix box running smbclient and then move the files back yourself. Works best when you just backup critical data directories. This is what I'm currently using for my Windows boxen. 2. SAMBA via smbmount: looks like a regular tar to AMANDA, but you have to keep the mount connected, and I'm not sure it will handle spaces in filenames as well. I never got this working the way I wanted, but that was mostly my lack of SAMBA knowledge. 3. NFS: There is an NFS server implementation available from Microsoft as part of Unix Services for Windows. In theory, you could NFS mount and backup with tar. It appears that most people use SAMBA instead of sending more money to Redmond. 4. Amanda-Win32 client: I looked at this and stopped looking when I got to the part where it doesn't work with amcheck. Maybe they've fixed that since spring '02, I haven't had time to keep up with it, but since my backups run overnight, I need amcheck to let me know all is well before I go home. 5. Amanda Client under Cygwin: A very recent development on the amanda-hackers list (Many thanks to Doug and Joshua) is a patch for compiling the real AMANDA source under Cygwin. I've gotten it working with some kludginess on one server, and all of the problems I'm having seem to be coming from my lack of understanding of Cygwin. This is under active development and looks very promising. If you're ok with installing Cygwin on your Windows machines, I'd give this a try. Any other thoughts? -Original Message- From: [EMAIL PROTECTED] [mailto:ahall;secureworks.net] Sent: Thursday, October 24, 2002 1:32 PM To: [EMAIL PROTECTED] Subject: Thoughts on Win32 Greetings, I am about to attempt to archive my Win32 hosts with amanda. I was wondering what the list users experience has been with samba and/or the Win32 amanda client. Thank you, Andrew
RE: Host name aliases in disklist
No way would I call this a bug. Because of the business AMANDA is in, answering multiple concurrent requests could be dangerous. There's no reason that your backup software shouldn't know every machine by one, and only one name, no matter how many names your users know it by. There are places in IT to obscure what's going on, this ain't one of them. If you're worried about your indexes being broken when you move a disk, how often do you really move a disk? Compare that to how often you restore, or better, how often you test your restores. You might be chasing a ghost, Scooby. It might make sense to also be prepared to do non-index recoveries. (Be prepared, test both monthly!) If you really want to do this, run multiple amdumps per day, and alter the disklist in between. Make sure one is done before the next starts. IIRC, The timeout is because inetd doesn't start a second amandad for the second connection request. So it simply goes unanswered. This might be a good thing to put some code into: On amandad startup, if it detects another amandad running, it returns some sort of HOST BUSY message. Probably easier to do in 2.5 with the NOP feature. -Original Message- From: Toralf [mailto:toralf;kscanners.com] Sent: Thursday, October 17, 2002 1:40 PM To: Anthony Valentine Cc: Amanda Mailing List Subject: Re: Host name aliases in disklist On Thu, 2002-10-17 at 03:03, Toralf Lund wrote: Hi, your assumption about the hostnames is correct. What you are doing is one of the big NO!NO!'s of amanda. Just like I suspected ;-/ Never list a single host in multiple concurent amanda runs, or, in your case, with different names in a single amanda configuration. amandad can only handle one request at the same time. Yes, I knew that. Which is why amdump/planner should do a host lookup and base it's decision on IP addresses. Only solution is to use only one hostname per physical host in disklist. I'd say the failure to handle aliases is a *BUG*! I don't think he is saying that it's the aliases that are a problem, but rather having the same host in the disklist twice as two separate aliases. The problem seems to be that Amanda doesn't understand the concept of aliases, i.e. that different hostnames may refer to the same physicl host. So, if that is true, you could have an entry for either amanda-server or fileserv, but not both only because they are the SAME host. Amanda can only handle one connection at a time, which would explain why amanda-server will time-out while fileserv is backed up fine, and then vise versa. It's still a bit surprising to me that I get a timeout. I would normally expect a host busy message. No other way out there. Again, this is really a shame. It causes extra work a whole lot of mess in the indexes etc. if you move disks around a bit. - Toralf -- UNIX was not designed to stop you from doing stupid things, because that would also stop you from doing clever things.
RE: restore hangs
Be careful that your file system can handle 9Gb files, some can't. And you were asking for it when you named it 'eris'. All hail discordia. -Original Message- From: Jerry [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 16, 2002 9:26 PM To: amanda mailing-list Subject: restore hangs anyone ever seen this? I am doing a restore by: /opt/amanda/sbin/amrestore -p /dev/rmt/0n eris | ufsrestore -ivbf 2 - It's a 9 GB dump approx. I get the interactive restore, I can restore files, but when I exit it hangs... I even let it sit for hours on the way home from work and after dinner. It never exits. When I control-c it, the process dies a few seconds after and I rewind the tape and everything is cool. Anyone with anything remotely similar. When I try to list the contents of the dump with restore -t same thing... hangs towards the end (I think it lists everything in the backup). The next thing I am going to try is to tape the file off tape (whoosh 9gb) and try to read it that way to see if it hangs as well. Jerry __ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos More http://faith.yahoo.com
RE: Backing up Windows shares...
Brian, Doug Kingston recently posted a patch (on amanda-hackers) to 2.4.3b4 to make it run under Cygwin. I'm trying to get it to work here on my test server, and once I learned about 'set CYGWIN=nontsec', it configured, but I'm having a library problem during make. If it works, I think it will be a very popular solution, as Doug's changes should be able to be merged into the main source and enabled with a configure option or detect. Paul -Original Message- From: Brian Kennedy [mailto:[EMAIL PROTECTED]] Sent: Tuesday, October 15, 2002 11:15 AM To: [EMAIL PROTECTED] Subject: Re: Backing up Windows shares... Is Jim Buttafuco still out there? He said he had rewritten the amanda client in perl to run on windows machines and wanted to release it a few weeks ago. I got real excited and then he dissapeared. If that's fallen through, does anyone have suggestions on backing up windows machines that has nothing to do with samba? Having engineers as users is troublesome, they know enough to THINK they know what they're doing. Commonly they change permisions so my backup user cannot access the files to back them up, so I need something that can run as a client process on the machine as local Administrator to bypass such nonsense. Steve Bertrand wrote: I have always had problems backing up windows shares with amanda, so I just continued to use Veritas for them, until I found an easy way to do it. If you mount the windows share into the Unix file system, it seems to be far more reliable, as well as faster and easier to do: # mount -t smbfs //winserver/share /mnt/mntpoint then add an entry for the mntpoint in amanda's disklist entry. Hope this makes someones life easier! Steve Bertrand
RE: Backing up Windows shares...
I looked at that and gave up when it looked like it couldn't do estimates. It also hadn't been updated in a while, and I'm looking forward to amanda 2.5.0. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Tuesday, October 15, 2002 12:17 PM To: [EMAIL PROTECTED] Subject: Re: Backing up Windows shares... There's amanda-win32 at sourceforge. I'm just in testing phase right now, takes a little work to put the pieces together, and I still need to figure out how to automate an install as much as possible. But has a small footprint and seems efficient. I apologize if this message does not come through clean. Forced to use Notes now . . . . -- toby bluhm philips medical systems, mr development, cleveland, ohio [EMAIL PROTECTED] 440-483-5323 Brian Kennedy [EMAIL PROTECTED] Is Jim Buttafuco still out there? He said he had rewritten the amanda client in perl to run on windows machines and wanted to release it a few weeks ago. I got real excited and then he dissapeared. If that's fallen through, does anyone have suggestions on backing up windows machines that has nothing to do with samba? Having engineers as users is troublesome, they know enough to THINK they know what they're doing. Commonly they change permisions so my backup user cannot access the files to back them up, so I need something that can run as a client process on the machine as local Administrator to bypass such nonsense. Steve Bertrand wrote: I have always had problems backing up windows shares with amanda, so I just continued to use Veritas for them, until I found an easy way to do it. If you mount the windows share into the Unix file system, it seems to be far more reliable, as well as faster and easier to do: # mount -t smbfs //winserver/share /mnt/mntpoint then add an entry for the mntpoint in amanda's disklist entry. Hope this makes someones life easier! Steve Bertrand
RE: tape changer
I don't use either, but I thought chg-mtx and chg-zd-mtx were different, and that difference was in the output of the two commands, hence what looks like a parsing error below. -Original Message- From: Martin A. Brooks [mailto:[EMAIL PROTECTED]] Sent: Thursday, October 03, 2002 5:41 AM To: [EMAIL PROTECTED] Subject: Re: tape changer At 11:27 03/10/2002 +0200, you wrote: Hm, are you shure /dev/sg0 is your changer? does it have scsi-id 0 or is the lowest id on your scsi-bus? if not count the devices with lower id's and use this number as x in /dev/sgx. Christoph I can successfully load and unload tape using mtx. So, yes, I'm sure sg0 is my changer. I've also switched to the chg-zd-mtx changer script as this seems more appropriate to my system. With tapedev as /dev/nst0 and changerdev as 0, I now get this error, which I'm not sure how to sort backup@amanda:/etc/amanda/Changer$ amtape Changer show amtape: could not get changer info: cannot determine first slot Regards Mart.
RE: tape changer
Here's a few things to try: 1. make the tape changer work separately from any AMANDA components. You might need zd-mtx or similar. (I use the ch driver and utilities from http://www.strusel007.de/linux/changer.html.) I'm guessing you're using some flavor of linux based on the device name /dev/sg0. 2. Make AMANDA work with the tape drive in the changer using the chg-manual changer. This means that before every backup, you will have to use the control commands from step 1 to load the right tape, but your backups will be working. 3. try the AMANDA changer script that matches the changer-driver you've selected. If you're using the one I linked to above, search the list archives for my chg-userland perl script that glues it to AMANDA, or let me know if you can't find it, and I'll re-post it. -Original Message- From: Martin A. Brooks [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 02, 2002 11:13 AM To: [EMAIL PROTECTED] Subject: tape changer Hi there I'm trying to use a Dell Powervault T130 with AMANDA and I'm having some difficulties. Specifically... backup@amanda:~$ amtape Changer reset amtape: could not reset changer: /dev/sg0: failed backup@amanda:~$ amtape Changer taper amtape: scanning for a new tape. amtape: could not get changer info: /dev/sg0: failed I'm using the chg-scsi changer option - could anyone indicate what I've missed? Regards Mart.
RE: disk-to-disk backups
Possible? Yes, with the current version of AMANDA. Recommended? I'd be leery of doing it to any drive I can't unplug, tuck under my arm, and run out of a burning building with. (Yes, I only have external tape drives.) A mix of disk for quick restores and tape for off-site would be entirely reasonable, if you really do that many restores. (Just remember to test restore both at least monthly.) -Original Message- From: Terri Eads [mailto:[EMAIL PROTECTED]] Sent: Thursday, September 26, 2002 3:19 PM To: [EMAIL PROTECTED] Subject: disk-to-disk backups We're considering doing our backups to raid systems now instead of to tape libraries. Does anyone have any positive or negative comments or experiences with the scenario that they are willing to share? (and is this even possible with amanda?) Thanks, -- Terri EadsUNIX Systems Administrator [EMAIL PROTECTED] Research Applications Program (303) 497-8425National Center for Atmospheric Research
RE: help on append to tape feature
Vijay, I really don't think you will have a lot of luck getting this change committed to CVS. There are some very good reasons why this has been discussed and discarded several times already: - Different tape hardware has different methods (and accuracy) in positioning for append. - If the tape fails during an append, you've not only put the current backup in jepoardy, you've destroyed any other backups that were on that tape. - Is your data really worth so little that you won't buy a few more tapes for it? That being said, AMANDA already supports what you're asking for, in a safer manner. You can perform several backups to holding disk, then flush them all to the same tape. This avoids the problem of tape positioning, and all of the backups remain on the holding disk until they are written to tape. The amanda-users and amanda-hackers archives have other people's takes on the matter. Here's Gyles entry from FOM: http://amanda.sourceforge.net/cgi-bin/fom?_highlightWords=append%20tapefile =29 -Original Message- From: Vijay [mailto:[EMAIL PROTECTED]] Sent: Monday, September 23, 2002 12:46 AM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: help on append to tape feature hi group, i'm planning to add append to tape feature in amanda. ie.. In case of smaller dump size. amanda needs an unused tape for dump. i would like amanda to dump these smaller tapes in an available tape by appending. so i would like to get help regarding this. i found that this module needs developers. i would like to contribute. i suppose the taper module of amanda needs changes. any help regarding this will be greatly appreciated. Vijay
RE: Install in RedHat 7.2 fails...... please, some help ....
The word 'and' doesn't mean anything to configure.
-Original Message-
From: Hadad [mailto:[EMAIL PROTECTED]]
Sent: Monday, September 23, 2002 6:14 PM
To: [EMAIL PROTECTED]
Subject: Install in RedHat 7.2 fails.. please, some help
I can't install Amanda in a RedHat 7.2.
Today I get the lastest stable package in amanda.org and my sintaxe
for configure is:
./configure --disable-libtool --without-client
--with-user=amanda and --with-group=amanda
Note: i want use Amanda only for backup files inside the same machine
where Amanda will be installed.
And this backups will be do it in my hard drive.
Well, my output is:
---
[root@localhost amanda-2.4.2p2]# ./configure
--disable-libtool --without-client --with-user=amanda and
--with-group=amanda
creating cache ./config.cache
checking host system type... Invalid configuration `and':
machine `and' not recognized
checking target system type... Invalid configuration `and':
machine `and' not recognized
checking build system type... Invalid configuration `and':
machine `and' not recognized
checking cached system tuple... ok
checking for a BSD compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking whether make sets ${MAKE}... yes
checking for working aclocal... found
checking for working autoconf... found
checking for working automake... found
checking for working autoheader... found
checking for working makeinfo... found
checking for non-rewinding tape device... /dev/null
checking for raw ftape device... /dev/null
checking for Kerberos and Amanda kerberos4 bits... no
checking for gcc... gcc
checking whether the C compiler (gcc ) works... yes
checking whether the C compiler (gcc ) is a cross-compiler... no
checking whether we are using GNU C... yes
checking whether gcc accepts -g... yes
checking for object suffix... o
checking for Cygwin environment... no
checking for mingw32 environment... no
checking for executable suffix... no
*
This machine, target type , is not known
to be fully supported by this configure script. If the
installation of Amanda on this system succeeds or needed
any patches, please email [EMAIL PROTECTED] with
the patches or an indication of the sucess or failure of
the Amanda installation on your system.
*
checking for ar... /usr/bin/ar
checking for mawk... no
checking for gawk... gawk
checking for gawk command line variable assignment... yes with -v
checking for bison... bison -y
checking for cat... /bin/cat
checking for compress... /usr/bin/compress
checking for dd... /bin/dd
checking for egrep... /bin/egrep
checking for getconf... /usr/bin/getconf
checking for gnuplot... no
configure: warning: *** You do not have gnuplot. Amplot will
not be installed.
checking for grep... /bin/grep
checking for gtar... /bin/gtar
checking for smbclient... no
checking for gzip... /bin/gzip
checking for Mail... /usr/bin/Mail
checking for mt... /bin/mt
checking for chio... no
checking for chs... no
checking for mtx... no
checking for lpr... /usr/bin/lpr
checking which flag to use to select a printer... -P
checking for pcat... no
checking for perl5... no
checking for perl... /usr/bin/perl
checking for sh... /bin/sh
checking for ufsdump... no
checking for dump... /sbin/dump
checking for ufsrestore... no
checking for restore... /sbin/restore
checking whether /sbin/dump supports -E or -S for estimates... S
checking for xfsdump... no
checking for xfsrestore... no
checking for vxdump... no
checking for vxrestore... no
checking for vdump... no
checking for vrestore... no
checking for large file compilation CFLAGS... -D_FILE_OFFSET_BITS=64
checking for large file compilation LDFLAGS...
checking for large file compilation LIBS...
checking for ranlib... ranlib
checking for ld used by GCC... /usr/bin/ld
checking if the linker (/usr/bin/ld) is GNU ld... yes
checking for BSD-compatible nm... /usr/bin/nm -B
checking whether ln -s works... yes
updating cache ./config.cache
loading cache ./config.cache within ltconfig
ltconfig: you must specify a host type if you use `--no-verify'
Try `ltconfig --help' for more information.
configure: error: libtool configure failed
[root@localhost amanda-2.4.2p2]#
---
Someone can help me please ???
Thanks !
--
Hadad mailto:[EMAIL PROTECTED]
RE: SAP R/3 backup with amanda
There is a tar-wrapper script for people who don't have GNUTar or need to do something non-tar, like dumping a database. Search the amanda-users list archives for it. It works with 2.4.x. AFAICS, Amanda out of the box does not support pre-/post-backup triggers (stop database before backup and start it when finished), nor does it support adapting a database specific backup tool. The 2.5 line may have some support for this. hauke
RE: special backupcase
I think you will have to set up multiple disklist / configs to try and force Amanda to do this. Hmm this seems abit complicated to me as im still trying to grasp the basic concepts of amanda? If you could point me towards some reading or howto i would be very grateful! If your limiting factor is bandwidth, you might be better off fiddling with the tape size until AMANDA is only tying up the DSL line for the amount of time you have the DSL line to yourself. AMANDA automatically balances the full backups across all of the tapes in a tapecycle so that backups are predictable in terms of time and tape used. You should also look at client-side compression, which could save a lot of time depending on how compressible your data is. Amanda herself determines the level and when that level will be done within the dumpcycle. She tries to balance the amount of data backed up each run so as to maximize tape usage. So what you are telling me is that i can't control wich days to do the full backup and which days to do incremental Yes, and most people don't want to, once they get the hang of it. Spreading the level 0 backups across tapes provides you with some protection in case a tape fails, better use of tape capacity, predictable backup durations, and warnings, rather than failures, when your backups get larger. (If you see a lot of backups being bumped, it's might be because your level 0's are getting bigger.) Each fs backed up MUST be able to fit on a tape. You need to specify multiple tapes / use of a tape changer would be desireable. Ok let me get this straight fs here means filesystem ok? a filesystem is per/machine? or is it per time amanda is running? So when amanda is running and doing a full backup all the filesystems listed in disklist should be able to fit on one tape? Im using the newest beta which will backup to our NAS box with 400GB available so i guess its just a question to fool amanda to think that im running some big DLT that can have +80GB on one tape or so? Is that right? Each partition on each machine has to be less than the size of one tape. Since you're using a file system for a tape drive, you can set the tape size arbitrarily. Tape size/speed is usually the bottleneck in a backup system, but it sounds like the network is the limit in your case. Use the tape size to control how much is backed up each night. You might also need to set the netusage parameter to some fraction of your DSL speed to keep AMANDA from flooding the line and timing out. Here's how I'd calculate it, your numbers will probably be different: SDSL 768 kilobits per second is about 76 kilobytes per second I'd set netusage to 60 KB, or about 80%. (I have confidence in my DSL provider, can you tell?) Assuming you can run your backups for 12 hours per night: 12 * 60 * 60 = 43200 seconds of run time per night 43200 * 60 = 2592000 kilobytes per night, or about 2.5 Gb. Cosmic ray particles crashed through the hard disk platter I prefer alpha particles. They can't get through aluminium foil, but lusers don't know that.
RE: tape rotation and no space left
Out of tape really means out of tape. Appends are really unlikely, as AMANDA
starts the tape write with a rewind followed by a write. No seek, no foul,
no append. (Unless you have the append patch.)
Is it possible you have hardware compression turned on? This can cause
problems if you also use software compression: the dumps tend to take more
space.
DDS3 is supposed to be good for 12Gb. Is it possible you have some DDS2
tapes mixed in? (This can happen with inherited systems)
If it were my tape, I'd put a note on the hard copy of the report, and watch
that tape next time it runs. If it gets the same error, I'd replace the
media. A reformat/retension might work, but those tapes are cheap enough
that it's not worth it.
Watch the next night's run to make sure this might not be a drive that needs
cleaning or adjustment.
-Original Message-
From: Keith Nasman [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 05, 2002 6:41 PM
To: [EMAIL PROTECTED]
Subject: tape rotation and no space left
I might have missed this in the FAQ-O-Matic.
I have a 20 tape rotation on which I do full backups each night(I'm
backing up about 10Gb on DDS3 tapes). I got this error this
morning. It
looks to me that amanda isn't overwriting DailySet100 but
appending it.
With 20 full backups, I really want amanda to check for the
right tape and
then overwrite it if it is the right one. How do I set this up?
Amanda works so well that I really just feed tapes and do a
amrecover once
in a while. :-)
Thanks,
Keith
A clip of my report follows:
These dumps were to tape DailySet100.
*** A TAPE ERROR OCCURRED: [[writing file: No space left on device]].
Some dumps may have been left in the holding disk.
Run amflush to flush them to tape.
The next tape Amanda expects to use is: DailySet101.
FAILURE AND STRANGE DUMP SUMMARY:
localhost //ntmachine/Users lev 0 STRANGE
localhost //ntmachine/Company lev 0 FAILED [out of tape]
localhost //ntmachine/Company lev 0 FAILED [data write:
Connection
reset by peer]
localhost //ntmachine/Company lev 0 FAILED [dump to tape failed]
STATISTICS:
Total Full Daily
Estimate Time (hrs:min)0:01
Run Time (hrs:min) 1:19
Dump Time (hrs:min)0:24 0:24 0:00
Output Size (meg) 969.7 969.70.0
Original Size (meg) 1764.2 1764.20.0
Avg Compressed Size (%)55.0 55.0--
Filesystems Dumped7 7 0
Avg Dump Rate (k/s) 682.7 682.7--
Tape Time (hrs:min)0:39 0:39 0:00
Tape Size (meg) 969.9 969.90.0
Tape Used (%) 9.89.80.0
Filesystems Taped 7 7 0
Avg Tp Write Rate (k/s) 426.6 426.6--
RE: Tape retirement
AFAIK, AMANDA never formats tapes. They are rewritten from block 1, but not formatted or retensioned. If you keep all your old backup reports, you can figure out how many times a tape has been used, and replace it when the time comes. If you'd like to add tape aging features to the software, please joing the amanda-hackers mailing list. I don't remember that being on the wish list, but I don't think anyone would complain about a useful feature, either. For my DDS3 tapes, I run them until I get errors on a tape twice in a row. Then it gets discarded. I have a stack of spares that I inherited, so replacements are easy. I also test restoring a random tape every month, so I know I'm not just blindly casting my bits to the ether. -Original Message- From: Keith Nasman [mailto:[EMAIL PROTECTED]] Sent: Thursday, September 05, 2002 6:45 PM To: [EMAIL PROTECTED] Subject: Tape retirement So you've got this pool of tapes that amanda rotates through. Does amanda keep track of the number of formats to the tapes and alert you to ones that should be retired? If amanda doesn't keep track of this what strategies are prevalent for DDS3 DAT tapes? Thanks, Keith
RE: What does -1 mean in disklist
Actually, it's a performance management thing. If you have several partitions on one physical drive, you can give them all the same spindle number, and AMANDA will only pull one backup from each spindle at a time, to keep from thrashing your drives. I use this on my older firewalls (200MHz or less) to keep the backups from taking over the machine. -Original Message- From: Gene Heskett [mailto:[EMAIL PROTECTED]] Sent: Tuesday, September 03, 2002 11:29 AM To: Martinez, Michael - CSREES/ISTM; [EMAIL PROTECTED] Subject: Re: What does -1 mean in disklist On Tuesday 03 September 2002 09:41, Martinez, Michael - CSREES/ISTM wrote: What does the -1 mean in the following disklist line: www /home2 nocomp-user -1 local According to the notes in the sample disklist, its a spindle number placeholder, whatever that is. Something to do with a raid array that allows each disk/spindle to be addressed individually. A method of backup up each single disk in a raid (I think). Possibly usefull to those with raids. Untested here, no raid. -- Cheers, Gene AMD K6-III@500mhz 320M Athlon1600XP@1400mhz 512M 99.13% setiathome rank, not too shabby for a WV hillbilly
RE: Questions
After much experimentation I have finally succesfully created an amanda backup. But I have not been able to figure if these features are possible. Congratulations and welcome to my personal favorite backup system. 1) I currently have enabled 4 dumps at a time. What I understand this meaning is the option inparallel does 4 cleint dumps at a time. If I have a disklist of 18 disk does that mean that all 18 disk get backed up to one tape or does that mean that every 4 disk get backed up to tape. This means that up to four dumps are going from the client to the holding disk at a time (no-holding disks go straight to tape, don't worry about that yet.) All of the disks being backed up will go to tape regardless of the inparallel value. 2) When sendszie is executed to estimate how big a directory or disk is is it possible to have sendzie work with the process that dumps to tape to judge if a directory is to big for that tape and load another tape for that backup AMANDA can't handle any partition that has a level 0 size bigger than one tape. The planner program collects the sizes of all of the backups to be done, and then bumps some up or down a level to (a) make sure all needed level 0 backups are done, and (b) fit as much data as possible on the tape. (This is why an accurate tapetype is important, and hardware compression isn't all it's cracked up to be.) If you set runtapes 2 in amanda.conf, AMANDA will spread the backups over the two tapes. I believe the version in development (2.5) will be able to split a partition across tapes. The current version cannot, and if it runs out of tape during a partition, it starts that partition from the beginning on the next tape, to insure it has a good backup. 3) Is it possible just to do fiolesize estimateions before issueing amdump. What I am trying to accomplish is to do estimations and have amdump pick up were it left off to save time. That doesn't make a lot of sense. Are you trying to run the estimates at 1:00 and the dump at 3:00? What if a large file appears in the meantime? Are you trying to resume a backup that was interrupted because it ran out of tape? The remaining partitions are already stored on the holding disk and can be flushed to the next tape. Are you trying to find out how much tape AMANDA will want tonight? You could run amdump with no tape in the drive, and let it run everything to the holding disk (if you have room). The report will then show the size of the backup, and you can then amflush the backups to tape. (This is also how some people use one tape for more than one days' worth of backups if they are DESPARATELY short on tapes.) Are you trying to use two different sizes of tape in the same AMANDA configuration? You might be better off separating the tapes into separate configurations by size, and using each size in a completely separate AMANDA configuration, backing up separate machines. 4) In regards to number 1 to elevate this problem I have been doing amdump configureationname partition. I am doing this for each disk in my disk list is it possible to have certain disk go to one tape and then push otehr disk to go on the second tape. amdump only takes one parameter, the name of the configuration. I expect your partiion parameter is being ignored. Again, if you use runtapes 2, AMANDA will put backups on two tapes as needed. Thanks for all your help Craig Hancock Glad to help, but if you can run an English-language spell check on your messages before posting, it will save everyone time in reading and replying. Thanks. Regards, Paul
RE: two schedules simultaneously?
1) Yes, you can run two schedules, just don't send requests from both at the same time. AMANDA clients assume that they are only talking to one server at a time. `cp -r /amanda/Config1/* /amanda/Config2/` or so. 2) FAQ-O-Matic, flaky but informative: http://amanda.sourceforge.net/fom-serve/cache/28.html 3) You might not want to do this, depending on WHY you think you should do it: - If you're building a second set of tapes to go off site, then a second configuration is probably the right way to go. - If you are doing this because AMANDA is only doing full backups every three weeks, you could just reduce dumpcycle to 5 and get full backups every week. - If you don't like AMANDA spreading full backups over all the tapes to reduce your loss in case of a media failure, then there are a variety of commercial backup packages that haven't figured this out yet. Or a force command noted in the above FOM. -Original Message- From: Wayne Byarlay [mailto:[EMAIL PROTECTED]] Sent: Thursday, August 29, 2002 10:02 AM To: [EMAIL PROTECTED] Subject: two schedules simultaneously? Greetings. I currently run Amanda with a 15-day dump cycle, Mon-Fri, called DailySet1. I want to know if it's possible to do the following: 1. Make DailySet1 only run Mon-Thur; 2. Create a WeeklySet1, with a 5-tape rotation, scheduled for Fridays, doing a level 0 on all filesystems. I was reading the book and it says you can do full archival dumps by creating a new AMANDA configuration with its own set of tapes. This sounds similar to what I want to do but not exactly. Does this mean create a whole 'nother instance of AMANDA in some other directory(my guess), or just create the WeeklySet1 file? The way it sounds, one instance of a total AMANDA configuration can only use one schedule set. any other thoughts on this, too, are indeed welcome. Thanks, Wayne Byarlay Purdue University Libraries
RE: Windows XP
Another possible permission issue is that some of those files may only be accessible to the special user Local System. You can use the CACLS command-line utility to find out what permissions are really on a file. -Original Message- From: JC Simonetti [mailto:[EMAIL PROTECTED]] Sent: Thursday, August 01, 2002 5:39 AM To: [EMAIL PROTECTED] Subject: Re: Windows XP Hi! Maybe not... Windows has a tricky perception of locked files when opened, so could I recommend you to create a Windows account (maybe amanda ;) ) but member of the backup operators group instead of administrators one? And of course make Amanda log into Windows as this account... In some cases this would help... -- Jean-Christian SIMONETTI email: [EMAIL PROTECTED] SysAdmin Wanadoo Portails phone: (+33)492283200 (standard) Sophia Antipolis, France -- On 1 Aug 2002 01:51:43 - [EMAIL PROTECTED] wrote: In this case, amanda is logging into Windows (via smbclient) as Administrator. Shouldn't that override any permission issues? I am trying to backup a Windows XP client (//hans/C$) with Amanda 2.4.2. It seems to work except for a number of files which are listed as errors below. Is there any way to get the files in error backed up with amanda? When I was engaged in my mighty struggle to conquor samba and windows 2k, I found just about all the errors were due to the amanda user not having permission. The exceptions were the system things like pagefile.sys and I think the $NtUninstall kinda thingies. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: erased a tape
I can't think of any reason not to just label the tape with the label it used to have. You'll need to use the -f option with amlabel (because amanda doesn't know the tape was erased) but other than that, the label and the physical media are still both available, there's no reason not to just put them back together. -Original Message- From: Hery Zo RAKOTONDRAMANANA [mailto:[EMAIL PROTECTED]] Sent: Thursday, August 01, 2002 7:35 AM To: [EMAIL PROTECTED] Subject: erased a tape Hi all, I have a running config for amanda now. Somehow, I had some out of tape errors for the last three dumps. I wanted to adjust my tapetype (I have a HP Surestore DDS2 4/8Go). Thought that it's a problem on my tapetype definition so I wanted to run tapetype -f /dev/nst0 on my tape server. Unfortunately, I had an amanda tape on and I think this has erased my datas on this tape (TestConf03). TestConf03 is what amanda is expecting for tonight's dump. I'm hesitating on two options: - re-label my erased tape to TestConf03 and insert it for tonight's dump - remove TestConf03 from my amanda cycle. Label this tape into TestConf(lastnumber)+1 and then use this new tape for tonight's dump. Is there something I missed or is there other way to do this? I think I'll opt for the second solution. Best regards. -- Hery Zo RAKOTONDRAMANANA 16 Rue Ratsimilaho, Antaninarenina, Madagascar Tel: +(261) 20 22 648 83 Fax: +(261) 20 22 661 83
RE: erased a tape
I didn't know it had gotten broken. It works on my version, which is admittedly a little old. (2.4.2-19991216-beta1) -Original Message- From: Gene Heskett [mailto:[EMAIL PROTECTED]] Sent: Thursday, August 01, 2002 11:29 AM To: Bort, Paul; [EMAIL PROTECTED] Subject: Re: erased a tape On Thursday 01 August 2002 11:04, Bort, Paul wrote: I can't think of any reason not to just label the tape with the label it used to have. You'll need to use the -f option with amlabel (because amanda doesn't know the tape was erased) but other than that, the label and the physical media are still both available, there's no reason not to just put them back together. The last time I tried to use the -f option for such a situation, it was ignored. Has this now been fixed in 2.4.3b3-20020702? [...] -- Cheers, Gene AMD K6-III@500mhz 320M Athlon1600XP@1400mhz 512M 99.09% setiathome rank, not too shabby for a WV hillbilly
RE: status of cygwin and amanda
It's been discussed, a couple of people have it working, but we don't have a set of patches suitable for public consumption yet. Since my SAMBA backups are 99% working, I haven't put a lot of time into this lately. It is possible to go there, but there isn't a map yet. -Original Message- From: Michael Perry [mailto:[EMAIL PROTECTED]] Sent: Thursday, July 25, 2002 10:43 AM To: [EMAIL PROTECTED] Subject: status of cygwin and amanda Hi all- I know there is a amanda gui client and applications for win32; but I have a working cygwin build on several systems primarily to use some mail tools like mutt. I also read in the archive that some folks have attempted or perhaps completed a port of the amanda-client to cygwin. I use cygwin on windows 2000 pro here. The cygwin mailing list points to a finished amanda-client port but I cannot find the software anywhere after the initial announcement by the person that did the port. Can anyone direct me to a location for a download of a ported amanda or perhaps give me some information on getting amanda-client working on cygwin? Thanks! -- Michael Perry [EMAIL PROTECTED]
RE: dumpcycle 0 not using holding space
weeks on 2 tapes. I need 2 tapes because I know at least one of the partitions will not fit on the tape. I intend to run amflush AMANDA can't split a partition across tapes, so if you're saying what I think you're saying, and one of your partitions is bigger than one of your tapes, you have a bigger problem than holding disk use. You'll need to split the partition into several tar-able chunks.
RE: Using multiple Tapechangers
The configuration you're describing is not a problem, I've been doing it for two years now. Each different configuration has its own changer setup, and they can be completely independent. -Original Message- From: Patrick Schumacher [mailto:[EMAIL PROTECTED]] Sent: Wednesday, July 24, 2002 4:47 AM To: Amanda-Users Mailingliste Subject: Using multiple Tapechangers hi. afaik, i can use different backup configurations to make different backups on the same server. right? is it possible to use 2 ore more tapechanger devices with different backup configurations (/backup0/, /backup2/ ...)? i know, normaly it must work but i must be sure ;). so want to backup server1 - server4 on changer dev. 1 and server5 - server10 on changer dev. 2 both with a unique/different configuration. regards Patrick Schumacher | technical department phone +49 (0)4192-8794 -440 | fax -290 | easynet DV GmbH | Achtern Dieck 9 | 24576 Bad Bramstedt # easynet is part of the easynet group plc | www.easynetgroup.net
RE: Exabyte eliant vs 85xx
I seem to remember that being true when we were considering upgrading our 8505, but I can't find any hard evidence. Sorry -Original Message- From: Gerhard den Hollander [mailto:[EMAIL PROTECTED]] Sent: Tuesday, July 23, 2002 3:22 AM To: [EMAIL PROTECTED] Subject: Exabyte eliant vs 85xx This is slightly off-topic, but does anyone know if the new(er) exabyte eliant drives are read/write compatible with the 85xx (specific the 8505) series ? Tia Kind regards, -- Gerhard den Hollander Phone :+31-10.280.1515 Global IT Support manager Direct:+31-10.280.1539 Jason Geosystems BV Fax :+31-10.280.1511 (When calling please note: we are in GMT+1) [EMAIL PROTECTED] POBox 1573 visit us at http://www.jasongeo.com 3000 BN Rotterdam JASON...#1 in Reservoir CharacterizationThe Netherlands This e-mail and any attachment is/are intended solely for the named addressee(s) and may contain information that is confidential and privileged. If you are not the intended recipient, we request that you do not disseminate, forward, distribute or copy this e-mail message. If you have received this e-mail message in error, please notify us immediately by telephone and destroy the original message.
RE: automatically ejecting tapes
eject isn't eject, it's 'offline', sorry. mt -f /dev/nst1 rewoffl ejects my DAT every time. (rewoffl is 'rewind and offline') This leads to the handy cron entry: amdump MyConfig1 mt -f /dev/nst1 rewoffl -Original Message- From: Mark Cooke [mailto:[EMAIL PROTECTED]] Sent: Thursday, July 18, 2002 10:13 PM To: [EMAIL PROTECTED] Subject: automatically ejecting tapes Hi, This is kinda off subject, but I was wondering is there a utility to eject tapes from their drive (DAT Drive)? As It would be good to (thru the use of a script) get be able to automatically eject the tape providing the backup has completed. I tried sifting through mt's man pages, but couldn't find any reference to ejecting tape at all. Thanks in advance Mark -- --- To steal ideas from one person is plagiarism; to steal from many is research.
RE: need help urgently!!!
Without your indexes, amrecover won't know what to do. Find the report(s) you got from the tape(s) you want to restore from. Pick your newest level 0, then find the incrementals that follow it. If your tape drive is /dev/nst0 (common for Linux), you can put the level 0 tape in and start with: amanda@yourserver$ mt -f /dev/nst0 rewind amanda@yourserver$ dd if=/dev/nst0 bs=32k count=1 (Should show the 'emergency restore' instructions and header block) If you know how many file systems are between this point on the tape and the one you want, some tape drives will allow you to seek past them: amanda@yourserver$ mt -f /dev/nst0 fsf 3 (Skips /dev/somedisk1 /dev/somedisk2 /dev/somedisk3) If that doesn't work, you'll have to step over them manually: amanda@yourserver$ dd if=/dev/nst0 of=/dev/null bs=32k (Will step over one partition on tape, repeat as needed) Now get the first block again: amanda@yourserver$ dd if=/dev/nst0 bs=32k count=1 (It should contain restore instructions, using dump's restore or tar x; follow them.) Good Luck. I hope this helps. P.S. Credit for the correct bits of the emergency restore instructions go to John R. Jackson, credit for bad bits to my failing memory. -Original Message- From: Carlos White [mailto:[EMAIL PROTECTED]] Sent: Thursday, July 18, 2002 9:20 PM To: [EMAIL PROTECTED] Subject: need help urgently!!! my server just lost a drive this drive contained just about everything i have been doing backups with amanda for some time right now, and a few restores, but that was when my backup info was on the server, i have had to reinstall from scratch on to a new drive, amanda is on it and running, but i now need to restore some data, i just need the info in the mail spool and everything in the /home folder can someone out there help me use amrecover to get it back, or am i just going to have to attempt to do a no amanda restore all help will be greatly appreciated, and the sooner the better thanks
RE:
AMANDA is still actively being developed and used around the world. The Book is still a very good starting point, as the core functionality hasn't changed. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, July 17, 2002 7:22 AM To: [EMAIL PROTECTED] Subject: Hy, I already worked with amanda a bit some while ago. Now, working for a different employer, we're thinking about starting to use it here, too. As i forgeot most of things to know about when using amanda, I was looking for the Book, but i saw it is already three years old. Can i still use it for learning to do Backup with amanda, or did the software changed significyntly in the meantime? Is Amanda at all still in use and in development (i saw the web isn't uopdatet for a while, too) TIA, Henning
RE: Archival tape backup configuration help
I suggested no-reuse instead of upping tapecycle (which I expect would also work) because it covered the case of re-using some tapes and not others. I wasn't sure from the original message whether some tapes would be re-used or not. Your solution is definitely easier if none of the tapes in the rotation will ever be re-used. Paul, If one wanted to store every tape, i.e. never reuse one, is there a difference between setting each to no-reuse after it is written or setting tapecycle 99 (some huge number)?
RE: My tape drive dissappeared after updating the operating system - now what?
Jeff, You didn't include the results of `mt -f /dev/nst0 status`, which might give some information about the state of the drive. Also, I hope you're using /dev/nst0 instead of /dev/st0 in your amanda configuration, or your backups may be no good. You can continue to do backups to your holding disk, which you can then flush to tape later. -Original Message- From: Jeff Q Silverman [mailto:[EMAIL PROTECTED]] Sent: Monday, July 15, 2002 12:06 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: My tape drive dissappeared after updating the operating system - now what? I have been using amanda without a problem for over two years. This weekend, I updated the operating system on my amandaa server and the tape drive broke. I assume that the problem is a software problem with the operating system. With a tape in the drive, the command cat /dev/st0 causes the process to hang in a sleep (wait for I/O) state. I assume that the cat command ought to output the contents of the tape to the next file mark, right? dmesg has the following information (edited for relevence, I hope) Linux version 2.4.9-12smp ([EMAIL PROTECTED]) (gcc version 2.96 2731 (Red Hat Linux 7.1 2.96-85)) #1 SMP Tue Oct 30 18:16:48 EST 2001 (scsi0) Adaptec AIC-7892 Ultra 160/m SCSI host adapter found at PCI 0/9/0 (scsi0) Wide Channel, SCSI ID=7, 32/255 SCBs (scsi0) Downloading sequencer code... 396 instructions downloaded scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.4/5.2.0 Adaptec AIC-7892 Ultra 160/m SCSI host adapter blk: queue c17fc418, I/O limit 4095Mb (mask 0x) Vendor: SEAGATE Model: ST318416N Rev: 0004 Type: Direct-Access ANSI SCSI revision: 03 blk: queue c17fc218, I/O limit 4095Mb (mask 0x) Vendor: ADIC Model: FastStor DLT Rev: 0133 Type: Medium Changer ANSI SCSI revision: 02 blk: queue c17fc018, I/O limit 4095Mb (mask 0x) Vendor: QUANTUM Model: DLT8000 Rev: 0232 Type: Sequential-Access ANSI SCSI revision: 02 blk: queue c15c9a18, I/O limit 4095Mb (mask 0x) Vendor: IBM Model: DDYS-T36950N Rev: S80D Type: Direct-Access ANSI SCSI revision: 03 blk: queue c15c9818, I/O limit 4095Mb (mask 0x) Vendor: QUANTUM Model: ATLAS10K2-TY734L Rev: DDD6 Type: Direct-Access ANSI SCSI revision: 03 blk: queue c15c9418, I/O limit 4095Mb (mask 0x) Vendor: SEAGATE Model: ST173404LWRev: 0002 Type: Direct-Access ANSI SCSI revision: 03 blk: queue c15c9218, I/O limit 4095Mb (mask 0x) Attached scsi disk sda at scsi0, channel 0, id 0, lun 0 Attached scsi disk sdb at scsi0, channel 0, id 5, lun 0 Attached scsi disk sdc at scsi0, channel 0, id 8, lun 0 Attached scsi disk sdd at scsi0, channel 0, id 9, lun 0 (scsi0:0:0:0) Synchronous at 20.0 Mbyte/sec, offset 31. SCSI device sda: 35885168 512-byte hdwr sectors (18373 MB) Partition check: sda: sda1 sda2 sda5 sda6 sda7 sda8 sda9 (scsi0:0:5:0) Synchronous at 160.0 Mbyte/sec, offset 63. SCSI device sdb: 71687340 512-byte hdwr sectors (36704 MB) sdb: sdb1 (scsi0:0:8:0) Synchronous at 160.0 Mbyte/sec, offset 127. SCSI device sdc: 143443640 512-byte hdwr sectors (73443 MB) sdc: sdc1 (scsi0:0:9:0) Synchronous at 160.0 Mbyte/sec, offset 63. SCSI device sdd: 143374738 512-byte hdwr sectors (73408 MB) sdd: sdd1 ... st: Version 20010812, bufsize 32768, wrt 30720, max init. bufs 4, s/g segs 16 Attached scsi tape st0 at scsi0, channel 0, id 4, lun 0 Eventually, the system crashes. The SCSI controller is an AIC7800. Questions: 1) Anybody have any good ideas about what is going wrong and how to fix it? 2) Am I testing the tape subsystem properly? 3) Am I correct in assuming that I cannot do anything else with Amanda until this problem is resolved? The tape drive is an ADIC FastStore 22, which is a DLT tape drive and a changer. Jeff Silverman, sysadmin for the Research Computing Systems (RCS) University of Washington, School of Engineering, Electrical Engineering Dept. Box 352500, Seattle, WA, 98125-2500 [EMAIL PROTECTED] http://rcs.ee.washington.edu/BRL/people/jeffs/
RE: My tape drive dissappeared after updating the operating syste m - now what?
ok, it's got a tape, but it's not rewound. Can you try this with a scratch tape in the drive? (This will erase everything on the tape.) mt -f /dev/nst0 rewind mt -f /dev/nst0 status dd if=/dev/zero of=/dev/nst0 bs=32k count=1 mt -f /dev/nst0 status mt -f /dev/nst0 rewind dd if=/dev/nst0 of=/tmp/zeroes mt -f /dev/nst0 rewoffl mt -f /dev/nst0 status /tmp/zeroes should be 32K long, and all 0x0s. -Original Message- From: Jeff Q Silverman [mailto:[EMAIL PROTECTED]] Sent: Monday, July 15, 2002 1:36 PM To: Bort, Paul Cc: 'Jeff Q Silverman'; [EMAIL PROTECTED] Subject: RE: My tape drive dissappeared after updating the operating syste m - now what? jeffs@truk /atm/cowiche/home/jeffs 1000 $ sudo mt -f /dev/nst0 status Password: SCSI 2 tape drive: File number=-1, block number=-1, partition=0. Tape block size 13344 bytes. Density code 0x61 (unknown to this mt). Soft error count since last status=0 General status bits on (101): ONLINE IM_REP_EN jeffs@truk /atm/cowiche/home/jeffs 1001 $ I symlinked /dev/tape to /dev/nst0. Jeff Silverman, sysadmin for the Research Computing Systems (RCS) University of Washington, School of Engineering, Electrical Engineering Dept. Box 352500, Seattle, WA, 98125-2500 [EMAIL PROTECTED] http://rcs.ee.washington.edu/BRL/people/jeffs/ On Mon, 15 Jul 2002, Bort, Paul wrote: Jeff, You didn't include the results of `mt -f /dev/nst0 status`, which might give some information about the state of the drive. Also, I hope you're using /dev/nst0 instead of /dev/st0 in your amanda configuration, or your backups may be no good. You can continue to do backups to your holding disk, which you can then flush to tape later. -Original Message- From: Jeff Q Silverman [mailto:[EMAIL PROTECTED]] Sent: Monday, July 15, 2002 12:06 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: My tape drive dissappeared after updating the operating system - now what? I have been using amanda without a problem for over two years. This weekend, I updated the operating system on my amandaa server and the tape drive broke. I assume that the problem is a software problem with the operating system. With a tape in the drive, the command cat /dev/st0 causes the process to hang in a sleep (wait for I/O) state. I assume that the cat command ought to output the contents of the tape to the next file mark, right? dmesg has the following information (edited for relevence, I hope) Linux version 2.4.9-12smp ([EMAIL PROTECTED]) (gcc version 2.96 2731 (Red Hat Linux 7.1 2.96-85)) #1 SMP Tue Oct 30 18:16:48 EST 2001 (scsi0) Adaptec AIC-7892 Ultra 160/m SCSI host adapter found at PCI 0/9/0 (scsi0) Wide Channel, SCSI ID=7, 32/255 SCBs (scsi0) Downloading sequencer code... 396 instructions downloaded scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.4/5.2.0 Adaptec AIC-7892 Ultra 160/m SCSI host adapter blk: queue c17fc418, I/O limit 4095Mb (mask 0x) Vendor: SEAGATE Model: ST318416N Rev: 0004 Type: Direct-Access ANSI SCSI revision: 03 blk: queue c17fc218, I/O limit 4095Mb (mask 0x) Vendor: ADIC Model: FastStor DLT Rev: 0133 Type: Medium Changer ANSI SCSI revision: 02 blk: queue c17fc018, I/O limit 4095Mb (mask 0x) Vendor: QUANTUM Model: DLT8000 Rev: 0232 Type: Sequential-Access ANSI SCSI revision: 02 blk: queue c15c9a18, I/O limit 4095Mb (mask 0x) Vendor: IBM Model: DDYS-T36950N Rev: S80D Type: Direct-Access ANSI SCSI revision: 03 blk: queue c15c9818, I/O limit 4095Mb (mask 0x) Vendor: QUANTUM Model: ATLAS10K2-TY734L Rev: DDD6 Type: Direct-Access ANSI SCSI revision: 03 blk: queue c15c9418, I/O limit 4095Mb (mask 0x) Vendor: SEAGATE Model: ST173404LWRev: 0002 Type: Direct-Access ANSI SCSI revision: 03 blk: queue c15c9218, I/O limit 4095Mb (mask 0x) Attached scsi disk sda at scsi0, channel 0, id 0, lun 0 Attached scsi disk sdb at scsi0, channel 0, id 5, lun 0 Attached scsi disk sdc at scsi0, channel 0, id 8, lun 0 Attached scsi disk sdd at scsi0, channel 0, id 9, lun 0 (scsi0:0:0:0) Synchronous at 20.0 Mbyte/sec, offset 31. SCSI device sda: 35885168 512-byte hdwr sectors (18373 MB) Partition check: sda: sda1 sda2 sda5 sda6 sda7 sda8 sda9 (scsi0:0:5:0) Synchronous at 160.0 Mbyte/sec, offset 63. SCSI device sdb: 71687340 512-byte hdwr sectors (36704 MB) sdb: sdb1 (scsi0:0:8:0) Synchronous at 160.0 Mbyte/sec, offset 127. SCSI device sdc: 143443640 512-byte hdwr sectors (73443 MB) sdc: sdc1 (scsi0:0
RE: Archival tape backup configuration help
amanda@tape$ amadmin MyConfig no-reuse MyOffsiteTapeLabel-1
Amanda will keep the record of the backup, but never ask for the tape again.
It is essentially read-only.
You should add another tape to your rotation (with a new name) to replace
it.
-Original Message-
From: Anthony Valentine [mailto:[EMAIL PROTECTED]]
Sent: Friday, July 12, 2002 5:08 PM
To: Jon LaBadie
Cc: Amanda Users
Subject: Re: Archival tape backup configuration help
If I wanted to do something similar, but I didn't want to reuse the
tapes at all (I need to store each one for 7 years), how would I set
Amanda up to do that?
Thanks in advance!
Anthony
On Mon, 2002-07-08 at 12:53, Jon LaBadie wrote:
On Mon, Jul 08, 2002 at 04:57:07PM -0400, Cory Visi wrote:
Ok, I have my daily tape backup set configured and working
perfectly.
Now I
want to setup an archival (available for offsite storage) backup
strategy.
Here is how I want it to work:
Week 1: Full backup to tapes 1 and 2
Week 3: Take tapes 1 and 2 offsite
Week 5: Full backup to tapes 3 and 4
Week 7: Replace tapes 1 and 2, take tapes 3 and 4 offsite
Week 9: Full backup to tapes 1 and 2
etc...
There will be 4 tapes total, 2 tapes per backup. amdump will be
run
every 4 weeks.
The difficulty is that a full backup of our disks do not fit on
one
tape. We can easily fit the excess in the holding space though.
The limitation is not the size of the total backup.
The limitation is the size of the largest disklist entry. No SINGLE
entry
in the disklist can span a tape. If you are using the same list (even
if
in a different file and config) you already know each individual
disklist
entry will fit onto a tape.
Aside from the weird crontab line, how would I configure Amanda to
handle
this strategy? This is what I have right now (I know it's wrong
because
it's not working at all):
dumpcycle 8 weeks
4 weeks
runspercycle 2
1
tapecycle 4 tapes
define dumptype comp-user-full {
comment Non-root partitions on reasonably fast machines (full
only)
record yes
Probably no so your dailies don't think a full was done and will
still do their own at the appropriate times.
index yes
skip-incr yes
Some suggest a dumpcycle of 0 to always force full.
compress client fast
priority medium
}
amdump email returns the following message:
driver: WARNING: got empty schedule from planner
and all the disks get SKIPPED.
I have a feeling skip-incr is not the right setting to use.
Anyone have any ideas?
Thank you for your help,
Cory Visi
End of included message
--
Jon H. LaBadie [EMAIL PROTECTED]
JG Computing
4455 Province Line Road(609) 252-0159
Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: Newbie: Tape doesn't eject?
AMANDA does not automatically eject tapes after a backup. If you want to do this, there is an easy way: When you put your amcheck and amdump commands in cron, do something like this: 30 16 * * 1-5 /usr/sbin/amcheck -m DailySet1 0 4 * * 2-6 /usr/sbin/amdump DailySet1 mt -f /dev/nst0 rewoffl Here's what it means: First line: 30 16: at 4:30 in the afternoon, * * : of any day of month or month of year, 1-5 : Monday through friday, /usr/sbin/amcheck: Check to see if we're ready for backups -m: and mail me any problems, DailySet1: for configuration DailySet1 Second line: 0 4: at 4:00 in the morning, * * : of any day of the month or month of year, 2-6 : Tuesday through Saturday, /usr/sbin/amdump: Run the backup, DailySet1: for configuration DailySet1, : and if that completes successfully, mt : send a tape command -f /dev/nst0: to the first non-rewinding SCSI tape device rewoffl : to rewind and eject. (cron tells stories, they're just in shorthand.) And that's how you get an eject afterwards. Don't worry about using only 10% for now. you can tune that (indirectly) later by adjusting dumpcycle. -Original Message- From: KEVIN ZEMBOWER [mailto:[EMAIL PROTECTED]] Sent: Tuesday, December 11, 2001 5:27 PM To: [EMAIL PROTECTED] Subject: Newbie: Tape doesn't eject? Just completed my first amanda backup successfully, mostly. However, I got an error on three of my five hosts about missing results. Looking this up in the searchable archives led me to the /tmp/amanda/sendsize.*.debug file, and this showed that I had a directory named /etc/amandates, instead of a file. I removed the directory, touched the file, then chmod and chown it to amanda:disk 0600. Should be fine, I think. However, when my backup finished, the report noted that the tape was only 10% full, and it didn't eject. Is this normal amanda behavior, to not eject a tape until it's full? I want to run amdump over again right away. Should I manually eject the tape (with mt) and insert a new one, or just start amdump again with the same tape in the drive? Thanks, again, for all your help. -Kevin Zembower - E. Kevin Zembower Unix Administrator Johns Hopkins University/Center for Communications Programs 111 Market Place, Suite 310 Baltimore, MD 21202 410-659-6139
RE: Skipping a tape
Yes, that would be the right way to do it. This is why many people on the list have recommended having and extra tape or two in the rotation as spares. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 13, 2002 7:35 PM To: Bort, Paul Cc: [EMAIL PROTECTED] Subject: Re: Skipping a tape Thanks for the answer. However, I belive I am hitting the issue 1) that you mentioned. My configuration is as follows: dumpcycle 2 weeks # the number of days in the normal dump cycle runspercycle 10 # the number of amdump runs in dumpcycle days tapecycle 22 tapes # the number of tapes in rotation Could I perhaps change tapecycle to 21 tapes for a short time? Thanks again, d Bort, Paul wrote: No need to edit the tapelist file directly, almost ever. In this case, you can tell AMANDA not to use the tape, then tell her it is available again later. Remove a tape from rotation: amanda@tape$ amadmin YOURCONFIG no-reuse YOURTAPE Return a tape to rotation: amanda@tape$ amadmin YOURCONFIG reuse YOURTAPE Notes: 1) AMANDA may ask for a new tape if this reduces the number of active tapes below the tapecycle value in amanda.conf. 2) Index information is preserved for restores, so don't worry about being able to restore from this tape 3) If you're alternating between a set of on-site and off-site tapes, you may want to add a third set that sits idle at your office waiting to be used. This gives you time to change tapes off-site. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 13, 2002 4:38 PM To: [EMAIL PROTECTED] Subject: Skipping a tape Hi, I am wondering if it is possible to safely skip a tape that happens to be unavailable (it was moved offsite but hasn't returned). I've looked at the $AMANDA/tapelist file and am tempted to edit, but am hesitating because I don't actually understand how Amanda uses the information. In particular, does the line position matter? The tape that Amanda wants is the last tape in tapelist, and it also has the earliest last-use date (the first field). If I wanted Amanda to just move to the next tape, should I just edit the date field? Move the line to the top of the file? Both? tia, d
RE: Backing up PostgreSQL?
Sorry, when the previous poster mentioned 'snapshot', I was thinking of SQL Server's 'dump', which is transactionally consistent, because it's done by the SQL engine. I thought Oracle had a similar method for producing a usable backup of the SQL server? -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Friday, June 14, 2002 2:53 PM To: Bort, Paul Cc: [EMAIL PROTECTED] Subject: RE: Backing up PostgreSQL? [ On Friday, June 14, 2002 at 12:20:21 (-0400), Bort, Paul wrote: ] Subject: RE: Backing up PostgreSQL? I don't know much about PostgresQL, but on MS SQL server and Oracle (IIRC), any update that would leave the database inconsistent should be inside a transaction. Any snapshot will not happen while a transaction is in progress; therefore the snapshot is consistent and restorable. I guess it depends on how sane your programmers are. Oracle, running on any snapshot-capable unix (including FreeBSD) and using a normal filesystem for storage, will not -- cannot possibly -- guarantee that a snapshot will not happen while a transaction is in progress. There is no possible interlock in any snapshot implementations I'm aware of between the kernel (which does the snapshot operation) and the user-land Oracle process(es). -- Greg A. Woods +1 416 218-0098; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Planix, Inc. [EMAIL PROTECTED]; VE3TCP; Secrets of the Weird [EMAIL PROTECTED]
RE: Ongoing write errors with taper
My guess then is that the tape drive itself is screwed. You could go through the SCSI dance*, but I doubt it would help. * SCSI Dance: Power down everything on that SCSI bus, including the host, and unplug, clean, and replug every cable on the SCSI bus. Power up all the peripherals before the host, and pray you didn't break anything. (See also TAPI two-step.) -Original Message- From: Nicki Messerschmidt [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 13, 2002 3:41 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: RE: Ongoing write errors with taper Bort, Paul [EMAIL PROTECTED] schrieb: [bitching about tape errors; logs as attached] My guess would be because tar and tapetype aren't looking for excessive write errors, they're looking for EOT. That might be te point. But even if I use tar or tapetype the errors should make it into /var/log/messages, because the kernel receives a message from the st module, am I right? And the kernel itself writes an entry into /var/log/message wehter the application responds to this signal or not is negligible... But please prove me wrong with this! Maybe the drive needs to be cleaned? Are there other factors (temperature, time of day?) that could be causing the problem? The tape is cleaned every day and as soon as the excessive write errors occur it states that it needs to be cleaned. This is very strange! Cheers Nicki -- Linksystem Muenchen GmbH [EMAIL PROTECTED] Schloerstrasse 10 http://www.link-m.de 80634 Muenchen Tel. 089 / 890 518-0 We make the Net work. Fax 089 / 890 518-77
RE: incremental bumped to level x incremental? I want full!
Yes.
dumpcycle 5 means that there has to be a level 0 backup within every 5
dumps.
dumpcycle 0 means that there has to be a level 0 backup every 0 dumps, which
the planner interprets as being perpetually due for a level 0.
You could probably set dumpcycle -1 if you wanted to always be OVERDUE for a
level 0 backup. :)
-Original Message-
From: Tom Van de Wiele [mailto:[EMAIL PROTECTED]]
Sent: Thursday, June 13, 2002 7:14 AM
To: Bort, Paul
Cc: [EMAIL PROTECTED]
Subject: RE: incremental bumped to level x incremental? I want full!
if I put dumpcycle 0 will it overwrite the tape with a *FULL* backup
every time?
On Wed, 2002-06-12 at 18:02, Bort, Paul wrote:
You might need to put 'dumpcycle 0' at the beginning of
amanda.conf instead
of inside a dumptype. (This is a guess. I put it at the
beginning and it
works there, but I haven't bothered to upgrade from version
2.4.2-19991216-beta1, because it 'ain't broke'.)
Levels are degrees of incremental-ness. Level 1 is all of
the files that
have changed since level 0, and level 2 is all of the files
that have
changed since level 1. A file system gets 'bumped' to a
lower level when the
space it would take at the higher level is needed for other
backups. If you
look at the planner (IIRC) debug file, it will show the
estimated size of
each backup at different levels and why it chose the levels it did.
-Original Message-
From: Tom Van de Wiele [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, June 12, 2002 10:43 AM
To: [EMAIL PROTECTED]
Subject: incremental bumped to level x incremental? I want full!
Hello list!
My objective: to make a FULL backup on tape every
weekday. If there is
data on the tape, I want it removed and/or overwritten by
new data...
First of all, I'm using version 2.4.2p2 on a Linux server (Dell
PowerEdge) with kernel 2.4.16
my 'disklist' file contains:
delta.eduline.be hda1 always-full
delta.eduline.be hda4 always-full
delta.eduline.be hda3 always-full
in my amanda.conf I have:
define dumptype always-full {
global
comment Full dump of this filesystem always
program GNUTAR
compress none
index yes
priority high
dumpcycle 0
}
The logs gave me this:
NOTES:
planner: Incremental of delta.eduline.be:hda3 bumped to level 2.
planner: Incremental of delta.eduline.be:hda4 bumped to level 2.
Incremental? I don't want incremental backups, I want a
FULL backup.
And what are these levels in combination with the
incremental term?
Kind regards
-- Tom Van de Wiele
RE: amanda and vpnd
There are several available approaches: (in no particular order) 1) reconfigure the VPN tunnel so that the AMANDA traffic is not subject to NAT. 2) use a separate VPN tunnel for AMANDA that bypasses NAT. 3) rebuild AMANDA with a specific port range (search the archives for details). 4) rebuild AMANDA without the port check (not recommended, but also in the archive.) -Original Message- From: adi [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 13, 2002 4:33 PM To: [EMAIL PROTECTED] Subject: Re: amanda and vpnd -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi again, if i can't make this to work, then can i make amanda accept if it is coming from a high numbered port or even any number port? Thanks. On Friday 14 June 2002 00:43, adi wrote: Hello, i have a problem running amcheck. My amanda server has a private ip and needs to backup a remote server(live ip). Amanda server connects to the vpnd server which is on a local lan and tunnel thru the vpn connection in order to do a backup. The remote machine which has to be backed up has a serial link established to the vpn server. The first time i tried running amcheck, it complains that it's(the vpn server) is coming from a port which is not secure. So i need to redirect connections coming from amanda server to a well-known port and then forward it to the amanda client. Correct? I wonder if anyone has been successfull at this attempt? This is my ipchains on the vpn server. /sbin/ipchains -A forward -i sl0 -s 192.168.1.85 -d 192.168.200.2 -j MASQ /sbin/ipchains -A input -j REDIRECT 600 -p udp -s 192.168.1.85 -d 192.168.200.2 10080 192.168.1.85 is the amanda server. 192.168.200.2 is the amanda client The vpn server serial ip is 192.168.100.2. This is the output of amcheck. Amanda Backup Client Hosts Check WARNING: zeus.vpn-remote: selfcheck request timed out. Host down? Client check: 1 host checked in 30.006 seconds, 1 problem found I've been cracking my head for 2 days now.. help needed, thanks. -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9CQFpInIYkBVpGqURAuEfAJ92h1BMgNCn5JO/gei+MGI3FsmvHgCdH6R4 TLu3DbvCM/gyQDolAqwvjUg= =DuEj -END PGP SIGNATURE-
RE: Skipping a tape
No need to edit the tapelist file directly, almost ever. In this case, you can tell AMANDA not to use the tape, then tell her it is available again later. Remove a tape from rotation: amanda@tape$ amadmin YOURCONFIG no-reuse YOURTAPE Return a tape to rotation: amanda@tape$ amadmin YOURCONFIG reuse YOURTAPE Notes: 1) AMANDA may ask for a new tape if this reduces the number of active tapes below the tapecycle value in amanda.conf. 2) Index information is preserved for restores, so don't worry about being able to restore from this tape 3) If you're alternating between a set of on-site and off-site tapes, you may want to add a third set that sits idle at your office waiting to be used. This gives you time to change tapes off-site. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 13, 2002 4:38 PM To: [EMAIL PROTECTED] Subject: Skipping a tape Hi, I am wondering if it is possible to safely skip a tape that happens to be unavailable (it was moved offsite but hasn't returned). I've looked at the $AMANDA/tapelist file and am tempted to edit, but am hesitating because I don't actually understand how Amanda uses the information. In particular, does the line position matter? The tape that Amanda wants is the last tape in tapelist, and it also has the earliest last-use date (the first field). If I wanted Amanda to just move to the next tape, should I just edit the date field? Move the line to the top of the file? Both? tia, d
RE: incremental bumped to level x incremental? I want full!
You might need to put 'dumpcycle 0' at the beginning of amanda.conf instead
of inside a dumptype. (This is a guess. I put it at the beginning and it
works there, but I haven't bothered to upgrade from version
2.4.2-19991216-beta1, because it 'ain't broke'.)
Levels are degrees of incremental-ness. Level 1 is all of the files that
have changed since level 0, and level 2 is all of the files that have
changed since level 1. A file system gets 'bumped' to a lower level when the
space it would take at the higher level is needed for other backups. If you
look at the planner (IIRC) debug file, it will show the estimated size of
each backup at different levels and why it chose the levels it did.
-Original Message-
From: Tom Van de Wiele [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, June 12, 2002 10:43 AM
To: [EMAIL PROTECTED]
Subject: incremental bumped to level x incremental? I want full!
Hello list!
My objective: to make a FULL backup on tape every weekday. If there is
data on the tape, I want it removed and/or overwritten by new data...
First of all, I'm using version 2.4.2p2 on a Linux server (Dell
PowerEdge) with kernel 2.4.16
my 'disklist' file contains:
delta.eduline.be hda1 always-full
delta.eduline.be hda4 always-full
delta.eduline.be hda3 always-full
in my amanda.conf I have:
define dumptype always-full {
global
comment Full dump of this filesystem always
program GNUTAR
compress none
index yes
priority high
dumpcycle 0
}
The logs gave me this:
NOTES:
planner: Incremental of delta.eduline.be:hda3 bumped to level 2.
planner: Incremental of delta.eduline.be:hda4 bumped to level 2.
Incremental? I don't want incremental backups, I want a FULL backup.
And what are these levels in combination with the incremental term?
Kind regards
-- Tom Van de Wiele
RE: Ongoing write errors with taper
My guess would be because tar and tapetype aren't looking for excessive
write errors, they're looking for EOT.
Maybe the drive needs to be cleaned? Are there other factors (temperature,
time of day?) that could be causing the problem?
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, June 12, 2002 1:13 PM
To: [EMAIL PROTECTED]
Subject: Ongoing write errors with taper
Hi listreaders,
perhaps can one of you explain me, why the taper of amanda
produces excessive
write errors on my Tandberg SLR50 but neither the tapetype
program nor a 24/7
running tar (to test the tapedrive) spit out this errors.
If this happens once more I think I will go mad... *g*
I think I append part of /var/log/messages and the amanda logfiles.
/var/log/messages
Jun 4 17:47:03: st: Version 20020205, bufsize 32768, wrt
30720, max init.
bufs 4, s/g segs 16
Jun 4 17:47:03: Attached scsi tape st0 at scsi0, channel 2,
id 6, lun 0
Jun 5 00:45:02: st0: Block limits 1 - 262144 bytes.
Jun 8 01:32:58: st0: Error with sense data: Info fld=0x40,
Current st09:00:
sense key Medium Error
Jun 8 01:32:58: Additional sense indicates Excessive write errors
Jun 8 01:32:58: st0: Error with sense data: Info fld=0x1,
Current st09:00:
sense key Medium Error
Jun 8 01:32:58: Additional sense indicates Excessive write errors
Jun 8 01:32:58: st0: Error on write filemark.
And I use this tapetype in the amanda.conf
define tapetype SLR {
comment SLR 50 Tapedefinition by Tapetype
length 45000 mbytes
filemark 0 kbytes
speed 1784 kps
}
Yeah, I know that this tapedrive only has 25Gb uncompressed
storage capacity,
but the error occures usually after about 2-10 Gb.
I hope that anyone has an Idea why this happens...
Cheers
Nicki Messerschmidt
RE: amrestore problems
Maybe /dev/ait2 != /dev/nst0 ? amrestore: could not open tape /dev/ait2: Permission denied snip $ ls -l /dev/nst0 crw-rw1 root disk 9, 128 Aug 30 2001 /dev/nst0 I can read for the tape with dd: $ mt -f /dev/ait2 fsf 1; dd if=/dev/ait2 bs=32k count=1 AMANDA: FILE 20020606 zambezi /boot lev 1 comp N program /bin/gtar To restore, position tape at start of file and run: dd if=tape bs=32k skip=1 | /bin/gtar -f... - etc, all the way thru the tape.
RE: amanda and windows
AMANDA kind of understands SMB already: you can specify a server and share to back up on a regular *nix client with //server/share as the mountpoint. There are other steps needed, but that's the start. You could smbmount, but there may be issues. Search the mailing list archive. As for Cygwin, some are using it, I'm trying to get it working, but haven't had much time lately. getfsent.c is the sticking point, needs to be edited to work in Cygwin. (Sorry I don't have anything that could be diff'ed yet.) -Original Message- From: Eric Bergeron [mailto:[EMAIL PROTECTED]] Sent: Wednesday, June 05, 2002 2:20 PM To: [EMAIL PROTECTED] Subject: amanda and windows My question is what is the prefered way to backup a win2k computer with amanda? I searched the FAQ-O-Matic and this mailing list's archives with no luck. Is the prefered method to smbmount C:\ and then tar what you want or do you install cgywin and amanda on the win2k computer and configure amanda as normal? If this is a RTFM please point me to the right document. Thank you for the help and time. -- Eric
RE: cygwin/dump
Thoughts: 1. The 'dump' utility is specific to the filesystem it's written for. Porting Linux 'dump' for ext2 to Cygwin doesn't make sense unless your Cygwin has local access to ext2 partitions that you want to back up. 2. There is already a tar in Cygwin, but I have no idea if it's a good one or a bad one. 3. If the underlying file system has locked a file, nothing can back it up without talking nicely to the underlying operating system. Running Cygwin as a user who is a member of the 'backup operator' group might help. 4. 'Compile Amanda on a windows machine' is an interesting proposition. I've gotten it to compile (client only) under Cygwin by commenting out a line in getfsent.c, IIRC. (something to do with mnttab) I don't have a server set up on that network to test it with yet. (There was no point in putting up the server until I had clients.) 5. I recommend (and use) a two-stage approach for backing up Windows boxes. Stage 1 is to have a complete install procedure for the OS and applications. Be able to rebuild from a formatted disk. Keep copies of the CDs and instructions on site and off. Stage 2 is to backup the data only with AMANDA and smbclient. 6. The only backup program I've ever seen that could restore a Windows NT box to a usable state is Ghost. If you really need fast restores, consider Ghosting the machine to a file and backing up that file. -Original Message- From: Tony [mailto:[EMAIL PROTECTED]] Sent: Wednesday, May 22, 2002 7:38 PM To: [EMAIL PROTECTED] Subject: cygwin/dump anyone have any experience/luck using cygwin utils for performing dumps from a windows box? One thought I had was that one could install cygwin and compile amanda on a windows machine ( b/c cygwin comes w/ gcc ) to get the box to 'appear' like a *nix box. I've had problems w/ using smbclient to perform backups b/c programs that are currently in use ( on a M$ platform ) can be copied. thoughts?? -Tony
RE: what to do if I put in the wrong tape
One other possibility is that you are running amcheck, and it's kicking the tape out. -Original Message- From: John Rosendahl [mailto:[EMAIL PROTECTED]] Sent: Monday, April 29, 2002 12:53 PM Cc: [EMAIL PROTECTED] Subject: Re: what to do if I put in the wrong tape I am using 2.4.3b3 and that is really not what happens, I can come up with a few reasons for this 1) I am using 2.4.3b3 and b means beta 2) I am backing up files that are much larger than my holding disks 40 gigs fo files for 4 gigs of holding disk 3) Could it be something with the tape changer, when ever I screw up I come to work to find that the tape has beed ejected. Thanks for your help -John Joshua Baker-LePain wrote: On Mon, 29 Apr 2002 at 10:11am, John Rosendahl wrote Is there a procedure that people use when they forget to replace the tape/put in the wrong tape. Amanda stalls when this happens, is there a way to get it going again(with the right tape in of course). Amanda shouldn't stall. It should do the backup to holding disk and then, in the email, tell you that a tape error occurred and you need to run amflush. Which you then do.
RE: firewire/IDE drive idea
There is development going on to allow AMANDA to write to any device as a backup medium. No need to get creative with the holding disk. You could then skip having to mount/unmount the drive, and really just change it like a tape, as long as the base OS doesn't freak out. I think it's in beta. (N.B. I'm still using AMANDA 2.4.2-19991216-beta1 because it was and is more solid than the commercial solution it replaced.) Also, because the backup medium is of greater general use (read: easy to pawn once stolen), you might want to consider physical security of the medium when away from your site, and possibly encrypting the backups. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Monday, April 29, 2002 2:42 PM To: [EMAIL PROTECTED] Cc: James Hanson Subject: firewire/IDE drive idea I'd like your opinion on an alternative to tape drives that would use amanda to organize backups. Let's say that our amanda tape server would have a firewire connection to an IDE drive. Since firewire is fast, and hot-pluggable, one could unmount the drive, disconnect the firewire, swap out the IDE drive with another each day, plug it back in, reformat and mount it. Let's assume we have a stock of about forty 60 G IDE hard drives. On a 30 day cycle, we swap these drives, and at the beginning of the month, put the most recent disk into cold storage for archive. Each night, the tape server goes out and dumps to tape, but there being no tape drive, sends the info instead to the holding disk, our firewire special. Looking at the costs, $100 for the firewire interface and $4200 for the hard drives gives you the complete system, assuming you have a linux box hanging around to mount it on. This compares favorably with a AIT system, which looks to be about $3000 for the drive and $100 per tape, 15 tapes needed for about a 2 week tapecycle. Advantages would be stability of medium, speed of access, nonproprietary nature (which contributes to redundancy, since any linux computer could mount the IDE drive). Disadvantages - bulk/weight of disks, sensitivity to damage from dropping, slightly more complicated procedure to change 'tapes', others I haven't thought of yet... So what do you think? - Feasible? - Anyone already doing this? (is this a solved problem?) - What about indexing? - What have I overlooked? -- John Rodkey, Information Technology, Westmont College [EMAIL PROTECTED]
RE: Please help -- planner askfor: lev out of range -1..10: 10
You could at least temporarily go to 'runtapes 2' in amanda.conf to make
more room in each run. If you can afford more tapes, using two a day might
help, or using two tape drives with chg-multi might be good too.
Lengthening the tapecycle is going to take a while (about two tapecycles,
I'm guessing) to level out as you would expect. It sounds like you need more
tape now, which a second tape and/or a second drive could do.
Also, rather than regenerating the disklist every night, you might not
confuse amanda as much if you simply checked the directory structure every
night, and warned when a disklist entry is close to the limit. This would
allow you to plan splits better, and having your disk list entries change
less often is a good thing if you have to restore.
-Original Message-
From: Brashers, Bart -- MFG, Inc. [mailto:[EMAIL PROTECTED]]
Sent: Monday, April 22, 2002 11:52 AM
To: '[EMAIL PROTECTED]'
Subject: Please help -- planner askfor: lev out of range -1..10: 10
I didn't get any responses from my posting last Thursday, so
I'm guessing
I'm really up the creek on this one.
If anyone has any suggestions as to how to fix this problem,
please let me
know. I've now gone 4 days without a backup, and I'm afraid
I'm going to
have to delete everything and start from scratch!
Thanks,
Bart
After many months of operation (THANK YOU to all the developers!) last
night's amanda Daily run gave me this error:
FAILURE AND STRANGE DUMP SUMMARY:
planner: FATAL error [planner askfor: lev out of range -1..10: 10]
beowulf/usr/local RESULTS MISSING
beowulf/var/mail RESULTS MISSING
(and so on, for every element of the disklist)
The only debug files from this (9 PM) run were
/tmp/amanda.amtrmlog.*.debug
/tmp/amanda/amtrmidx.*.debug
but they showed nothing particularly interesting or helpful.
The latter
said
could not open index directory
/var/lib/amanda/Daily/index/db02/_root/
but that's because I recently added that machine-directory combination
(oops, forgot it) and amanda hasn't managed to schedule a
full dump, and
won't incrementally dump a new disk.
The relevant parts of my amanda.conf file are:
dumpcycle 3 weeks
runspercycle 15
tapecycle 17 tapes
define dumptype mydumptype {
program GNUTAR
compress none
} (plus more of the usual options)
I searched the archive, but could not find anything directly
relevant. This
lev out of range error happened to me once before, when I was first
learning and setting up amanda. That time, I just deleted
all the logs and
tapelists and everything, and started from scratch. I would
like to avoid
repeating that.
The only other (possibly) relevant detail is that I
constantly get errors
like:
FAILURE AND STRANGE DUMP SUMMARY:
beowulf/home/proj/mm5 lev 1 FAILED [dump larger than
tape, skipping
incremental]
beowulf/var/mail lev 1 FAILED [dumps way too big, must skip
incremental dumps]
(and so on, for ~30 or so directories)
I've written a script that runs (via crontab) every afternoon and
re-generates the disklist. It starts with a list of directories (/etc
/var/log /var/lib /usr/spool /root /home /data). For every
directory it
finds that contains more than 25Gb, it recursively breaks it
up and adds the
sub-directories to the disklist. This is because /home is
currently 153Gb,
and would never get dumped since it's larger than a tape.
I have about 200Gb to backup to a DLT8000 (40 Gb) tape drive.
I recently
went from a 2 to 3 week dumpcycle, hoping that the extra
space would get rid
of the dumps way too big error.
Please cc me with any hints as to how to clear this up, as I
only get daily
digests and I don't want to wait until tomorrow to fix this...
Bart
--
Bart Brashers MFG Inc.
Air Quality Meteorologist 19203 36th Ave W Suite 101
[EMAIL PROTECTED]Lynnwood WA 98036-5707
http://www.mfgenv.com 425.921.4000 Fax: 425.921.4040
RE: Please help -- planner askfor: lev out of range -1..10: 10
Thanks for your responses, Jon and Paul, I don't think going to 'runtapes 2' in amanda.conf is going to fix the lev out of range error, since that part happens long after my error. The email message starts out with FAILURE AND STRANGE DUMP SUMMARY: planner: FATAL error [planner askfor: lev out of range -1..10: 10] beowulf/usr/local RESULTS MISSING beowulf/var/mail RESULTS MISSING and there's nothing in /tmp/amanda/dump/20020419/ and the like. The planner fails, so nothing is ever gtar'd, so there's nothing to write to tape(s). Any ideas on how to clear the lev out of range error? Then I'll tackle the other errors stemming from having way to much data to backup compared to my tapesize. Bart, increasing the number of tapes should address this problem, because the planner is trying to figure out what level to run for each backup, and for unknown reasons is bouncing that backup down to a lower level than dump can handle. (Or one part of planner understands that 10 is the limit and another thinks that 9 is the limit.) Additional tape space will allow planner to go with a higher level for that (and other) backups. Paul
RE: dos partion
dump is fs-specific. If your dump is for ext2 file systems, it won't understand FAT or FAT32 at all, and bails rahter than risking something. (The specific error is because it is looking for a superblock (or sblock), and the place that ext2 would have put one has something else in FAT, which doesn't make sense as an sblock number) change the dump type to comp-user-tar, make sure that comp-user-tar has the options you want, and you should be set. Oh, and make sure you have a non-broken version of tar. 1.13.19 is recommended. -Original Message- From: Tom Beer [mailto:[EMAIL PROTECTED]] Sent: Monday, April 22, 2002 2:12 PM To: [EMAIL PROTECTED] Subject: dos partion Hi, I want to back a dos partition on a dual boot machine. I mount this parition /mnt/dos and the disklist entry is vaio.system ad0s1 comp-user sendsize.debug states bad sblock number entiry dump terminated. Any pointers? Thanks Tom
RE: Amanda and Encryption
Another option, depending on how many machines are on the public network, is a VPN tunnel, like FreeSWAN or CIPE. (This also provides a secure channel for other useful things like monitoring and remote control, if desired). -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Monday, April 22, 2002 3:45 PM To: Chad Morland Cc: [EMAIL PROTECTED] Subject: Re: Amanda and Encryption Hello, Encryption is possible via gnupg and gzip. Please read the info at the the following url: http://security.uchicago.edu/tools/gpg-amanda/ I am in the process of writting a document to cover this. It is about 90% done and is not up to date with 2.4.3. ( I dont know if there are any differences with 2.4.3, I assume none) If you would like I can forward you what I have. Its info is complete but it needs to be proof read again, so understand its not perfect. I am currently running this setup and love it. It has not messed up on me once. Andrew On Mon, 22 Apr 2002, Chad Morland wrote: I was wondering if it is possible to encrypt the traffic that amanda sends over a network. I have data that I need to backup and some of the data will have to travel over the public internet. Because this data contains sensitive information I do not want the data to be sent unencrypted. Has anyone configured amanda to use SSL/TLS or even SSH? Thanks. -CM
RE: Amanda and NT shares
As far as I know, Samba won't change that bit just because it's tar that's asking for the file instead of a user. gnutar has its own mechanism (gnutar-lists) for tracking which files should be backed up during an incremental. So to answer your questions: 1) I don't think so 2) Nothing If we could figure out how to get AMANDA to compile under Cygwin (someone on the list said they had done it, but didn't say how) then I would have more interesting answers. -Original Message- From: Mike Hendrix [mailto:[EMAIL PROTECTED]] Sent: Tuesday, April 16, 2002 10:50 AM To: [EMAIL PROTECTED] Subject: Amanda and NT shares I am trying to figure something and am looking for assistance. I have checked the newsgroups without success. The question is: When I backup an NT/2000 share does amanda/tar even touch the archive bit on a file?? Can someone explain exactly what happens?? --- I have not failed. I've just found 10,000 ways that won't work. - Thomas Edison Michael Hendrix [EMAIL PROTECTED] Systems Engineer / SysAdmin Team Leader Logical Net / Capital Net (518) 292-4509
RE: NT share basckup
Did you --with-smbclient when you built AMANDA? Here's the batch file I used for building Amanda on my server, which does work with SAMBA: #!/bin/bash ./configure \ --with-user=amanda \ --with-group=disk \ --with-config=MYCONFIG \ --with-tape-device=/dev/nst0 \ --with-changer-device=/dev/sch0 \ --exec-prefix=/usr \ --sysconfdir=/ \ --with-smbclient=/usr/bin/smbclient \ --with-gnutar=/bin/tar \ --with-samba-user=username \ --with-gnutar-listdir=/amanda/Krakow1/gnutar-list make make install I think the --with-smbclient and --with-samba-user options will be most helpful. -Original Message- From: David Flood [mailto:[EMAIL PROTECTED]] Sent: Friday, April 12, 2002 10:04 AM To: [EMAIL PROTECTED] Subject: NT share basckup I've set up a share on my NT box. I can do: 'smbclient //PC-NAME/SHARE -U username' after entering the password I can see the contents of the share. So I set up this share to be backed up by amanda. I have: Created a share on the PC Created /etc/amandapass with the correct information. Created a dumptype for this backup which is using GNUTAR as required. Added the entry to the disklist in the format: TAPESEVER //PC-NAME/SHARE Dumptype where TAPESERVER is the server doing the backup but also has samba. when I run amcheck config I get the following: ERROR: father: [This client is not configured for samba: //wasp/backup] ERROR: father: [GNUTAR program not available] ERROR: father: [can not read/write /opt/amanda/var/amanda/gnutar-lists/.: No such file or directory] ERROR: father: [SMBCLIENT program not available] Has anyone got any suggestions. David Flood Systems Administrator
RE: Using Remote Tape Drive on Windows NT
Unconfirmed rumor has it that Amanda can run under Cygwin. Whether this
includes the server side and/or tape writes is unknown. This might neatly
solve your problem.
If you do get it working, please, please let us know how.
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Thursday, March 28, 2002 11:33 AM
To: [EMAIL PROTECTED]
Subject: Using Remote Tape Drive on Windows NT
SHORT VERSION:
I would like to use amanda to backup a handful of
linux and windows NT boxes where the tape drive is
installed on one of the Windows NT boxes. Can anyone
point me in the right direction?
Is there a better free (as in beer) solution for my
particular problem?
LONG VERSION:
I have at my disposal both a single DLT drive as well
as a fancier tape robot. Unfortunately, both of these
are installed on a Windows NT server and due to
political constraints this can not be easily
changed at this time. My group is responsible for
administering all the servers involved. It is with
hardware changes that the political barriers are
encountered.
A portion of the tapes in the tape robot have been
configured with software that makes them appear as a
simple filesystem. (This may or may not be useful
information.)
Is there any way to run an amanda server on one of the
Linux boxes but use the tape drive(s) on the Windows
NT server? I am not opposed to writting a days worth
of code to accomplish this task.
Is there a better free (as in beer) solution for my
particular problem. {Large software purchases also
encounter political heat.}
RE: question to the audience.....
For short scripts and such perhaps they should also go in the FAQ-o-Matic? -Original Message- From: Don Potter [mailto:[EMAIL PROTECTED]] Sent: Tuesday, March 26, 2002 9:54 AM To: Matthew Boeckman; [EMAIL PROTECTED] Subject: Re: question to the audience. That would be a good idea...something similar to the plugins that BigBrother has. Not endoresed by BigBrother but made availbale as user contributions. Matthew Boeckman wrote: I think it's a great idea. Would it be possible to have a page at amanda.org to post them instead of the list? I'd love to go to www.amanda.org/scripts (or something) instead of searching for perl regex'es in the amanda-users archive... just a thought... Don Potter wrote: I'm sure we all have a selection of scripts that we use to make our backup lives that much easier. Would there be any objections posting any scripts that we find to be prudent to this mailer. Majority of mine are quick and nasty bourne scripts which can be pasted in, but they could prove to be useful out there. Just inquiring to see if there is a interest. Don
RE: amflush and irc?
It depends on how often you require full backups. I was getting really low tape utilization and big swings in amadmin 'balance' results, so I cut my dumpcycle and runspercycle in half, and now I get more consistent tape usage, and level 0 backups twice as often as a side benefit. Amanda's tape fill planning degrades gracefully when there is excess tape, nothing to worry about, but it still degrades. -Original Message- From: Jon LaBadie [mailto:[EMAIL PROTECTED]] Sent: Thursday, March 21, 2002 2:30 PM To: [EMAIL PROTECTED] Subject: Re: amflush and irc? On Thu, Mar 21, 2002 at 12:16:10PM -0600, Brandon D. Valentine wrote: On Thu, 21 Mar 2002, Jon LaBadie wrote: You won't. After running a while amanda will spread the level 0's over all the dump days. So if your once a week tape now gets 40GB and your daily incrementals now get 2, each will average about 8 GB daily. Actually it's even better than that. Each tape should average a full 40GB. Amanda will do its best to fill each and every tape to capacity, promoting as many full dumps per run as possible. If you have 40GB of data and a 40GB tape drive, you should get a full set of level 0s on every run. YMMV, but that has not been my experience. My tapes are 12GB (dds3) and although I have about 50GB (data precompression, not disk) to back up, my nightly tape usage is about 7GB. I thought the planner attempted to get a consistant nightly backup, not a filled-tape backup. Our differing observations may be the result of different amanda.conf settings. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: Confusion on Dumps Configuration
Sometimes 'dump' is one run of amdump, sometimes it is the 'dump' utility that makes a backup of a filesystem. The number of filesystems that fit on a tape will vary by their size and compressability. Your bump factors in amanda.conf will also affect it. Levels are degrees of incremental-ness in Amanda. Level 0 is a full backup. Level 1 is everything that changed since the last level 0. Level 2 is everything that changed since the last level 1. This goes on to level 9. Amanda automatically picks the 'best' level to use for each file system based on a bunch of things, including how often you want full backups (dumpcycle), how much a backup of each file system will take at each level, and how compressible that backup is likely to be, based on historical records. Your best bet for tape fitting is usually to add file systems gradually over several amdump runs, so that the initial level 0 for each new filesystem can be fit on the tape, bumping existing filesystems to level 1 or lower to make room if needed. I like to add the biggest file systems first, but that's me. -Original Message- From: Mark Schoonover [mailto:[EMAIL PROTECTED]] Sent: Thursday, March 21, 2002 3:51 PM To: 'Joshua Baker-LePain'; Mark Schoonover; 'Jenn Sturm'; 'Bort, Paul' Cc: Amanda-User (E-mail) Subject: RE: Confusion on Dumps Configuration Josh, Jenn Paul, Thanks for the quick replies! I have found the error of my ways, which leads me to a few more questions! The way I understand it is a 'dump' is one cron job of amdump running. OK no problem there. Now, how do I estimate how many filesystems I can backup to a single tape?? Trial and error?? Not sure what a level 0 is offhand. The best I can figure out on my own is an entire dump of all filesystems in one run. Josh, can I bother you for a copy of your tapetype?? Thanks again! .mark
RE: filemarkers
The AMANDA filemarks setting covers the amount of space that ending one tar or dump and starting the next takes. Some tape drives don't take much space for a filemark, some take lots. AMANDA uses this number as part of estimating how much tape is needed and when it should bump backups down levels. If you don't have a high enough filemark setting, you could get unexplained random EOT errors. If you have it set too high, planner will bump backups to higher levels than needed. I would leave it set to whatever tapetype likes, but changing it isn't immediately fatal either, if you have some room to spare on your tapes. -Original Message- From: Paul Yeatman [mailto:[EMAIL PROTECTED]] Sent: Wednesday, March 13, 2002 1:02 AM To: Amanda user's group Subject: filemarkers Hi, Now that I've added a DDS4 tape drive to my backup arsenal and will be using DDS4 tapes, I'm requestioning the filemark parameter in tapetypes. The DDS3 drive I've used all this time has had a tapetype with a filemark of 0. I believe this is what the 'tapetype' test produced and, when I checked with other tapetypes and possibly people on the list that such a value wasn't ludicrous, I went with it...and I've experienced no problems because of it. But, looking over the tapetypes in general again, this value is usual non-zero which is making me requestion my decision. Why do so many people use a non-zero filemark value (one tapetype I saw for a DLT used one over a meg)? What's the point? If 0 works fine (is this only true for me?), why ever use anything else? I'm missing the point of any advantage it serves. Just curious if anyone had any input or a good answer for this. A quick search for filemark in the archives was futile as I mostly conjured up nothing but tapetype listings instead of discussion on the matter. -- Paul Yeatman (858) 534-9896 [EMAIL PROTECTED]
RE: Restoring w/o amanda
instead of the | /bin/tar -f..., try dumping the output of dd to a file: dd if=$tape bs-32k skip=1 /tmp/ddoutput Alternatively, rewind the tape to the beginning, and start dd'ing blocks until you get to one that you can't read. Rewind again, and 'skip=' the number of blocks you could read. -Original Message- From: Brad Tilley [mailto:[EMAIL PROTECTED]] Sent: Monday, March 11, 2002 1:50 PM To: [EMAIL PROTECTED] Subject: Restoring w/o amanda Hello, We are doing a disaster recover check, and we would like to test amanda restoration without amanda being present. I have followed the directions in 'Unix Backup and Recovery' (O'Reilly) by W. Curtis Preston (a very good book... well worth the money), but I can't get any further than reading the amanda header that contains the backup info. Here's what I do: [root@reg root]# dd if=/dev/st0 bs=32k skip=2 count=1 AMANDA: FILE 20020304 reg /root lev 0 comp N program /bin/tar To restore, position tape at start of file and run: dd if=$tape bs=32k skip=1 | /bin/tar -f... - 1+0 records in 1+0 records out [root@reg root]# So, I cd to a /tmp directory and try to run the command as it appeared in the amanda header, but I always get a tar error. I think I am probably overlooking something simple in tar, but I don't know what. I've never done this before, so please don't flame me too much if I'm doing something stupid. Thank you, Brad
RE: tapedev=otherhost:/dev/nst0 ?
Then make the spare Linux box the one that uses the Samba client to pull data from the Windows boxes. We've been doing that here for a while with no problems. Sample disklist amandaserver /dev/hda1 amandaserver /dev/hda2 spareserver /dev/hda1 spareserver //win1/cshare spareserver //win1/dshare -Original Message- From: Jordi Vidal [mailto:[EMAIL PROTECTED]] Sent: Friday, March 08, 2002 2:39 PM To: Joshua Baker-LePain Cc: Amanda List Subject: Re: tapedev=otherhost:/dev/nst0 ? Thank for your reply I know the client best option but it dont fit my needs. The clients are Windowses, and I have a spare Linux not in production to carry the load compression. On Fri, 8 Mar 2002, Joshua Baker-LePain wrote: Can I build amanda server in a host without tape and use some other host tapes? The goal is to save the tapeserver (the machine that has the tape device) from CPU load when using compression best Why not just use compress client best? That makes the clients do the compression, not the tape server.
RE: Disaster Recovery on Windows
Backup only the data. Keep a copy of the install CDs and install instructions on site and off site. Practice your re-install. (We can re-install a box in under four hours with Citrix and all apps.) -Original Message- From: Jan Boshoff [mailto:[EMAIL PROTECTED]] Sent: Monday, February 25, 2002 3:58 PM To: Amanda Users Subject: Disaster Recovery on Windows Hi all I just had a thought and would like to probe the great minds of this list. Would anyone do a disaster recovery type of restore of a Winbox that was backed up using smbclient?I'm currently backing up the complete winbox, but it occured to me that I don't know how I would restore if something happened to the Winbox that we needed to restore the complete drive. I mean, you need to have Win installed for smbclient to talk to it accross the network. Any thoughts would be helpful! Thanks again! Jan -- Jan Boshoff PhD Student, Chemical Engineering Univ. of Delaware, DE USA www.che.udel.edu/research_groups/nanomodeling
RE: Problems with dumps
Paul, A couple of things might help track down where the problem is coming from. First, if you can add a third tape to a run, that will indicate whether the problem is with AMANDA's tape size estimate or not. Next, do you have enough holding disk space for that backup? You can test this by configuring that entry in disklist to use a backup type that does not go to the holding disk. That should help narrow down where the limitation is. Paul -Original Message- From: Paul Lussier [mailto:[EMAIL PROTECTED]] Sent: Thursday, January 03, 2002 11:35 AM To: [EMAIL PROTECTED] Subject: Problems with dumps Hi all, I'm having trouble getting one of my clients backed up. There are 13 file systems on the client which need to be dumped, totalling about 12GB of data. I'm getting error messages like the following for several of the file systems: hacluster1 /dev/sda12 lev 0 FAILED [dumps too big, but cannot\ incremental dump skip-incr disk] I know that amanda seems to think that the dumps are too big, and failing these file systems because I've dis-allowed incremental backups. However, I've also specified the use of 2 tapes for the backups, and amanda doesn't seem to be filling both: Estimate Time (hrs:min)0:10 Run Time (hrs:min) 9:26 Dump Time (hrs:min)7:51 7:49 0:02 Output Size (meg) 54711.454709.12.2 Original Size (meg) 89777.289755.3 22.0 Avg Compressed Size (%)60.9 61.0 10.2 (level:#disks ...) Filesystems Dumped 33 28 5 (1:5) Avg Dump Rate (k/s) 1981.3 1991.6 15.7 Tape Time (hrs:min)6:50 6:50 0:00 Tape Size (meg) 54712.454710.02.4 Tape Used (%) 156.3 156.30.0 (level:#disks ...) Filesystems Taped33 28 5 (1:5) From the 'Tape Used' it appears that amanda is only filling 50% of the second tape. The 'Tape Size' seems to indicate I'm only filling about 55GB worth of tape. I'm using a DLT7000 drive with DLT4 tapes. I should be able to get 70GBs worth of data across 2 tapes, no? So, I should be able to get another 15GBs onto the second tape, by my calculations, which is fine, since there's less than 12GB currently failing. Any ideas? Thanks, -- Seeya, Paul God Bless America! ...we don't need to be perfect to be the best around, and we never stop trying to be better. Tom Clancy, The Bear and The Dragon
RE: Tape changer question
What you are asking for is technically possible, and not all that difficult, but ill-advised. If something blows up during the backup on Sunday, you are back to the previous Sunday's backups, if you still have them. (If you don't have them, you're really toast.) Unless bludgeoned otherwise, AMANDA will automatically distribute full backup (level 0) of the different file systems across a series of tapes, so that a catastrophic failure of the tape server only costs you a small impact in restorability. The indexes keep track of which backups are on which tapes, allowing you to restore disks or files by selection. AMANDA will then tell you which tape(s) you need. This will also give you more predictable run times for your backups. -Original Message- From: Ian Eure [mailto:[EMAIL PROTECTED]] Sent: Friday, February 15, 2002 3:16 PM To: [EMAIL PROTECTED] Subject: Tape changer question Hi there. I have a HP C1557A (SureStore DAT 24x6) tape changer that I'm using with Amanda. What I'd like to do is this: Full backup on Sunday night, split across the 6 tapes in the changer Incremental backups Monday-Saturday, each on one of the 6 tapes in the changer. Do I need two different backup types to do this? E.g. one Weekly backup job, and a Daily job that backs up to the first tape in the changer, and moves to the next tape each day after? Any advice is appreciated.
RE: Tape changer question
if you spread the full backups out as AMANDA is inclined to do, you can probably go to runtapes 1 and she will make every effort to fill that tape with backup goodness every night. Since you have runtapes 6, she will try to fill six tapes every night. This doesn't sound like what you want. From what you've said, you have more than 7 tapes. Tapecycle should be the total number of tapes you have to use, plus maybe a couple spares. (I like to keep it at exactly the number I rotate, and have blanks ready to amlabel in case of trouble.) runspercycle should probably be 6 if you want to use all six tapes and change them once a week. dumpcycle is how often you want a guaranteed set of level 0 backups. One week is probably good for now. I think runspercycle * runtapes * dumpcycle (in days) == tapecycle is a good starting point. As for the flush, it didn't happen because AMANDA didn't find the number (runtapes) of empty tapes she was expecting. You can use amflush after changing runtapes to flush the backups. There is more info on-line at http://www.backupcentral.com/amanda.html. Amanda is configured to use my changer thusly: -- snip -- runtapes 6 # explained in WHATS.NEW tpchanger chg-zd-mtx # the tape-changer glue script, see TAPE.CHANGERS changerdev /dev/sg0 tapetype HP-CHANGER labelstr ^GH-((I)?(F)?)-[0-9][0-9]-[0-9]$ -- snip -- And for rotations: -- snip -- dumpcycle 1 weeks # the number of days in the normal dump cycle tapecycle 7 tapes # the number of tapes in rotation bumpsize 20 MB # minimum savings (threshold) to bump level 1 - 2 bumpdays 1 # minimum days at each level bumpmult 4 # threshold = bumpsize * (level-1)**bumpmult -- snip --
RE: Reg:Do we have to label the cleaning tape?
iii) How do I eject my magazine? I used the eject button in the tape drive to eject the magazine to place the cleaning tape into the magazine, but it loads the first tape and not the one(say 4) which was there before I ejected(which seems quite obvious). But how do I bring back to the original configuration(ie 4) First do an amtape /config/ rest, followed by an amcheck /config/ which should leave the drive with the next required tape loaded if it is in the magazine. Otherwise reload the magazine with the next set of tapes and repeat. Amanda needs a base camp to start from by doing the reset after you've had the magazine out. On some drives (Seagate robots) you can push-button up the slot to load if the same tape is still in the same slot as before the magazine was ejected. I can confirm that if the same slot is loaded before and after the magazine change, AMANDA is ok. When I do a mag change on my Exabyte EXB-10h, I just eject the tape that is in the drive, take out the mag, put in the mag, and put the tape from the same slot in the mag in the drive. No need to reset, and the regularly scheduled amcheck steps to the next tape and everything is fine.
RE: Missing Tape has Backups Screwed
amadmin YOURCONFIG no-reuse MISSINGTAPE Then use the next tape in sequence. Just remember to mark it 'reuse' before it is next scheduled. -Original Message- From: Stephen Carville [mailto:[EMAIL PROTECTED]] Sent: Friday, February 08, 2002 2:37 PM To: Amanda Users Subject: Missing Tape has Backups Screwed One of my tapes did not come back from offsite in time for today's backup (Some winidiot sent if off for four weeks instead of one) so today's backup failed. How can I get amanda to just skip that tape and go on to the next one in the sequence? When I try amcheck, amanda thinks all of the tapes in the changer are active tapes and will not write to them. Even if I load the next tape in the sequence mnaually, amcheck rejects it. How can I tell which tapes it will write too? dumpcycle 7 days runspercycle 5 tapecycle 15 tapes runtapes 2 So far, backups have only needed one tape per run so it seems to me that tapes from two weeks ago should no longer be active but amanda thinks they are. -- Stephen Carville UNIX and Network Administrator DPSI (formerly Ace USA Flood Services) 310-342-3602 [EMAIL PROTECTED]
RE: Missing Tape has Backups Screwed
Oops. Do you have runtapes 2 ? I think you would have to mark all of the tapes that are in this run as no-reuse to get it to cycle around. Do you have runtapes * runspercycle = tapecycle? (ie, runtapes=2, runspercycle=5, tapecycle=10 or something like that?) I should have been more explicit about there needing to be spare tapes in the rotation for no-reuse to do what you would expect. From what I've seen on the list, it's a good idea to have tapecycle runtapes*runspercycle, for just this reason. You can temporarily pull a tape from the rotation without confusing AMANDA. (For example, I have runtapes=1, runspercycle=5, tapecycle=30, so I have six complete backup sets, one for each of the last six weeks. I picked those numbers based on the number of tape magazines I had for my changer.) Good Luck. -Original Message- From: Stephen Carville [mailto:[EMAIL PROTECTED]] Sent: Friday, February 08, 2002 4:21 PM To: Bort, Paul Cc: Amanda Users Subject: RE: Missing Tape has Backups Screwed On Fri, 8 Feb 2002, Bort, Paul wrote: - amadmin YOURCONFIG no-reuse MISSINGTAPE - - Then use the next tape in sequence. - - Just remember to mark it 'reuse' before it is next scheduled. Didn't work. $ amadmin daily1 tape The next Amanda run should go onto tape C83173 or a new tape. The next Amanda run should go onto tape C83174 or a new tape. $ amadmin daily1 no-reuse C83173 amadmin: marking tape C83173 as not reusable. $ amtape daily1 label C83174 amtape: scanning for tape with label C83174 amtape: slot 9: date 20020121 label C83174 (exact label match) amtape: label C83174 is now loaded. amdump reports *** A TAPE ERROR OCCURRED: [new tape not found in rack]. When I try 'amcheck daily1' it reports that _all_ the tapes are active. Even tapes that 'amtape daily1 info' says are not part of the current backup. -- Stephen Carville UNIX and Network Administrator DPSI (formerly Ace USA Flood Services) 310-342-3602 [EMAIL PROTECTED]
RE: Amanda install reality check
1. REBDA (Read Everything Before Doing Anything) 2. Be prepared to run the configure/install process a few times until you get it the way you want. 3. Remember to do both the server setup and client setup on the server. 4. Start by just backing up the backup server. 5. Start by changing tapes by hand. Add the changer once all the clients are working. 6. /tmp/amanda/*debug 7. Yes it's a complex install, but the reward is industrial-strength cross-platform backup and restore that you can hack to your specifications. (No Backup Exec bitterness here, no.) 8. Build your own. Whoever made the RPM or DEB didn't have your network in mind. Any other suggestions from the list? -Original Message- From: W. D. [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 06, 2002 4:03 PM To: [EMAIL PROTECTED] Cc: KEVIN ZEMBOWER Subject: Re: Amanda install reality check Any hints, tips or gotchas that would be helpful for the 'uninitiated'? (Especially FreeBSD HP SureStore DAT 40) At 14:09 2/6/2002, KEVIN ZEMBOWER, wrote: For-what-it's-worth dept.: In the year that I've been a full-time Unix system administrator, I guess I've installed 40-50 different packages, mostly from source. Amanda was the second most time-consuming and difficult; only Xwindows was harder for me. -Kevin Zembower - E. Kevin Zembower Unix Administrator Johns Hopkins University/Center for Communications Programs 111 Market Place, Suite 310 Baltimore, MD 21202 410-659-6139 gene [EMAIL PROTECTED] 02/06/02 01:49PM snip I seem to be really struggling to get this to work. I did think it would be easier than this ;-) /snip Gene Start Here to Find It Fast!© - http://www.US-Webmasters.com/start.htm
