Using HD for backup
Hi, How can I use, for exmple, the "/mnt/backups" path for storing backups, instead of a tape? Thanks.
tapetype Tandberg-DLT8000-40
Hi,
here are the defenitions for the tandberg DTL8000. I run tapetype and got numbers
lower than
hw specs, bu I think thats normal. If I encounter problems with this defintions, I
will post.
define tapetype Tandberg-DLT8000-40 {
comment "Tandberg DTL8000 with DTLtape IV-Cartridge"
length 4 mbytes
filemark 32 kbytes
speed6000 kps
}
--
Henning Heinold
Re: Using HD for backup
Adolfo Pachn wrote: Hi, How can I use, for exmple, the "/mnt/backups" path for storing backups, instead of a tape? Thanks. In simple terms: specify /mnt/backups as your holding disk. Make sure you have set your reserve option in configuration, otherwise amanda will not make level 0 backups to holding disk. And specify a tapetype that suits your needs. Amanda will take the tapetype to balance and plan your backups, but the free space on "/mnt/backups" as a typetype wont do any good to amanda. Take one that is higher than your biggest level 0 dump. Amrecover has no problems restoring from your holding disk, but you have to make sure there is enough room for the backups. (but perhaps i missunderstood you and you want to mount something other than a harddisk under /mnt/backups) I had this problem, cause there was intended to keep lev0 on tape and lev1 on hard disk. There is this two-configuration setting wich are using the same dirs in the FAQ, but i didn't want to flush the incrementals, so i wrote a shell script to delete the backup files along with their indexes and logs. (i hope there is no "flush-to-dev-null-on-demand-option" or "clean-up-holding-disk" somewere that i have missed ... ) The script is based on determining the size of backups in the holding disk and if necessary deleting some of them. I modified it a little bit to post it on the list and added a testing option which causes it to _not_ delete anything by default. It is currently only tested and used on a SuSE 7.0 Linux System. The Script could also come handy as a fall-back to ensure the space left in the holding disk if there's no tape online. (imagine a little test if there is a tape available and if not, make some room in the holding disk, so the backups could be done) Now I do sometimes flush the backups, as the level 0 backups sometimes don't fit on one tape and I _have_ to flush manually ;-) Simon Mayr PS Great new "incronly" option! Very good for a two-configuration setting. #!/bin/sh # # author: simon mayr # email : [EMAIL PROTECTED] # # purpose: make room in the holding disk without flushing the backups to tape # # usage: define all the parameters in the script #place it in front of amanda in your cronjob, so that in #worst case it can not delete the backup you've just made #!!! test before real use # # needed: some sanity checking, so you should really test before use! # perhaps some usefull return values # command line parameters # # Use at your own risk. # I am NOT responsible for anything YOU do with this script # feel free to modify and copy , bla bla ... # # # how it works: # # find out how much space we occupy in the holding disk # is it too much? # no # - end script # yes # - find the oldest amanda dir in holding disks # delete it # delete corresponding indexfileslogfiles # start over again ( with a maximum of trys to avoid infinity loops) # CONFIGURATION (parameters): # on or off # turn off when you are sure nothing bad will happen TESTING=on # your configuration (directoryname) for the not-going-to-tape-backups CONFDIR=DailySet1 # the holding disk(s) (seperate with "\ ") # eg HOLDING=/disk1/\ /path/to/disk2/\ /disk3/ HOLDING=/holding_disk/ # this is the main option in this script, # it controls if anything at all will be done # adjust to this value: # allowed_usage_of_the_holding_disk(s) - desired_min_free_space_for_one_backup # eg 40gig holdingdisk, 12gig backup, 4gig for safety # means 40-(12+4)=24gigs=24000 in mbytes MAXMB=24000 # maximum number of deleted dates # if anything is in your holding disk, that can not be deleted # the script will try again and again # OR if your MAXMB is too low you would possibly loose all your backups MAXWHILE=8 # END OF CONFIGURATION # use du to get the size of our images # this will NOT work, if your du produces other output than # number [directory] # with this options COUNT=0 for VAR in `du $HOLDING -ms | cut -f 1` do COUNT=$(($VAR+$COUNT)) done # this is for testing, because # the while loop only produces output if there is anything to do # so you get only mail from your cronjob, when something was deleted if ! test "off" = "$TESTING" then echo $COUNT MB used, $MAXMB allowed fi # the main loop while test $MAXWHILE -gt 0 test $COUNT -gt $MAXMB do echo $COUNT MB used, $MAXMB allowed # DELDATE contain the oldest date from your backups in the # holding disks # (a bit clumsy, but it works for me ) DELDATE= for DIRNAME in $HOLDING do cd $DIRNAME for VAR in `ls -d [0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9] 2 /dev/null` do test $VAR -lt $DELDATE DELDATE=$VAR done done if test $DELDATE = then echo and there is nothing I could delete to lower the disk usage echo Terminating now! exit 0 fi # tell what to do echo removing $DELDATE # remove DELDATE in all
Re: Unable to skip Files on tape?
On Sun, Jan 14, 2001 at 09:24:19PM +0100, Frank Schiebel wrote: Hello Jean-Louis! Try to configure amanda with "--with-broken-fsf" Great, it works now! Thanks a lot for the hint! Whats the matter with broken fsf. Is it a problem with the tape device or with amanda or a combination of the two? That's a problem with your tape driver. /* * tapefd_fsf_broken -- handle systems that have a broken fsf operation * and cannot do an fsf operation unless they are positioned at a tape * mark (or BOT). This shows up in amrestore as I/O errors when skipping. */ Jean-Louis -- Jean-Louis Martineau email: [EMAIL PROTECTED] Departement IRO, Universite de Montreal C.P. 6128, Succ. CENTRE-VILLETel: (514) 343-6111 ext. 3529 Montreal, Canada, H3C 3J7Fax: (514) 343-5834
Will amanda do what I need?
I have a small peer to peer network with several windows 95 machines and a windows 2000 machine and a linux machine. Ideally, what I would like to be able to do is to backup everything, on a regular, to a very large (and prehaps removable) ata hard drive on the linux box. If I can't do that, the next best solution would be to back up everything to something like an Onstream 50 GB backup tape drive on the linux box. (Presently, my only tape drive is a 4 GB dat, which is virtually useless with today's large drives.) I am looking for the lowest cost, practical solution. Is Amanda what I am looking for? If not, can anyone suggest a solution. (Another possibility is Novastore Novanet-8, but this is pretty expensive, especially when you add the cost of a large tape drive. TIA, Ed
hmm. What's wrong with this picture?
STATISTICS:
Total Full Daily
Dump Time (hrs:min)0:48 0:48 0:00 (0:00 start, 0:00
idle)
Output Size (meg) 927.1 927.10.0
Original Size (meg) 0.00.00.0
Avg Compressed Size (%) -- -- --
Tape Used (%) 5.65.60.0
Filesystems Dumped3 3 0
Avg Dump Rate (k/s) 331.2 331.2--
Avg Tp Write Rate (k/s) 331.1 331.1--
---
Who says you can't create value from nothing huh?
:o)
minimum number of tapes needed?
Hello, Does amanda need a minimum number of tapes to finish a dumpcycle of, say, 7 days? My situation is as follows: There are two hosts to be backed up, one having a du of 7G and the other 3.5G. Day to day changes is less than 1% of the total du for each host. I have a single tape drive with 15G of uncompressed capacity (~30G compressed). To start, I plan on having a dumpcycle of 7 with a runcycle of 5. Can I set the tapecycle to 1 (or 2) and still have things backed up properly? The reason I have this restriction is because the machines in question are at a colocation facility, so changing tapes everyday is inconvenient at best. Going by the simple math in the UBR amanda chapter, I have: 11G / 7 = 1.6G 11G / 7 * 1(.01) = 0.02G 11G / 7 * 2(.01) = 0.03G 11G / 7 * 3(.01) = 0.05G 11G / 7 * 4(.01) = 0.07G 11G / 7 * 5(.01) = 0.08G 11G / 7 * 6(.01) = 0.10G = 2.0G I assume that the first run will do a complete level 0 dump, which will be 11G. So over a period of one week, 13G of tape would be used. Does my math make sense here? What pitfalls should I expect by using only one tape per week? I've also read that the tapecyle should be ((Runs + 1) * #tapes), so in my case, this would be two tapes. So I suppose this means that I should at the least use 2 tapes per dumpcycle (which is one week)? Also, by using one (or two) tapes means that amanda has to fit multiple images onto a single tape. Will this be a problem? How easy is it to recover from a tape with multiple images? Thanks for helping a newbie, Antony
Re: RedHat 7.0
On Tue, 16 Jan 2001, Chris Baker wrote: Can anyone help me, since I have upgraded two of my machines to RedHat 7.0 I am getting errors with amanda stating in its Mail Report that the servers FAILED!: - FAILURE AND STRANGE DUMP SUMMARY: web /export lev 0 FAILED [Request to web timed out.] web /home lev 0 FAILED [Request to web timed out.] web /etc lev 0 FAILED [Request to web timed out.] ns2 /etc lev 1 FAILED [no estimate or historical data] - Anyone got any suggestions?, I am running Amanda v2.4.1 First, this should probably have gone to amanda-users, and not amanda-hackers, but that nit aside, my guess is the upgrade "upgraded" your /etc/inetd.conf file and you will need to put back the entries for amandad, etc. Please direct further discussion on this to [EMAIL PROTECTED], and not to amanda-hackers. Marc Mengel [EMAIL PROTECTED]
RedHat 7.0
Can anyone help me, since I have upgraded two of my machines to RedHat 7.0 I am getting errors with amanda stating in its Mail Report that the servers FAILED!: - FAILURE AND STRANGE DUMP SUMMARY: web /export lev 0 FAILED [Request to web timed out.] web /home lev 0 FAILED [Request to web timed out.] web /etc lev 0 FAILED [Request to web timed out.] ns2 /etc lev 1 FAILED [no estimate or historical data] - Anyone got any suggestions?, I am running Amanda v2.4.1 Regards, Chris
Dump not running
Hi, Whenever I run 'amcheck' I get [root@paperbark sbin]# ./amcheck hlink-data2 Amanda Tape Server Host Check - /usr/local/amanda/dumps: 1900471 KB disk space available, that's plenty. amcheck-server: slot 8: date 20010112 label hlink-data12 (first labelstr match) NOTE: skipping tape-writable test. Tape hlink-data12 label ok. Server check took 6.004 seconds. Amanda Backup Client Hosts Check Client check: 1 host checked in 0.565 seconds, 0 problems found. (brought to you by Amanda 2.4.1p1) which to me would appear as though the dump is going to run correctly, however the cronjob that I have set up for amdump never actually runs. I have tried changing my disklist to only backup the /tmp directory and this appears to run fine, but if I try to backup my /user directory 'amdump' never appears to finish. so i tried to run amverify and it just does this [root@paperbark sbin]# ./amverify hlink-data2 Tape changer is chg-multi... 1 slot... Verify summary to root Defects file is /tmp/amverify.24428/defects amverify hlink-data2 Tue Jan 16 14:14:28 EST 2001 Loading current slot... Using device aborted!for device to go ready... cat: /tmp/amverify.24428/tapelist: No such file or directory and I eventually have to push ctrl-c to stop it, it nevers becomes 'ready' what should I be looking at to try and work out what is happening cheers Nathan
Diagnosing client-side errors
I am trying to back up a single partition having just installed Amanda. I thought I'd try with this in my `disklist': scooby sda8 always-full I get the following error, but can't work out what I'm doing wrong. Any tips? In general, how can I diagnose client-side problems? Thanks, B. Amanda Backup Client Hosts Check ERROR: scooby: [can not access sda8 (sda8): No such file or directory] Client check: 1 host checked in 0.014 seconds, 1 problem found (brought to you by Amanda 2.4.2)
Re: RedHat 7.0
Anyone got any suggestions?, I am running Amanda v2.4.1 First, this should probably have gone to amanda-users, and not amanda-hackers, but that nit aside, my guess is the upgrade "upgraded" your /etc/inetd.conf file and you will need to put back the entries for amandad, etc. Red Hat Linux 7 uses `xinetd', so the format of the configuration file for the superserver has changed quite dramatically. You need to create a file in /etc/xinetd.d/ similar to the other files in that directory. It should be fairly self-explanatory. Cheers, Ben
amcheck with gnutar
Hello, I am trying to amcheck with gnutar. And, I've got this error: - Amanda Backup Client Hosts Check ERROR: dirac: [can not execute /usr/local/libexec/runtar: Permission denied] ERROR: dirac: [can not read/write /usr/local/var/amanda/gnutar-lists/.: Permissi on denied] Client check: 1 host checked in 0.103 seconds, 2 problems found (brought to you by Amanda 2.4.2) How can I do to get Permission allowed? Takayuki Murai -$BB<0f!!N4G7(B- [EMAIL PROTECTED]
No Subject
Hello, I am trying to amcheck with gnutar. And, I've got this error: - Amanda Backup Client Hosts Check ERROR: dirac: [can not execute /usr/local/libexec/runtar: Permission denied] ERROR: dirac: [can not read/write /usr/local/var/amanda/gnutar-lists/.: Permissi on denied] Client check: 1 host checked in 0.103 seconds, 2 problems found (brought to you by Amanda 2.4.2) The files of permissions are: -rwsr-x--- 1 rootamanda 52300 Jan 15 17:06 runtar drw-rw-rw- 2 amanda amanda 512 Jan 16 15:58 gnutar-lists How can I do to get Permission allowed? Takayuki Murai -$BB<0f!!N4G7(B- [EMAIL PROTECTED]
Re: amandad busy
* Alexandre Oliva ([EMAIL PROTECTED]) wrote: On Jan 9, 2001, Mitch Collinsworth [EMAIL PROTECTED] wrote: This sounds like you have more than one amanda server (or more than one instance on the same server) trying to run backups on that client at the same time. Or you have the same client listed with different names in the disklist. Turns out I was being an idiot--we have two tapestacks on the same machine and run two amanda configs at the same time. I had some of the disks in that host in one config and the other partition in the other config, so the two were fighting over who got to back it up. Now if I could just get the stupid box to *not* take 30+ hours to backup a 15 GB partition I'd be in good shape (and I have hard set the interface to 100/FD in solaris and on the switch). Thanks for all the help! -- .michael lea _/_/ when in danger or in Associate System Engineer _/_/doubt, run in circles, [EMAIL PROTECTED] _/_/ scream and shout
