Re: Listed-incremental interaction
On Mon, 6 Aug 2001, John R. Jackson wrote: GNU tar. Instead, it creates a copy of the file (xxx.new). In the case of a full dump, it will copy /dev/null to the file. Otherwise it copies oh, I see. the live name for the next run. But in the case of no-record, the rename is bypassed and the file effectively thrown away. ah, that's good. Thank you. I will verify that on the next backup :)
problem running amrecover
Hi, so as of late im running amanda to make nice shiny backups of all my servers. Now ive tried for the first time to actually recover soemthing and lo and behold, im running into problems. what im trying to do is recover a few files of one of the partitions i back up whats happening is the following: AMRECOVER Version 2.4.1p1. Contacting server on backup ... 220 backup AMANDA index server (2.4.2p2) ready. 200 Access OK Setting restore date to today (2001-08-07) 200 Working date set to 2001-08-07. 200 Config set to work. 200 Dump host set to xxx. $CWD '/home/httpd/htdocs/tsv_entwicklung/templates/spiele/spielplan' is on disk 'sda8' mounted at '/home'. 200 Disk set to sda8. /home/httpd/htdocs/tsv_entwicklung/templates/spiele/spielplan WARNING: not on root of selected filesystem, check man-page! amrecover ls 2001-08-07 . 2001-08-07 index.html 2001-08-07 index_00_01.html 2001-08-07 index_alt.html amrecover add index.html Added /httpd/htdocs/tsv_entwicklung/templates/spiele/spielplan/index.html amrecover extract Extracting files using tape drive /dev/null on host backup. The following tapes are needed: ABT451 Restoring files into directory /home/httpd/htdocs/tsv_entwicklung/templates/spiele/spielplan Continue? [Y/n]: Load tape ABT451 now Continue? [Y/n]: EOF, check amidxtaped.debug file. amrecover: short block 0 bytes UNKNOWN file amrecover: Can't read file header extract_list - child returned non-zero status: 1 amidxtaped.debug: amidxtaped: debug 1 pid 19204 ruid 502 euid 502 start time Tue Aug 7 17:18:34 2001 amidxtaped: version 2.4.2p2 SECURITY USER root bsd security: remote host xxx user root local user amanda amandahosts security check passed 6 amrestore_nargs=6 -h -p /dev/null xxx ^sda8$ 20010807 Ready to execv amrestore with: path = /usr/local/amanda/sbin/amrestore argv[0] = amrestore argv[1] = -h argv[2] = -p argv[3] = /dev/null argv[4] = xxx argv[5] = ^sda8$ argv[6] = 20010807 amrestore: missing file header block amrestore: WARNING: not at start of tape, file numbers will be offset amrestore: 0: reached end of tape: date ^L amidxtaped: amrestore terminated normally with status: 1 Rewinding tape: done amidxtaped: pid 19204 finish time Tue Aug 7 17:18:34 2001 not having much experience with amanda and all im a bit stumped as to whats going on here anyone got any ideas? thx, -- Arne Hueggenberg [EMAIL PROTECTED] System Administrator Sports Bytes GmbH http://www.sports-and-bytes.com
amrecover issues
If I do amrecover on the command line,
# amrecover
AMRECOVER Version 2.4.2p2. Contacting server on mybackup ...
220 mybackup AMANDA index server (2.4.2p2) ready.
200 Access OK
Setting restore date to today (2001-01-01)
200 Working date set to 2001-01-01.
200 Working date set to 2001-01-01.
amrecover: Unexpected server end of file
#
Amanda complains of amrecover: Unexpected server end of file My amanda tape server
is mybackup server 10.1.1.10. I want to restore a file from its /var backup. If I do
this
# amrecover -C Daily
AMRECOVER Version 2.4.2p2. Contacting server on
mybackup ...
220 mybackup AMANDA index server (2.4.2p2) ready.
200 Access OK
Setting restore date to today (2001-01-01)
200 Working date set to 2001-01-01.
200 Config set to Daily.
501 No index records for host: mybackup. Invalid?
Trying mybackup ...
501 No index records for host: mybackup. Invalid?
Trying localhost ...
501 No index records for host: localhost. Invalid?
amrecover sethost 10.1.1.10
200 Dump host set to 10.1.1.10.
amrecover setdisk /var
Scanning /dumps/amanda...
200 Disk set to /var.
No index records for disk for specified date
If date correct, notify system administrator
amrecover history
200- Dump history for config Daily host 10.1.1.10
disk /var
201- 2001-01-01 0 DailySet10 1
200 Dump history for config Daily host 10.1.1.10
disk /var
amrecover
I know the dumps are working for /var. I get a successful backup email and amadmin
works too except for amrecover. Weird huh? My /etc/hosts file is correct. I have index
generation enabled too and there are files created in my index dir.
I also know daemons and hosts file are ok
# more /etc/hosts
127.0.0.1 mybackup localhost
10.1.1.10 mybackup.test.com mybackup
#more /home/amanda/.amandahosts
mybackup amanda
mybackup root
mybackup.test.com amanda
mybackup.test.com root
10.1.1.10 amanda
10.1.1.10 root
Thourgh the includedir /etc/xinetd.d directive, the 3 services also work
::
amanda
::
service amanda
{
socket_type= dgram
protocol = udp
wait = yes
user =amanda
group =disk
server =/usr/local/libexec/amandad
disable=no
}
::
amandaidx
::
service amandaidx
{
socket_type = stream
protocol= tcp
wait= no
user=root
groups =yes
server =/usr/local/libexec/amindexd
}
::
amidxtape
::
service amidxtape
{
socket_type = stream
protocol= tcp
wait= no
user= root
groups = yes
server =
/usr/local/libexec/amidxtaped
server_args = amidxtaped
}
The xinetd /var/log/messages confirm that xinet serves
the four processes just fine
mybackup xinetd[11344]: xinetd Version 2.1
mybackup xinetd[11344]: libwrap
mybackup xinetd[11344]: options compiled in.
mybackup xinetd[11344]: Started working: 3 available services
mybackup xinetd: xinetd startup succeeded
I also have the correct entries in /etc/services
amanda 10080/udp #amanda backup services
kamanda 10081/tcp #amanda backup services(Kerberos)
kamanda 10081/udp #amanda backup services (Kerberos)
amandaidx 10082/tcp #amanda backup services
amidxtape 10083/tcp #amanda backup services
My /etc/hosts.allow has all:all
Here is my amanda.conf and my disklist
-
define dumptype global {
index yes
}
define dumptype myvartest {
comment /var directory
compress client best
holdingdisk yes
index yes
}
As for the disklist,
#more disklist
10.1.1.10 /var myvartest
The backups work, but restores using amrecover don't. What could be wrong?
Thanks.
Free, secure Web-based email, now OpenPGP compliant - www.hushmail.com
RE: trouble backing up with tar
For some strange reason that doesn't work. The backup always fails with: sendsize: getting size via dump for /dev/vx/dsk/tmp_perforce/vol01 level 0 sendsize: running /usr/sbin/ufsdump 0sf 1048576 - /dev/vx/rdsk/tmp_perforce/vol01 running /usr/local/libexec/killpgrp-2.4.1p1 DUMP: Warning - super-block on device `/dev/vx/rdsk/tmp_perforce/vol01' is corrupt - run fsck DUMP: The ENTIRE dump is aborted. Can anybody tell me why ufsdump is being used instead of vxdump? I built the amanda client on the machine with the disk and configure found vxdump and vxrestore. Regards Silke Jolet, John John.Jolet@mTo: 'Silke Mueller' [EMAIL PROTECTED] edic.comcc: Subject: RE: trouble backing up with tar 03.08.01 14:53 use dump. it runs dump on the 2.6 machine, then pipes the stuff through to the amanda server. I'm backing up aix boxes using aix's backup to a linux amanda server. the dump program is local to the client. -Original Message- From: Silke Mueller [mailto:[EMAIL PROTECTED]] Sent: Friday, August 03, 2001 5:48 AM To: [EMAIL PROTECTED] Subject: trouble backing up with tar Hi, I want to back up a 40 GB disk with gnutar and a DLT 7000. There are approximately 1.300.000 files on the disk and the dump exits with the error message driver: result time 8235.753 from dumper0: FAILED 01-2 [data timeout]. There are no other errors. Btw. the etimeout is 15000. I don't use dump, because the disk is vxfs on a Solaris 2.6 machine and my backup server is a Solaris7 machine without any vxfs. Any ideas how to solve this? Thanks Silke
Re: amanda + tape changer + solaris 8
Thomas Hepper [EMAIL PROTECTED] wrote: Hi, On Fri, Aug 03, 2001 at 09:48:08AM +0200, thomas graichen wrote: John R. Jackson [EMAIL PROTECTED] wrote: the backtrace looks like it tries to access the labelfile somehow - but i did not define one in chg-scsi.conf #labelfile /usr/local/amanda/etc.local/amanda/fptsys/labelfile OK, what happens if you define it, may be you hit a bug :-), no you hit a bug, will fix it. In the meantime try to set labelfile. didn't help (at least with an empty labelfile it still coredumps) - but i got chg-scsi to do what i wanted it to do without crashing by commenting out some calls to MapBarCode (which i think you are working on fixing so that it plays nice without a labelfile :-) can you please tell me then you have fixed it so that i may move towards clean code? thanks for all so far and in advance for the stuff coming t -- thomas graichen [EMAIL PROTECTED] ... perfection is reached, not when there is no longer anything to add, but when there is no longer anything to take away. --- antoine de saint-exupery
amrecover issue
Greetings. I just got amanda up and backing things up fine. The next step is obviously to test restores, and I've come up with some questions/concerns in this area. first, I have NO tape drive yet - I'm trying to pry some dollars out of managements' hands for this, but no luck so far - so all backups are being done to holding disks only (RAID 0+1). anyway, I wanted to restore a file from yesterdays backup as a test. from log.20010805.0: SUCCESS dumper utl-atl-06 /export/home 20010805 0 [sec 59.666 kb 393280 kps 6591.3 orig-kb 393260] so a full backup was done successfully. I went ahead and verified the file in question was in the index file: [siteops@utl-atl-06 _export_home]$ zgrep sysedge.cf * 20010731_0.gz:07330357423/./siteops/sysedge.cf 20010805_0.gz:07331665340/./siteops/sysedge.cf ok, looks good to me! I su over to root and go into amrecover inside the /export/home directory [root@utl-atl-06 home]# amrecover Daily AMRECOVER Version 2.4.2p2. Contacting server on utl-atl-06 ... 220 utl-atl-06 AMANDA index server (2.4.2p2) ready. 200 Access OK Setting restore date to today (2001-08-06) 200 Working date set to 2001-08-06. 200 Config set to Daily. 200 Dump host set to utl-atl-06. $CWD '/export/home' is on disk '/export/home' mounted at '/export/home'. 200 Disk set to /export/home. /export/home amrecover history 200- Dump history for config Daily host utl-atl-06 disk /export/home 201- 2001-08-05 0 /backup/hd2/20010805/utl-atl-06._export_home.0 0 201- 2001-08-04 1 /backup/hd2/20010804/utl-atl-06._export_home.1 0 201- 2001-08-03 1 /backup/hd1/20010803/utl-atl-06._export_home.1 0 201- 2001-08-02 1 /backup/hd3/20010802/utl-atl-06._export_home.1 0 201- 2001-08-01 1 /backup/hd2/20010801/utl-atl-06._export_home.1 0 201- 2001-07-31 0 /backup/hd2/20010731/utl-atl-06._export_home.0 0 200 Dump history for config Daily host utl-atl-06 disk /export/home amrecover setdate ---05 200 Working date set to 2001-08-05. amrecover ls 2001-08-05 app/ amrecover hmm.. not good - here is a directory listing for /export/home: [root@utl-atl-06 home]# ls -la total 26 drwxr-xr-x 6 root root 512 Dec 13 2000 . drwxrwxr-x 3 root sys 512 Dec 11 2000 .. drwxr-xr-x 5 root other512 Dec 15 2000 app drwxr-xr-x 8 aribaariba512 Dec 27 2000 ariba drwx-- 2 root root8192 Dec 11 2000 lost+found drwxr-xr-x 6 siteops siteops 512 Aug 6 12:19 siteops so we are missing a couple of directories in the interactive part of amrecover. Being semi-concerned at the integrity of the backups, I decided to try a manual restore, and so extracted the gzip-tar image from the holding disk with dd. Then unzipped it, and verified that the file was there: [root@utl-atl-06 20010805]# dd if=utl-atl-06._export_home.0 bs=32k skip=1 of=/tmp/out 12290+0 records in 12290+0 records out [siteops@utl-atl-06 /tmp]$ tar tvf out | grep sysedge.cf -rw--- siteops/siteops 2678 2001-04-20 16:41 07331665340/./siteops/sysedge.cf extracting the file and diff'ing the original confirms that it is ok. any suggestions? is this a by-product of only using holding disks, or something else? (note that I got similar results for all the filesystems I backed-up - some directories would appear in amrecover, some would not. Thanks in advance! Matt +- Matthew Galer [EMAIL PROTECTED] 770-453-9001 x149
Re: Backing up filesystem to large for one tape
Kris Boulez [EMAIL PROTECTED] writes: What do people think about my proposed solution to this problem Problem --- Filesysteem needs to be backed up using amanda. Level 0 doesn't fit on a single tape anymore (after compression) Solutions - a) By a tape unit which can hold more data (expensive) b) Wait for multi-tape support in amanda (ETA ?) c) Split the filesystem on the server d) let amanda believe that the filesystem is split - create two dumptype, each with a seperate exclude list - add two entries to the disklist with seperate dump type (will amanda backup the same partition twice ?) - (if the above step doesn't work) create a new /dev/rdsk/.. entry pointing to the same device but with a different name Do you think d) can be made to work ? It can. Works ok here. Takes a little work before the backup is run, bus is fairly straightforward, especially if you use a version of Amanda, which let's you define dumptypes in the disklist (2.4.2?). Also requires you to use tar since Amanda doesn't support excluding with dump (even if the dump program supports it). Kris, /Jens -- Jens Bech Madsen The Stibo Group, Denmark
getting an estimate
Is there a good way to run just the estimating part of amanda, so that I can get an idea of the size that a dump is going to be, without actually running amdump? The one idea that I've had is to run amdump with a tapedev of null: (I'm using the tapeio branch), but that would still force my hosts to send backups over the network, just so I can get an estimate. There's got to be a better way. Ideas? Dan -- The more advertising I see, the less I want to buy. Dan Debertin [EMAIL PROTECTED] www.nodewarrior.org
