Results missing
Hi I'm having problems with my amanda installation. It has run without any problems for quite a while, but suddenly it gives the RESULTS MISSING error on alle servers in the backup-report. The log.* from the lates run is as follows: --- SNIP --- START planner date 20020530 START driver date 20020530 INFO planner Adding new disk firewall.eos:/etc. INFO planner Adding new disk firewall.eos:/afs/eos.dk/admin/hostdb. INFO planner Adding new disk afs1.eos:/etc. INFO planner Adding new disk afs1.eos:/var. INFO planner Adding new disk afs1.eos:/service. START taper datestamp 20020530 label EOS-005 tape 0 INFO planner Incremental of lemon.eos.dk:/home bumped to level 2. INFO planner Incremental of lemon.eos.dk:/ bumped to level 2. INFO planner Incremental of tinus.eos.dk:/home bumped to level 2. INFO planner Incremental of afs1.eos:/data02/cvsroot bumped to level 2. FINISH planner date 20020530 FATAL driver schedule line 4: syntax error INFO taper tape EOS-005 kb 0 fm 0 [OK] --- SNIP --- And the corresponding amdump file (only the part with errors is included): --- SNIP --- GENERATING SCHEDULE: afs1.eos /var 11839 0 1970:1:1:0:0:0 11525 384 afs1.eos /etc 11839 0 1970:1:1:0:0:0 2440 81 afs1.eos /service 11839 0 1970:1:1:0:0:0 5 0 firewall.eos /afs/eos.dk/admin/hostdb 11839 0 1970:1:1:0:0:0 0 0 6778732 firewall.eos /etc 11839 0 1970:1:1:0:0:0 0 0 6778732 svabonius.eos.dk / 206 0 1970:1:1:0:0:0 432210 382 1 2001:10:28:0:3:56 43620 263 firewall.eos /d106 7 4 2002:5:22:23:31:45 1 0 firewall.eos / 6 0 1970:1:1:0:0:0 537151 1823 2 2002:5:20:23:36:32 13831 70 afs1.eos /data01 3 0 1970:1:1:0:0:0 4339878 7708 2 2002:5:25:23:39:13 1043 6 firewall.eos /c0t1d0s7 3 0 1970:1:1:0:0:0 354601 1632 2 2002:5:22:0:31:46 19602 161 maconomyserver.eos.dk / 3 0 1970:1:1:0:0:0 288089 896 1 2002:5:18:23:11:52 14811 100 lemon.eos.dk / 3 0 1970:1:1:0:0:0 116063 186 2 2002:5:27:23:40:20 3539 9 afs1.eos //adm/backup 3 0 1970:1:1:0:0:0 45806 469 1 2002:5:18:23:37:49 82011 2562 maconomyserver.eos.dk /var 3 0 1970:1:1:0:0:0 1555 18 1 2002:5:18:23:11:31 1318 36 svabonius.eos.dk /etc 3 0 1970:1:1:0:0:0 1473 6 1 2002:5:18:23:11:46 462 2 firewall.eos /ext 2 1 2002:5:21:23:11:30 2826747 11977 firewall.eos /var 2 1 2002:5:22:0:34:52 132735 804 tinus.eos.dk /home 2 2 2002:5:27:23:32:26 124405 389 afs1.eos /data02/eos 2 1 2002:5:28:0:24:5 69644 829 afs1.eos /data02/cvsroot 2 2 2002:5:27:23:37:16 63666 535 svabonius.eos.dk /home 2 1 2002:5:28:6:54:56 48871 125 lemon.eos.dk /home 2 2 2002:5:27:23:32:24 2209 7 tinus.eos.dk / 2 1 2002:5:27:23:45:17 2023 21 svabonius.eos.dk /usr/bb18d1 2 1 2002:5:27:23:32:24 31 0 firewall.eos /pack/eoswww 2 1 2002:5:22:23:11:30 1 0 proxy.eos / 1 0 1970:1:1:0:0:0 259941 1355 1 2002:5:18:23:11:39 962 99 driver: schedule line 4: syntax error driver: adding holding disk 0 dir /amanda-cache/eos size 33306364 reserving 33306364 out of 33306364 for degraded-mode dumps taper: DONE [idle wait: 1110.422 secs] taper: writing end marker. [EOS-005 OK kb 0 fm 0] amdump: end at Thu May 30 01:18:39 CEST 2002 --- SNIP --- It's quite obvious that the driver is upset about the results the planner generated, but what could have coursed the error? Kind regards -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Any automagic exclusion of filesystems?
Hi I'm changing the dump-method on some of my filesystems from ufsdump to gtar (so I can restore these filesystems on non-Sun hardware/OS). I used to backup via the mount-point, and that worked ok, but will I get double amount of data now? Hmm... how the do I explain this... :) I have the following filesystems on the box: / /data /usr /var . . In the disklist i have /, /data, /usr and /var defined as seperat disks, using gtar as dump-method. So the question is, will the backup of / also include /data, /usr and /var, or will amanda exclude these by default? Hope it makes sence :) Kind regards... -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Re: The skinny on indexing
On Wed, 2002-05-22 at 18:57, Kirk Strauser wrote: I've been using Amanda for a while, but I'm not terribly clear on the details of indexing. My impression to date is that indexing is required if you which to use amrecover, but is not if you're willing to hand-restore filesystems with a dump or tar image from a tape. Not 100% correct as I see it. If you're using index, you'll be able to browse through your indexes and select individual files for restore. If you not use indexes you will only be able to restore whole filesystems with amrestore, or use the native tools such as tar or dump, combined with mt. Why would you *not* want indexing enabled? To save diskspace for indexing on filesystems containing your OS or similar (/usr and such). Why *would* you want indexing enabled? To be able to restore individual files with amrecover. Note: I'll be perfectly happy for an RTFM answer if it also includes a pointer to documentation that concisely answers my questions. :) I can recomend John and Alexandres chapter about amanda from Unix Backup Recovery (can be found on http://www.backupcentral.com/amanda.html). It's an execelent introduction to amanda. Now that I'm at it, the whole book is recommendable. The man-pages is also worth a read. Kind regards, Jens Rohde
Re: The skinny on indexing
On Wed, 2002-05-22 at 22:26, Kirk Strauser wrote: OK. So you *can* still use amrecover, just not in the FTP-like mode. No it's actually amrestore. amrecover is using amrestore to manage the restore after selection of files. Kind regards, Jens Rohde
Re: Question in case of disaster
On Tue, 2002-05-07 at 10:18, Radu Filip wrote: Hi I'm not in a disaster situation, fortunatelly, but I would like to test how to proceed, just in case. Good idea :) (1) dd if=$TAPE bs=32k skip=1 of=/tape_content produce what in /tape_content? An ISO image that I can mount it with mount -o loop? Depending on which program you used for the backup you will get a tar or a dump-archive. So extract with suitable program. (2) What is mentioned ufsrestore program? I tried to find what RPM (I ran RedHat) package belongs to but I did not find anything. ufsdump is Solaris' version of dump. Viceverse ufsrestore is the Solaris version of... restore :) Kind regards... -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Re: LTO Ultrium Unit
On Fri, 2002-04-19 at 21:18, Marcos Teodoro Dias Junior wrote: Some one here have been used this tape unit with Amanda using Linux as S.O? I'm using the IBM equivalent on a Sun-server, and it works just fine. I've added a tapetype entry to the FAQ-O-matic. Kind regards... -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Problems with amrecover
Hi I'm stuck! - I've tried to get amrecover to work. There's no problem in selecting host, disk and files to recover, but when I try to do the actual recover I get the following error: amrecover extract Extracting files using tape drive /dev/nst1 on host devel. The following tapes are needed: EOS-011 EOS-012 Restoring files into directory /tmp Continue? [Y/n]: y Load tape EOS-011 now Continue? [Y/n]: y EOF, check amidxtaped.debug file on devel. amrecover: short block 0 bytes UNKNOWN file amrecover: Can't read file header extract_list - child returned non-zero status: 1 Continue? [Y/n]: The debug file contains: amidxtaped: debug 1 pid 3112 ruid 1028 euid 1028 start time Thu Oct 18 11:17:09 2001 amidxtaped: version 2.4.2 SECURITY USER root bsd security: remote host afs1.eos user root local user amanda amandahosts security check passed 6 amrestore_nargs=6 -h -p /dev/nst1 svabonius.eos.dk ^/etc$ 20011017 Ready to execv amrestore with: path = /afs/eos.dk/sw/pack/amanda-2.4.2/sbin/amrestore argv[0] = amrestore argv[1] = -h argv[2] = -p argv[3] = /dev/nst1 argv[4] = svabonius.eos.dk argv[5] = ^/etc$ argv[6] = 20011017 amrestore: could not open tape /dev/nst1: Permission denied amidxtaped: amrestore terminated normally with status: 2 Rewinding tape: no tape online amidxtaped: pid 3112 finish time Thu Oct 18 11:17:09 2001 I've searched the mail-list archives but didn't find anything that applied to my problem (or at least I don't think so :) ). I have no problems with restoring files using mt and dd on the same tape/tape drive, so it is functioning. The server is running RedHat Linux 7.0, and I've tried both amanda 2.4.2 and 2.4.2p2. Same result. So what do I do wrong here? -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Probs with amrecover
Hi I'm trying to recover som data from a backup, but it does't really work. Here's a cut of the session: --- CUT --- AMRECOVER Version 2.4.2. Contacting server on afs1.eos.dk ... 220 afs1 AMANDA index server (2.4.2) ready. 200 Access OK Setting restore date to today (2001-07-13) 200 Working date set to 2001-07-13. 200 Config set to eos. 501 No index records for host: afs1. Invalid? Trying afs1 ... 501 No index records for host: afs1. Invalid? Trying localhost.localdomain ... 501 No index records for host: localhost.localdomain. Invalid? Trying localhost ... 501 No index records for host: localhost. Invalid? amrecover sethost afs1.eos 200 Dump host set to afs1.eos. amrecover setdisk //lemon/IntraToolsFiles Scanning /amanda-cache/eos... 20010410: found Amanda directory. 20010411: found Amanda directory. 20010412: found Amanda directory. 20010413: found Amanda directory. 20010415: found Amanda directory. 20010417: found Amanda directory. 20010513: found Amanda directory. 20010525: found Amanda directory. 20010605: found Amanda directory. 20010606: found Amanda directory. 20010608: found Amanda directory. 200 Disk set to //lemon/IntraToolsFiles. amrecover ls 2001-07-13 IntraToolsBackup.zip amrecover add IntraToolsBackup.zip Added /IntraToolsBackup.zip amrecover lcd /tmp amrecover extract Extracting files using tape drive /dev/nst0 on host devel. The following tapes are needed: EOS-008 Restoring files into directory /tmp Continue? [Y/n]: y Load tape EOS-008 now Continue? [Y/n]: y Error connecting to tape server: Invalid argument --- CUT --- both the amanda user and root are allowed to connect to the amandasystem, with all variants of the machines hostname. Any idears? - somebody really need there stuff :) Kind regards... -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Re: Probs with amrecover
On Fri, 13 Jul 2001, John R. Jackson wrote: Hi John Then first things first. If amrecover is giving you grief, and you really, really need that file back now, don't use it. Amrecover was able to tell you the right tape, so mount it by hand and do the amrestore piped into smbclient (or whatever) by hand. Thanks. That worked (dd and mt rules) :) - I'll take a look at amrecover later. The client and the server are on the same machine, so there should not be any network-problems :) I can really recomend the book on http://www.backupcentral.com/amanda.html. They have a really good chapter about recovering without amrecover. But the guys have there data, and I am at home, so everyone is happy :) Thanks... -- Jens Rohde, System Administrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Aarhus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Strange warning i logfile
Hi I've solved my problem with strange danish characters. I'm archiving the files in a zip-archive, so I can actually make the backup now :) But I get this strange warning in the logfile: /-- afs1.eos //lemon/IntraToolsFiles lev 0 STRANGE sendbackup: start [afs1.eos://lemon/IntraToolsFiles level 0] sendbackup: info BACKUP=/usr/bin/smbclient sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc |/usr/bin/smbclient -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end ? SSL: Error error setting CA cert locations: error:::lib(0) :func(0) :reason(0) ? trying default locations. | added interface ip=10.0.0.5 bcast=10.255.255.255 nmask=255.0.0.0 | tar: dumped 1 files and directories | Total bytes written: 39588352 sendbackup: size 38661 sendbackup: end \ What does it meen, and can I remove it somehow? -- Jens Rohde, Systemsadministrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Århus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788
Probs with danish characters on NT machine
Hi I'm having a problem backing up files on a NT-machine, using smbmount. A cut from the log: --- CUT --- /-- afs1.eos //lemon/IntraToolsFiles lev 0 STRANGE sendbackup: start [afs1.eos://lemon/IntraToolsFiles level 0] sendbackup: info BACKUP=/usr/bin/smbclient sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc |/usr/bin/smbclient -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end ? SSL: Error error setting CA cert locations: error:::lib(0) :func(0) :reason(0) ? trying default locations. | added interface ip=10.0.0.5 bcast=10.255.255.255 nmask=255.0.0.0 ? ERRDOS - ERRbadfile opening remote file \subproject\82\De_5_Losninger.sda (\subproject\82\) ? ERRDOS - ERRbadfile opening remote file \subproject\87\ref_opstartsmode.sdw (\subproject\87\) | tar: dumped 153 files and directories | Total bytes written: 57367552 sendbackup: size 56023 sendbackup: end \ --- CUT --- The two files are both using the special danish character 'ø' (oe) in the filename. Same error occur on an other NT-machine. Any surgestions? Kind regards, -- Jens Rohde, Systemsadministrator --- Eastfork Object Space (EOS), Margrethepladsen 3, 8000 Århus C, Denmark Tel: +45 8732 8787 / Fax: +45 8732 8788