amdump creates no backup

Printable View

Show 40 post(s) from this thread on one page

May 24th, 2011, 03:06 AM
bv_chris

amdump creates no backup

Hello again dear Community.

Maybe with this problem someone can help.

I got my 2 virtual linux machines today, so my task was to set up amanda-client and -server.

I did this and after quiet some time i was abel to amcheck and amdump.

But when i try to amrecover on the client system it says No Index specified blabla...

In the amanda.conf the dumptype i use is "test" with index YES
My /test directory on amanda-client contains 15files with various size.

when i run amdump he does something. the logfile looks fine:

Code:

DISK planner amanda-client /test START planner date 20110524 START driver date 20110524 STATS driver startup time 0.022 FINISH planner date 20110524 time 0.078 START taper datestamp 20110524 label testbu-1 tape 0 SUCCESS dumper amanda-client /test 20110524 1 [sec 0.018 kb 3 kps 165.1 orig-kb 210] SUCCESS chunker amanda-client /test 20110524 1 [sec 0.046 kb 3 kps 751.8] SUCCESS taper amanda-client /test 20110524 1 [sec 0.001 kb 64 kps 45779.7 {wr: writers 2 rdwait 0.000 wrwait 0.000 filemark 0.000}] INFO taper tape testbu-1 kb 64 fm 1 [OK] FINISH driver date 20110524 time 0.212 ~ ~

a ls -oa on my index dir says:

Code:

amanda@amanda-server:/ $ ls -oa /backups/testbu/index/amanda-client/_test/ total 32 drwxr-sr-x 2 amanda 4096 May 24 11:53 . drwxr-sr-x 3 amanda 4096 May 24 11:32 .. -rw------- 1 amanda 79 May 24 11:32 20110524_0.gz -rw------- 1 amanda 48 May 24 11:53 20110524_1.gz

so there is something. when i try to extract/list the contents of the *.gz files it shows me nothing.

when i try to amrecover on the amanda-client :

Code:

root@amanda-client:/test # /usr/sbin/amrecover -C testbu -s amanda-server AMRECOVER Version 2.5.0p2. Contacting server on amanda-server ... 220 amanda-server AMANDA index server (2.5.0p2) ready. 200 Access OK Setting restore date to today (2011-05-24) 200 Working date set to 2011-05-24. Scanning /backups/amanda... 20110524111152: found Amanda directory. 200 Config set to testbu. 200 Dump host set to amanda-client. Trying disk / ... Trying disk rootfs ... Can't determine disk and mount point from $CWD '/test' amrecover> setdisk /test 200 Disk set to /test. No index records for disk for specified date If date correct, notify system administrator amrecover>

histroy shows:

Code:

amrecover> history 200- Dump history for config "testbu" host "amanda-client" disk "/test" 201- 2011-05-24 0 testbu-1:1 201- 2011-05-24 1 testbu-2:1 201- 2011-05-24 1 testbu-3:1 201- 2011-05-24 1 testbu-4:1 201- 2011-05-24 1 testbu-5:1 201- 2011-05-24 1 testbu-1:1 200 Dump history for config "testbu" host "amanda-client" disk "/test"

so he obviously did something.

anyone got the same problem?

amcheck is ok too:

Code:

amanda@amanda-server:/ $ /usr/sbin/amcheck testbu Amanda Tape Server Host Check ----------------------------- Holding disk /backups/amanda: 709360 KB disk space available, using 606960 KB slot 1: read label `testbu-1', date `20110524' slot 2: read label `testbu-2', date `20110524' NOTE: skipping tape-writable test Tape testbu-2 label ok Server check took 0.114 seconds Amanda Backup Client Hosts Check -------------------------------- Client check: 1 host checked in 0.061 seconds, 0 problems found (brought to you by Amanda 2.5.0p2)
May 24th, 2011, 04:45 AM
bv_chris

i dig a little more:
amdump.* :

Code:

driver: start time 0.022 inparallel 4 bandwidth 2000 diskspace 607268 dir OBSOLETE datestamp 20110524 driver: drain-ends tapeq FIRST big-dumpers sssS dumper: pid 19087 executable dumper3 version 2.5.0p2 planner: time 0.075: got partial result for host amanda-client disk /test: 0 -> 230K, 1 -> -2K, -1 -> -2K planner: time 0.078: got partial result for host amanda-client disk /test: 0 -> 230K, 1 -> 10K, -1 -> -2K changer: got exit: 0 str: 1 5 1 changer_query: changer return was 5 1 changer_query: searchable = 0 changer_find: looking for NULL changer is searchable = 0 changer: opening pipe to: /usr/lib64/amanda/chg-disk -slot current planner: time 0.079: got result for host amanda-client disk /test: 0 -> 230K, 1 -> 10K, -1 -> -2K planner: time 0.079: getting estimates took 0.078 secs FAILED QUEUE: empty DONE QUEUE: 0: amanda-client /test ANALYZING ESTIMATES... pondering amanda-client:/test... next_level0 0 last_level 0 (due for level 0) (picking inclevel for degraded mode) picklev: last night 0, so tonight level 1 curr level 0 size 230 total size 3326 total_lev0 230 balanced-lev0size 115 INITIAL SCHEDULE (size 3326): amanda-client /test pri 2 lev 0 size 230 DELAYING DUMPS IF NEEDED, total_size 3326, tape length 2000 mark 1000 delay: Total size now 3326. PROMOTING DUMPS IF NEEDED, total_lev0 230, balanced_size 115... planner: time 0.079: analysis took 0.000 secs amdump.*

So i think something goes wrong at the bolt line.
But i must admit i have not a clue to what that means.

if more is needed just feel free to ask :D
May 24th, 2011, 05:58 AM
bv_chris

Am i just stupid, that no one posts?
Or is this a widely known bug that no one knows a answer too?

What really pisses me of, is that there are no sender*debug files. any post i read, every guy says look in them. but amanda creates no such file. not in the logdir not anywhere....
May 24th, 2011, 06:21 AM
martineau

run: amadmin testbu find

You will find that only level 1 are available, amrecover can't works if there is no full backup (level 0)

redo a full backup: amadmin testbu force amanda-client; amdump testbu
never overwrite the latest full dump.
May 24th, 2011, 07:03 AM
bv_chris

i did:

Code:

amanda@amanda-server:~ $ /usr/sbin/amadmin testbu find Scanning /backups/amanda... date host disk lv tape or file file part status 2011-05-24 amanda-client /etc 0 testbu-2 1 -- OK 2011-05-24 amanda-client /etc 0 testbu-4 1 -- OK 2011-05-24 amanda-client /etc 0 testbu-1 1 -- OK 2011-05-24 amanda-client /test 0 testbu-1 1 -- OK 2011-05-24 amanda-client /test 0 testbu-2 1 -- OK 2011-05-24 amanda-client /test 0 testbu-3 1 -- OK 2011-05-24 amanda-client /test 0 testbu-4 1 -- OK 2011-05-24 amanda-client /test 0 testbu-5 1 -- OK 2011-05-24 amanda-client /test 0 testbu-1 1 -- OK

aren't all of them lvl 0? i specified my dumptype to be "dumpcycle 0" and strategy NOINC so he must make full backups.

a quick undertanding question:
if i delete the content of tapelist , clear the backups relabel the taps and reset to slot 1. it should be like a fresh start right? or do i need to clear the contents of the dirs curinfo and index + all logs?
May 24th, 2011, 07:15 AM
martineau

The result is different that the history of amrecover.
This show you have three backup on testbu-1 on file 1, this is impossible.

As you do more that one run in a day, you must set 'usetimestamps' to 'yes', otherwise the recovery is not possible with amrecover.
May 24th, 2011, 07:19 AM
bv_chris

Mhh. How do i reset my database? so that it is a fresh one?

just deleting?
May 24th, 2011, 07:23 AM
martineau

relabel all volumes with 'amlabel -f'
May 24th, 2011, 07:31 AM
bv_chris

ok i relabeld the tapes, and forced a lvl0 dump

Code:

amanda@amanda-server:~ $ /usr/sbin/amadmin testbu find Scanning /backups/amanda... date host disk lv tape or file file part status 2011-05-24 amanda-client /test 0 testbu-5

the option "usetimestamps" doesnt work. amanda.conf runs on errors:

Code:

"/etc/amanda/testbu/amanda.conf", line 36: configuration keyword expected "/etc/amanda/testbu/amanda.conf", line 36: end of line expected

i tried
usetimestamps yes
usetimestamps "yes"
usetimestamps true
usetimestamps "true"

doesnt work. i will try a recover now lets see how that works.
but i appreciate your posts. thanks a lot.

i'm out till thursday. i will update my status.
thanks again
May 24th, 2011, 08:34 AM
martineau

usetimestamps is in 2.5.1.

with 2.5.0, you should not do to runs in a day.

Show 40 post(s) from this thread on one page