slow restoring + restoring files causes tar to die

Hi!

don't know if I sent this twice! in that case, pardon me for the inconvenience.

I backup a windows big volume with about 30000 files using, among other, 
smbclient.
so in my disklist there lies 

        amanda.my_domain.com           //principal.my_domain.com/DADES          
       nocomp-user-gnutar


the dumptype definition used is:

        define dumptype nocomp-user-gnutar {
                comment "Principal samba backups"
                program "GNUTAR"
                holdingdisk yes
                index yes
                options compress, index
                priority high
        }


as you can see I DO use compression.the tape type used is:

        define tapetype principalHDD {
                comment "30Gb disk"
                length 37500 mbytes
        }

I have some firewire external HDDs with five 37500 mbytes 'tapes'
when trying to restore the backup I have two problems:

        amrecover> add io3data
        Added dir /io3data at date 2004-11-05
        Added dir /io3data at date 2004-11-01
        amrecover> extract
        amrecover: warning: using /dev/null as the tape device will not work

        Extracting files using tape drive /dev/null on host 
amanda.my_domain.com.
        The following tapes are needed: CINTA04
                                CINTA07

        Restoring files into directory /tmp/
        Continue [?/Y/n]? Y

        Extracting files using tape drive /dev/null on host 
amanda.my_domain.com.
        Load tape CINTA04 now
        Continue [?/Y/n/s/t]? t
        New tape device [?]: file:/mnt/cinta/CINTA04
        Using tape "file:/mnt/cinta/CINTA04" from server amanda.my_domain.com.
        Continue [?/Y/n/s/t]? Y

amanda keeps processing the 27Gb file for about three or four hours and then I 
got this error:

        tar: Unexpected EOF in archive
        tar: Error is not recoverable: exiting now
        extract_list - child returned non-zero status: 2
        Continue [?/Y/n/r]? Extracting files using tape drive 
file:/mnt/cinta/CINTA04 on host amanda.my_domain.com.
        Load tape CINTA07 now
        Continue [?/Y/n/s/t]? EOF, check amidxtaped.<timestamp>.debug file on 
amanda.my_domain.com.
        amrecover: short block 0 bytes
        UNKNOWN file
        amrecover: Can't read file header
        extract_list - child returned non-zero status: 1
        Continue [?/Y/n/r]? amrecover>
        amrecover>


in the amidxtaped debug file I have:


        amidxtaped: debug 1 pid 13118 ruid 87 euid 87: start at Wed Nov 10 
10:15:16 2004
        amidxtaped: version 2.4.4p2
        amidxtaped: time 0.000: > SECURITY USER root
        amidxtaped: time 0.000: bsd security: remote host amanda.my_domain user 
root local user amanda
        amidxtaped: time 0.000: amandahosts security check passed
        amidxtaped: time 0.000: > CONFIG=principal
        amidxtaped: time 0.000: > LABEL=CINTA08
        amidxtaped: time 0.000: > FSF=28
        amidxtaped: time 0.000: > HEADER
        amidxtaped: time 0.000: > DEVICE=file:/mnt/cinta/CINTA08
        amidxtaped: time 0.000: > HOST=^amanda.my_domain$
        amidxtaped: time 0.000: > DISK=^//principal.my_domain/DADES$
        amidxtaped: time 0.000: > DATESTAMP=20041108
        amidxtaped: time 0.000: > END
        amidxtaped: time 0.000: amrestore_nargs=0
        amidxtaped: time 0.000: Ready to execv amrestore with:
        path = /usr/sbin/amrestore
        argv[0] = "amrestore"
        argv[1] = "-p"
        argv[2] = "-h"
        argv[3] = "file:/mnt/cinta/CINTA08"
        argv[4] = "^amanda.my_domain$"
        argv[5] = "^//principal.my_domain/DADES$"
        argv[6] = "20041108"
        amrestore: WARNING: not at start of tape, file numbers will be offset
        amrestore:   0: skipping dns.my_domain.com._etc.20041108.1
        
[..]

        amrestore:  27: restoring 
amanda.my_domain.__principal.my_domain_DADES.20041108.0
        gzip: stdin: invalid compressed data--format violated
        Error 32 (Broken pipe) offset 1387102208+32768, wrote 0
        amrestore: pipe reader has quit in middle of file.
        amidxtaped: time 27004.971: amrestore terminated normally with status: 2
        amidxtaped: time 27004.971: rewinding tape ...
        amidxtaped: time 27004.971: done
        amidxtaped: time 27004.971: pid 13118 finish time Wed Nov 10 17:45:21 
2004




I was wondering if someone already had this problem and how would you solve it.
moreover, I'd like to be able to restore some files a bit faster. any ideas? I 
think the whole 27Gb file is parsed by tar in the restoring process , so 
splitting it in smaller independent chunks will help.

BUT I cannot make, let's say 30 wilcarded entries in my disklist of the style:
        hostname        /path/to/large-volume/[A-B]*            user-tar
        hostname        /path/to/large-volume/[C-D]*            user-tar
        ...
(found in the amanda faq'o'matic 
(http://amanda.sourceforge.net/cgi-bin/fom?_recurse=1&file=5)) because I'm 
using smbclient to backup and these directories are highly dynamic.

any ideas?


TIA


sam


-- 
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm