flexbackup-help — Support/help/usage

[flexbackup-help] afio: "-" [offset 6]: Fatal error: ???

[Jeff B. <je...@je...>]

Hello folks,

I'm using Redhat 9.  Everything was peachy.  flexbackup worked great,
then suddenly I'm getting this error.  Can anyone give me some
guidance.  I need to restore something from these tapes.  I've tried two
out of 7 that we have.

Thanks.  <<<jeff<at>jeffborders>>>.com


[root@stilson archive]# flexbackup -extract
 
flexbackup version 1.2.0 (http://flexbackup.sourceforge.net)
/etc/flexbackup.conf syntax OK
 
|------------------------------------------------------------
| Checking 'buffer' on this machine... Ok
| Checking /bin/sh on this machine... bash2
|------------------------------------------------------------
| Logging output to "flexbackup.extract.200406030946.log"
| Reading from CURRENT TAPE POSITION
|------------------------------------------------------------
At block 0.
|------------------------------------------------------------
| buffer -m 10m -p 75 -s 10k -t -u 100 -i "/dev/nst0" | afio -i -z -x -D
\
|   /usr/bin/flexbackup -P gzip -Q -d -Q -q -Z -v   -b 10k -
|------------------------------------------------------------
afio: "-" [offset 6]: Fatal error:
afio: "-": Unrecognizable archive
Kilobytes Out 10
|------------------------------------------------------------
At block 2.
|------------------------------------------------------------
[root@stilson archive]#

Re: [flexbackup-help] adding titles to zip archives

[Charlie B. <cha...@e-...>]

On Tue, 1 Jun 2004, Rohan Carly wrote:

> I have been examining how flexbackup attempts to attach a title to zip
> archives. (I'm talking about zip, not gzip).
...
> This is a bad solution for two reasons:
> 
> 1. It requires write access to the volume being backed up.
> 2. Some filesystems don't allow filenames to contain colons. (Windows
> filesystems shared over SMB is one example).
> 
> Could you please modify flexbackup to apply titles by simply running "zip"
> again with the "-z" option after it's done the compression.
> 
> BTW, thanks for your fine program.

A fine way for you to provide thanks, and get just what you want, would be 
for you to submit a patch. :-)

--
Charlie

[flexbackup-help] adding titles to zip archives

[Rohan C. <se...@ro...>]

I have been examining how flexbackup attempts to attach a title to zip
archives. (I'm talking about zip, not gzip).

Flexbackup relies on the -@ switch to supply a list of files to zip. From
the zip man page:
> -@     Take  the  list of input files from standard
>             input. Only one filename per line.

This rules out being able to use -z.
>  -z     Prompt for  a  multi-line  comment  for  the
>              entire zip archive.  The comment is ended by
>              a line containing just a period, or  an  end
>              of  file condition (^D on UNIX, ^Z on MSDOS,
>              OS/2, and  VAX/VMS).   The  comment  can  be
>              taken from a file:
>
>                    zip -z foo < foowhat

To work around this, flexbackup creates an empty file using "touch". From
the backup_zip() sub in the flexbackup source:
>    # Kludge a title by replacing / with - in the title
>    # then touch a file in the dir we are going to back up.
>    $title =~ s%/%-%g;
>    $title =~ s% %_%g;
>    $cmd = "$main::path{touch} \"$dir/$title\"";
>    push(@cmds, &maybe_remote_cmd($cmd, $remote));

This is a bad solution for two reasons:

1. It requires write access to the volume being backed up.
2. Some filesystems don't allow filenames to contain colons. (Windows
filesystems shared over SMB is one example).

Could you please modify flexbackup to apply titles by simply running "zip"
again with the "-z" option after it's done the compression.

BTW, thanks for your fine program.

__
Rohan Carly

[flexbackup-help] Directory ownership not saved

[Rohan C. <se...@ro...>]

I've found that flexbackup does not correctly preserve 
the UID/GID of directories, despite the use of "tar --
same-owner". This seems to be due to the interaction 
between "find" and "tar".

The problem can be fixed by dropping the "-depth" 
argument to find, so that the directories appear in the 
file list ahead of the contents of the directories.

I was able to fix the problem by commenting out the 
following line in the file_list_cmd() subroutine:

$cmd .= "-depth ";

I use flexbackup v1.2.1, GNU find 4.1.7 and GNU tar 
1.13.25.


P.S. I submitted this sometime ago on flexbackup's "bugs" section on
Sourceforge but I'm not sure that anyone uses that.

__
Rohan Carly

[flexbackup-help] remote directory

[patoch <pl...@ir...>]

Hi
I'm trying to use  flexbackup 1.2.1-2. It's seem to be a very powerfull 
software.
Is it possible that $device should be  a remote directory?
It's working with a remote tape drive but when I am trying to backup on 
a remote directory, I've got this message:
 Can't write to <remote host:/path/directory

Thanks for your help

Patoch

Re: [flexbackup-help] Easy way of checking a backup

[Dirk H. <dh...@ca...>]

Hi,

Andrea Francesconi schrieb:

> Jeff MacDonald wrote:
>
>> if you put it in cron, you should get an email every time it runs with=
 a
>> full log output.
>
>
> For big backups, cron mail isn't very light (1,2MB of backup log).

If it was that small! A test backup of our /home partition resulted in=20
11 MB of email - and some spurious errors I could not explain to myself..=
=2E

> From the Flexbackup ToDo list from site:
>
> "A shorter summary status that's better than doing
> egrep "(error|of set|Backup of|written)"
> of logfile.  Needs to handle all configurations...."

Hm. I'll have to reconsider the trade-off between flexbackup=20
(convenient, clever, but no reliable backup verify) against pure afio=20
(to do file searching stuff on my own, fiddle with the command line, but =

being able to guarantee that the backup tape is at least readable).

Is there someone how found a procedure to _guarantee_ the physical=20
health of a backup tape by using flexbackup after the backup is done. I=20
thought of something like

mt bsf 1 ; afio -t ... /dev/ntape ; [ $? =3D "0" ] && echo "success" ||=20
echo "failure"

If one does not rewind the tape, this should create a file list and read =

the complete latest backup from the tape for this. Any physical read=20
errors would be recognized.

Comments?

Best regards,
Dirk

--=20
--- Dirk Hillbrecht, cantamen GmbH --- dh...@ca...
--- Hausmannstra=DFe 9-10, 30159 Hannover, http://www.cantamen.de
--- Tel.: +49/511/5902626-0, Fax: +49/511/5902626-4

Re: [flexbackup-help] Easy way of checking a backup

[Andrea F. <afr...@si...>]

Jeff MacDonald wrote:

> if you put it in cron, you should get an email every time it runs with a
> full log output.

For big backups, cron mail isn't very light (1,2MB of backup log).
 From the Flexbackup ToDo list from site:

"A shorter summary status that's better than doing
egrep "(error|of set|Backup of|written)"
of logfile.  Needs to handle all configurations...."


fRANz

Re: [flexbackup-help] Easy way of checking a backup

[Jeff M. <in...@bi...>]

if you put it in cron, you should get an email every time it runs with a
full log output.

jeff.


On Wed, 2004-05-12 at 11:34, Dirk Hillbrecht wrote:
> Hello,
> 
> flexbackup seems to be exactly the tool I was searching for for quite a 
> while now. I will almost certainly use it for our daily backup 
> procedures here in the company.
> 
> There is only one thing I am not aware of so far: How does one ensure 
> that the backup succeeded, esp. regarding probable physical errors. I 
> thought that flexbackup does this automatically after each backup (as it 
> should be...), but it seems to do not. How can this gap be closed? You 
> know: There is nothing worse than a write-only backup...
> 
> Best regards,
> Dirk
-- 
Bored @ Work ?
http://www.notforwork.com

[flexbackup-help] Easy way of checking a backup

[Dirk H. <dh...@ca...>]

Hello,

flexbackup seems to be exactly the tool I was searching for for quite a=20
while now. I will almost certainly use it for our daily backup=20
procedures here in the company.

There is only one thing I am not aware of so far: How does one ensure=20
that the backup succeeded, esp. regarding probable physical errors. I=20
thought that flexbackup does this automatically after each backup (as it =

should be...), but it seems to do not. How can this gap be closed? You=20
know: There is nothing worse than a write-only backup...

Best regards,
Dirk

--=20
--- Dirk Hillbrecht, cantamen GmbH --- dh...@ca...
--- Hausmannstra=DFe 9-10, 30159 Hannover, http://www.cantamen.de
--- Tel.: +49/511/5902626-0, Fax: +49/511/5902626-4

[flexbackup-help] Multi-volume Restore...?

[vasudeva <vas...@me...>]

Recently I decided to try out the multi-volume support with mbuffer.
The backup went fine -- five separate directories across two
volumes.

When I went to restore, though, I got problems.  I can restore from
any backup file except the one that spans across the tapes.  In my
case, this is file number 5.  I've included the command I'm using
and the output, as well as attached my flexbackup.conf.



***
(root@lenin)    (/home/vasudeva)        (09:04:06)
flexbackup -volumes 2 -num 5 -extract -onefile db/pkg/x11-plugins/gkrellm-reminder-2.0.0/CC

flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
/etc/flexbackup.conf syntax OK

path afio = /home/vasudeva/temp/afio/afio-2.4.8beta1/afio
|------------------------------------------------------------
| Checking 'mbuffer' on this machine... Ok
| Checking /bin/sh on this machine... bash2
|------------------------------------------------------------
| Logging output to "flexbackup.extract.200403300904.log"
| Positioning tape at file number 5
/dev/nst0: Input/output error
|------------------------------------------------------------
At block 349629.
|------------------------------------------------------------
| Extracting single file
db/pkg/x11-plugins/gkrellm-reminder-2.0.0/CC
| mbuffer -q -m 10M -p 75 -s 10240 -f -n 2 -i "/dev/nst0" | \
|   /home/vasudeva/temp/afio/afio-2.4.8beta1/afio -i -w
/tmp/extract.11376 \
|   -z -x -D /usr/bin/flexbackup -P gzip -Q -d -Q -q -Z -v   -b 10k
-
|------------------------------------------------------------
error: inputThread: error reading: Input/output error
***



That's where it will hang, for as long as I let it.  It doesn't
matter which of the two tapes tape I try this command on -- it fails
both ways.



/></

--
linkswarm.com :: Collaborative Insolence
megarad.com :: Killing You With Technology
vasudeva.megarad.com :: For The Faint of Heart

Re: [flexbackup-help] serious error : flex is core dumping

[Jeff M. <in...@bi...>]

Actually I'm running it on FreeBSD 5

Jeff.

On Tue, 2004-04-06 at 13:02, Corey Mann wrote:
> I'm gonna guess your running Red Hat 9... the perl and perl-NDBM_File 
> packages conflict (I just tried updating it (with yum) yesterday and 
> came across it by chance, I don't have a solution to it, just pointing 
> you in the right direction)
> 
> 
> --Corey
> 
> Jeff MacDonald wrote:
> > On a whim, i took my logs,index,data and timestamps and moved them out
> > of the way. Bingo it started working again.
> > 
> > So I figure I'll wait 6 more weeks see if it dies again, if it does I'll
> > either mail this list again, or try amanda.
> > 
> > Anyone have any suggestions, or is anyone interested at all in looking
> > deeper into this ?
> > 
> > Jeff.
> > 
> > On Tue, 2004-04-06 at 09:09, Jeff MacDonald wrote:
> > 
> >>Hi,
> >>
> >>Subject says it all, I've been running flex fine on a machine for about
> >>6 weeks, then suddenly in my cron emails it tells me that it's core
> >>dumping. It core's whether I dump a local or remote file system etc..
> >>
> >>-su-2.05b$ flexbackup -set gollum -level 3
> >>
> >>flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
> >>/etc/flexbackup.conf syntax OK
> >>
> >>|------------------------------------------------------------
> >>| Checking /bin/sh on this machine... unknown (probably Bourne Shell)
> >>|------------------------------------------------------------
> >>| Doing level 3 backup of set gollum using tar
> >>| Found directory index key 200402121142
> >>| Backup set "gollum" (/usr/home /etc/ /root/ /usr/local/etc/)
> >>|------------------------------------------------------------
> >>Segmentation fault (core dumped)
> >>
> >>Here's what gdb says [tho i have no clue what it means.
> >>
> >>#0  0x281e447a in memmove () from /usr/lib/libc.so.5
> >>#1  0x00000039 in ?? ()
> >>#2  0x281d43b2 in __split_page () from /usr/lib/libc.so.5
> >>#3  0x281d77a5 in __expand_table () from /usr/lib/libc.so.5
> >>#4  0x281d4972 in __addel () from /usr/lib/libc.so.5
> >>#5  0x281d73db in __hash_open () from /usr/lib/libc.so.5
> >>#6  0x281d7009 in __hash_open () from /usr/lib/libc.so.5
> >>#7  0x2817c6b2 in dbm_store () from /usr/lib/libc.so.5
> >>#8  0x28249c7d in XS_NDBM_File_STORE ()
> >>   from /usr/local/lib/perl5/5.6.1/mach/auto/NDBM_File/NDBM_File.so
> >>#9  0x080a01e3 in Perl_pp_entersub ()
> >>#10 0x0809a638 in Perl_runops_standard ()
> >>#11 0x0805c187 in perl_call_sv ()
> >>#12 0x0805bd42 in perl_call_sv ()
> >>#13 0x0805bb43 in perl_call_method ()
> >>#14 0x08093888 in Perl_magic_setnkeys ()
> >>#15 0x08093ba0 in Perl_magic_setpack ()
> >>#16 0x080920dc in Perl_mg_set ()
> >>#17 0x0809aa82 in Perl_pp_sassign ()
> >>#18 0x0809a638 in Perl_runops_standard ()
> >>#19 0x0805b7e6 in perl_run ()
> >>#20 0x0805b4cb in perl_run ()
> >>#21 0x08058eab in main ()
> >>#22 0x08058d65 in _start ()
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>-------------------------------------------------------
> >>This SF.Net email is sponsored by: IBM Linux Tutorials
> >>Free Linux tutorial presented by Daniel Robbins, President and CEO of
> >>GenToo technologies. Learn everything from fundamentals to system
> >>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> >>_______________________________________________
> >>flexbackup-help mailing list
> >>fle...@li...
> >>https://lists.sourceforge.net/lists/listinfo/flexbackup-help
> > 
> > 
> > 
> > 
> > -------------------------------------------------------
> > This SF.Net email is sponsored by: IBM Linux Tutorials
> > Free Linux tutorial presented by Daniel Robbins, President and CEO of
> > GenToo technologies. Learn everything from fundamentals to system
> > administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> > _______________________________________________
> > flexbackup-help mailing list
> > fle...@li...
> > https://lists.sourceforge.net/lists/listinfo/flexbackup-help

Re: [flexbackup-help] serious error : flex is core dumping

[Corey M. <ma...@pu...>]

I'm gonna guess your running Red Hat 9... the perl and perl-NDBM_File 
packages conflict (I just tried updating it (with yum) yesterday and 
came across it by chance, I don't have a solution to it, just pointing 
you in the right direction)


--Corey

Jeff MacDonald wrote:
> On a whim, i took my logs,index,data and timestamps and moved them out
> of the way. Bingo it started working again.
> 
> So I figure I'll wait 6 more weeks see if it dies again, if it does I'll
> either mail this list again, or try amanda.
> 
> Anyone have any suggestions, or is anyone interested at all in looking
> deeper into this ?
> 
> Jeff.
> 
> On Tue, 2004-04-06 at 09:09, Jeff MacDonald wrote:
> 
>>Hi,
>>
>>Subject says it all, I've been running flex fine on a machine for about
>>6 weeks, then suddenly in my cron emails it tells me that it's core
>>dumping. It core's whether I dump a local or remote file system etc..
>>
>>-su-2.05b$ flexbackup -set gollum -level 3
>>
>>flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
>>/etc/flexbackup.conf syntax OK
>>
>>|------------------------------------------------------------
>>| Checking /bin/sh on this machine... unknown (probably Bourne Shell)
>>|------------------------------------------------------------
>>| Doing level 3 backup of set gollum using tar
>>| Found directory index key 200402121142
>>| Backup set "gollum" (/usr/home /etc/ /root/ /usr/local/etc/)
>>|------------------------------------------------------------
>>Segmentation fault (core dumped)
>>
>>Here's what gdb says [tho i have no clue what it means.
>>
>>#0  0x281e447a in memmove () from /usr/lib/libc.so.5
>>#1  0x00000039 in ?? ()
>>#2  0x281d43b2 in __split_page () from /usr/lib/libc.so.5
>>#3  0x281d77a5 in __expand_table () from /usr/lib/libc.so.5
>>#4  0x281d4972 in __addel () from /usr/lib/libc.so.5
>>#5  0x281d73db in __hash_open () from /usr/lib/libc.so.5
>>#6  0x281d7009 in __hash_open () from /usr/lib/libc.so.5
>>#7  0x2817c6b2 in dbm_store () from /usr/lib/libc.so.5
>>#8  0x28249c7d in XS_NDBM_File_STORE ()
>>   from /usr/local/lib/perl5/5.6.1/mach/auto/NDBM_File/NDBM_File.so
>>#9  0x080a01e3 in Perl_pp_entersub ()
>>#10 0x0809a638 in Perl_runops_standard ()
>>#11 0x0805c187 in perl_call_sv ()
>>#12 0x0805bd42 in perl_call_sv ()
>>#13 0x0805bb43 in perl_call_method ()
>>#14 0x08093888 in Perl_magic_setnkeys ()
>>#15 0x08093ba0 in Perl_magic_setpack ()
>>#16 0x080920dc in Perl_mg_set ()
>>#17 0x0809aa82 in Perl_pp_sassign ()
>>#18 0x0809a638 in Perl_runops_standard ()
>>#19 0x0805b7e6 in perl_run ()
>>#20 0x0805b4cb in perl_run ()
>>#21 0x08058eab in main ()
>>#22 0x08058d65 in _start ()
>>
>>
>>
>>
>>
>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by: IBM Linux Tutorials
>>Free Linux tutorial presented by Daniel Robbins, President and CEO of
>>GenToo technologies. Learn everything from fundamentals to system
>>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
>>_______________________________________________
>>flexbackup-help mailing list
>>fle...@li...
>>https://lists.sourceforge.net/lists/listinfo/flexbackup-help
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> flexbackup-help mailing list
> fle...@li...
> https://lists.sourceforge.net/lists/listinfo/flexbackup-help

Re: [flexbackup-help] serious error : flex is core dumping

[Jeff M. <in...@bi...>]

On a whim, i took my logs,index,data and timestamps and moved them out
of the way. Bingo it started working again.

So I figure I'll wait 6 more weeks see if it dies again, if it does I'll
either mail this list again, or try amanda.

Anyone have any suggestions, or is anyone interested at all in looking
deeper into this ?

Jeff.

On Tue, 2004-04-06 at 09:09, Jeff MacDonald wrote:
> Hi,
> 
> Subject says it all, I've been running flex fine on a machine for about
> 6 weeks, then suddenly in my cron emails it tells me that it's core
> dumping. It core's whether I dump a local or remote file system etc..
> 
> -su-2.05b$ flexbackup -set gollum -level 3
> 
> flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
> /etc/flexbackup.conf syntax OK
> 
> |------------------------------------------------------------
> | Checking /bin/sh on this machine... unknown (probably Bourne Shell)
> |------------------------------------------------------------
> | Doing level 3 backup of set gollum using tar
> | Found directory index key 200402121142
> | Backup set "gollum" (/usr/home /etc/ /root/ /usr/local/etc/)
> |------------------------------------------------------------
> Segmentation fault (core dumped)
> 
> Here's what gdb says [tho i have no clue what it means.
> 
> #0  0x281e447a in memmove () from /usr/lib/libc.so.5
> #1  0x00000039 in ?? ()
> #2  0x281d43b2 in __split_page () from /usr/lib/libc.so.5
> #3  0x281d77a5 in __expand_table () from /usr/lib/libc.so.5
> #4  0x281d4972 in __addel () from /usr/lib/libc.so.5
> #5  0x281d73db in __hash_open () from /usr/lib/libc.so.5
> #6  0x281d7009 in __hash_open () from /usr/lib/libc.so.5
> #7  0x2817c6b2 in dbm_store () from /usr/lib/libc.so.5
> #8  0x28249c7d in XS_NDBM_File_STORE ()
>    from /usr/local/lib/perl5/5.6.1/mach/auto/NDBM_File/NDBM_File.so
> #9  0x080a01e3 in Perl_pp_entersub ()
> #10 0x0809a638 in Perl_runops_standard ()
> #11 0x0805c187 in perl_call_sv ()
> #12 0x0805bd42 in perl_call_sv ()
> #13 0x0805bb43 in perl_call_method ()
> #14 0x08093888 in Perl_magic_setnkeys ()
> #15 0x08093ba0 in Perl_magic_setpack ()
> #16 0x080920dc in Perl_mg_set ()
> #17 0x0809aa82 in Perl_pp_sassign ()
> #18 0x0809a638 in Perl_runops_standard ()
> #19 0x0805b7e6 in perl_run ()
> #20 0x0805b4cb in perl_run ()
> #21 0x08058eab in main ()
> #22 0x08058d65 in _start ()
> 
> 
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> flexbackup-help mailing list
> fle...@li...
> https://lists.sourceforge.net/lists/listinfo/flexbackup-help

[flexbackup-help] serious error : flex is core dumping

[Jeff M. <in...@bi...>]

Hi,

Subject says it all, I've been running flex fine on a machine for about
6 weeks, then suddenly in my cron emails it tells me that it's core
dumping. It core's whether I dump a local or remote file system etc..

-su-2.05b$ flexbackup -set gollum -level 3

flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
/etc/flexbackup.conf syntax OK

|------------------------------------------------------------
| Checking /bin/sh on this machine... unknown (probably Bourne Shell)
|------------------------------------------------------------
| Doing level 3 backup of set gollum using tar
| Found directory index key 200402121142
| Backup set "gollum" (/usr/home /etc/ /root/ /usr/local/etc/)
|------------------------------------------------------------
Segmentation fault (core dumped)

Here's what gdb says [tho i have no clue what it means.

#0  0x281e447a in memmove () from /usr/lib/libc.so.5
#1  0x00000039 in ?? ()
#2  0x281d43b2 in __split_page () from /usr/lib/libc.so.5
#3  0x281d77a5 in __expand_table () from /usr/lib/libc.so.5
#4  0x281d4972 in __addel () from /usr/lib/libc.so.5
#5  0x281d73db in __hash_open () from /usr/lib/libc.so.5
#6  0x281d7009 in __hash_open () from /usr/lib/libc.so.5
#7  0x2817c6b2 in dbm_store () from /usr/lib/libc.so.5
#8  0x28249c7d in XS_NDBM_File_STORE ()
   from /usr/local/lib/perl5/5.6.1/mach/auto/NDBM_File/NDBM_File.so
#9  0x080a01e3 in Perl_pp_entersub ()
#10 0x0809a638 in Perl_runops_standard ()
#11 0x0805c187 in perl_call_sv ()
#12 0x0805bd42 in perl_call_sv ()
#13 0x0805bb43 in perl_call_method ()
#14 0x08093888 in Perl_magic_setnkeys ()
#15 0x08093ba0 in Perl_magic_setpack ()
#16 0x080920dc in Perl_mg_set ()
#17 0x0809aa82 in Perl_pp_sassign ()
#18 0x0809a638 in Perl_runops_standard ()
#19 0x0805b7e6 in perl_run ()
#20 0x0805b4cb in perl_run ()
#21 0x08058eab in main ()
#22 0x08058d65 in _start ()

[flexbackup-help] restore problems

[Andy <an...@ac...>]

Hi All,
I've got backups up and running now, but have a question about restore,
As suggested by Roger (And lots of help by him, many thanks), I've setup
a restore using the following (the first line his suggestion to set the
tape up). Then the afio, but the problem is shown below. A lot of files
show the broken pipe error, which I'm guessing is because the original
file wasn't compressed when it was written, and now trying to uncompress
isn't working on that file.
once the restore finished, the file exists, but it has 0 bytes. Is there
an extra flag/option to afio that I should be adding to fix this, or any
ideas???!

mt -f /dev/nht0 fsf 1
afio -i -k -x -P bzip2 -Q -d -Z -v /dev/nht0

afio: "shares/misc/del/temp/temp/fax to kath re 1 cheviot way.doc": Broken =
pipe
afio: "inentry xwait()": Exit 2
shares/misc/del/temp/temp/fax to kath re 1 cheviot way.doc -- uncompressed
bzip2: (stdin) is not a bzip2 file.
afio: "shares/misc/del/temp/temp/Fax.walter-univ.north 6.3.03..doc": Broken=
 pipe
afio: "inentry xwait()": Exit 2
shares/misc/del/temp/temp/Fax.walter-univ.north 6.3.03..doc -- uncompressed
shares/misc/del/temp/temp/My Pictures/P1010032.JPG -- okay


Info:
shares/misc/del/temp/temp/Fax.walter-univ.north 6.3.03..doc being the
file that i'm testing, as that exists normally, but when restored has 0
bytes due to the errors shown (I'm guessing).

--
Andy
<an...@ac...>  http://www.acomm-hosting.co.uk/

Web Design from =A3200  Hosting from =A380 and domains only =A315
ADSL for =A320 per month at http://tinyurl.com/iyej

[flexbackup-help] RE: loader

[Corey M. <ma...@pu...>]

"mtx next" unloads the current tape and then loads a tape from the next 
higher numbered slot -- my autoloader also automatically loads the first 
tape on startup (no need for the first mtx load; which it wouldn't 
matter anyway as "mtx next" will put in a tape... mtx load requires a 
slot number; unload does not (tape goes back into the slot it came from)

You could do mtx unload at the end, but an mtx next is probably 
preferable (make the next tape the first of the next set)

My autoloader does not do barcodes so this isn't an issue there. It also 
has a sequential mode which will span the tapes for me automatically 
(make them act as one tape)

As a suggestion, you might want to have a tape-change script (e.g. echo 
"Please insert new tape, then press enter"; read) or "mtx next" or 
however the end user would want to change the tapes... when you hit the 
end of the medium (that is an ideal solution though)

Hope this helps,
	--Corey

sample output (tape 1 is already loaded as if it were power-cycled):
[root@backup root]# mtx next
Unloading Data Transfer Element into Storage Element 1...done
[root@backup root]# mtx inquiry
Product Type: Medium Changer
Vendor ID: 'SEAGATE '
Product ID: 'LTO LDR CLL1600 '
Revision: 'S17r'
Attached Changer: No
[root@backup root]# mtx status
   Storage Changer /dev/changer:1 Drives, 8 Slots ( 0 Import/Export )
Data Transfer Element 0:Full (Storage Element 2 Loaded)
       Storage Element 1:Full
       Storage Element 2:Empty
       Storage Element 3:Full
       Storage Element 4:Full
       Storage Element 5:Full
       Storage Element 6:Full
       Storage Element 7:Full
       Storage Element 8:Full
[root@backup root]# mtx next
Unloading Data Transfer Element into Storage Element 2...done
[root@backup root]# mtx status
   Storage Changer /dev/changer:1 Drives, 8 Slots ( 0 Import/Export )
Data Transfer Element 0:Full (Storage Element 3 Loaded)
       Storage Element 1:Full
       Storage Element 2:Full
       Storage Element 3:Empty
       Storage Element 4:Full
       Storage Element 5:Full
       Storage Element 6:Full
       Storage Element 7:Full
       Storage Element 8:Full
[root@backup root]# mtx next
Unloading Data Transfer Element into Storage Element 3...done
[root@backup root]# mtx status
   Storage Changer /dev/changer:1 Drives, 8 Slots ( 0 Import/Export )
Data Transfer Element 0:Full (Storage Element 4 Loaded)
       Storage Element 1:Full
       Storage Element 2:Full
       Storage Element 3:Full
       Storage Element 4:Empty
       Storage Element 5:Full
       Storage Element 6:Full
       Storage Element 7:Full
       Storage Element 8:Full
[root@backup root]# mtx last
Unloading Data Transfer Element into Storage Element 4...done
[root@backup root]# mtx status
   Storage Changer /dev/changer:1 Drives, 8 Slots ( 0 Import/Export )
Data Transfer Element 0:Full (Storage Element 8 Loaded)
       Storage Element 1:Full
       Storage Element 2:Full
       Storage Element 3:Full
       Storage Element 4:Full
       Storage Element 5:Full
       Storage Element 6:Full
       Storage Element 7:Full
       Storage Element 8:Empty
[root@backup root]# mtx first
Unloading Data Transfer Element into Storage Element 8...done
[root@backup root]# mtx status
   Storage Changer /dev/changer:1 Drives, 8 Slots ( 0 Import/Export )
Data Transfer Element 0:Full (Storage Element 1 Loaded)
       Storage Element 1:Empty
       Storage Element 2:Full
       Storage Element 3:Full
       Storage Element 4:Full
       Storage Element 5:Full
       Storage Element 6:Full
       Storage Element 7:Full
       Storage Element 8:Full

[ Stephane Gauthier wrote: ]
  > I'm currently researching for a cheap backup solution for a cluster 
of 8
  > machines using an HP Surestore DAT 40x6e autoloader and was 
wondering if
  > you were planning on adding generic support  for loaders? (via the 
mtx util)

  I don't have access to such a device, or knowledge of how to operate them,
  so I'd need someone to work out the details, modify things, test, etc.

  About all I have is an mtx manpage :)

  One archive spanning multiple tapes would be hard, but bolting on mtx
  support to normal operation should not be too bad.

  Using flexbackups' method of "each set == new tape if doing level 0", 
do we
  just use "mtx next" instead of the manual prompting that's there now?

  It is mainly a matter of something like:
     - "mtx load" to start off,
     - "mtx next" to switch to a new tape
     - "mtx unload" when done
     - use "scsitape" rather than "mt"?

  Or is there more to it?  Just doing a quick search I can't find anything
  like an "mtx howto" document.  From what I understand some of these things
  get quite fancy and can do bar-code labels and such....


  (Anyone who knows how changers+mtx work is welcome to help out)


  --
   Edwin Huffstutler
   edwinh@co...   GnuPG Key ID: AE782DC9

Re[2]: [flexbackup-help] ide-scsi problems

[Andy <an...@ac...>]

On Fri, 19 Mar 2004 02:36:32 -0800
Roger Pryor <rp...@pr...> wrote:

RP> On Fri, 19 Mar 2004 09:18:36 -0000 (GMT)
RP> "Andy" <an...@ac...> wrote:
RP> 
RP> > Hi All,
RP> > I've done quite a few searches and read previous posts but none seem to
RP> > help out my situation.
RP> > 
RP> > I'm using the IDE SCSI emulation (Tape drive is Seagate Model:
RP> > STT20000A) dmesg shows the tape driver loads, identifies the unit and
RP> > then spits out: st: Version 20030403, bufsize 32768, max init. bufs 4,
RP> > s/g segs 16 Attached scsi tape st0 at scsi0, channel 0, id 0, lun 0
RP> > st0: Error with sense data: Current st09:00: sense key Illegal Request
RP> > Additional sense indicates Invalid command operation code
RP> > 
RP> 
RP> Hi:
RP> 
RP> I use the same tape drive.  Don't use SCSI, use hd?=ide-tape in your boot
RP> package.  Then set the blksize to 32k, and then use the 
RP> -test-tape-drive option to verify that it works. Much faster to test. 
RP> 
RP> This drive has one problem:  if the write operation fails, you usually
RP> have to power-cycle the machine (ugh) to restore the drive to sanity.  If
RP> you can fix that one, please let me know how.
RP> 
RP> The relevant bits from my config file are:
RP> 
RP> $type = 'afio';
RP> # Compression
RP> $compress = 'gzip'; # one of false/gzip/bzip2/zip/compress/hardware
RP> $compr_level = '4'; # compression level (1-9) (for gzip/bzip2/zip)
RP> 
RP> $device = '/dev/tape';        (this is linked to /dev/nht0
RP> 
RP> $blksize = '32';
RP> 
RP> $mt_blksize = "0";
RP> 

No luck unfortunately, doing exactly as Roger said, and the test still
fails as follows:


flexbackup -test-tape-drive

flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
/etc/flexbackup.conf syntax OK

|------------------------------------------------------------
| Checking /bin/sh on this machine... bash2
|------------------------------------------------------------
| Testing will *erase* the tape currently in the drive!
| Hit CTRL-C to abort within 10 seconds...
|------------------------------------------------------------
| If for some reason this program does not exit within a few minutes,
| Hit CTRL-C, and try adjusting $blksize, $pad_blocks, or $mt_blksize.
|------------------------------------------------------------
| Rewinding & erasing tape...
| Creating index key 200403191920
|------------------------------------------------------------
At block 1.

Writing test file #1
4+1 records in
4+1 records out
At block 6.
Writing test file #2
0+1 records in
0+1 records out
At block 8.
Writing test file #3
4+1 records in
4+1 records out
At block 13.

Rewinding...
Skipping index label...
At block 1.

Reading test file #1
5+1 records in
5+1 records out
At block 6.
Reading test file #2
1+1 records in
1+1 records out
At block 8.
Reading test file #3
5+1 records in
5+1 records out
At block 13.

Comparing...
Files /usr/bin/flexbackup and /tmp/test1.3771 differ
Files /etc/flexbackup.conf and /tmp/test2.3771 differ
Files /usr/bin/flexbackup and /tmp/test3.3771 differ

FAILURE! Problem with tape driver or parameters.  Please see the FAQ
or try changing the $blksize, $pad_blocks, or $mt_blksize settings.


Any more suggestions?? I'm going mad!!

Andy.

Re: [flexbackup-help] ide-scsi problems

[Roger P. <rp...@pr...>]

On Fri, 19 Mar 2004 09:18:36 -0000 (GMT)
"Andy" <an...@ac...> wrote:

> Hi All,
> I've done quite a few searches and read previous posts but none seem to
> help out my situation.
> 
> I'm using the IDE SCSI emulation (Tape drive is Seagate Model:
> STT20000A) dmesg shows the tape driver loads, identifies the unit and
> then spits out: st: Version 20030403, bufsize 32768, max init. bufs 4,
> s/g segs 16 Attached scsi tape st0 at scsi0, channel 0, id 0, lun 0
> st0: Error with sense data: Current st09:00: sense key Illegal Request
> Additional sense indicates Invalid command operation code
> 

Hi:

I use the same tape drive.  Don't use SCSI, use hd?=ide-tape in your boot
package.  Then set the blksize to 32k, and then use the 
-test-tape-drive option to verify that it works. Much faster to test. 

This drive has one problem:  if the write operation fails, you usually
have to power-cycle the machine (ugh) to restore the drive to sanity.  If
you can fix that one, please let me know how.

The relevant bits from my config file are:

$type = 'afio';
# Compression
$compress = 'gzip'; # one of false/gzip/bzip2/zip/compress/hardware
$compr_level = '4'; # compression level (1-9) (for gzip/bzip2/zip)

$device = '/dev/tape';        (this is linked to /dev/nht0

$blksize = '32';

$mt_blksize = "0";



--------------------------------------------------------------------------
PLEASE NOTE: NEW EMAIL ADDRESS, please change your nicknames, thanks.
Roger Pryor P. Eng.                Email: rp...@pr...
Pryor and Pryor Inc.           Telephone: (+1) 604-685-2621
602 - 1230 Comox Street,        Internet: http://www.pryor-and-pryor.com
Vancouver, B.C., V6E 1K7, Canada

[flexbackup-help] ide-scsi problems

[Andy <an...@ac...>]

Hi All,
I've done quite a few searches and read previous posts but none seem to
help out my situation.

I'm using the IDE SCSI emulation (Tape drive is Seagate Model: STT20000A)
dmesg shows the tape driver loads, identifies the unit and then spits out:
st: Version 20030403, bufsize 32768, max init. bufs 4, s/g segs 16
Attached scsi tape st0 at scsi0, channel 0, id 0, lun 0
st0: Error with sense data: Current st09:00: sense key Illegal Request
Additional sense indicates Invalid command operation code

But from what i've found this is normal (!).

When using flexbackup:

flexbackup -test-tape-drive

flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
/etc/flexbackup.conf syntax OK

|------------------------------------------------------------
| Checking /bin/sh on this machine... bash2
|------------------------------------------------------------
| Trying "mt defblksize" instead of "mt setblk"
Error setting block size
Neither of these commands worked:
  mt -f /dev/nst0 setblk 0
  mt -f /dev/nst0 defblksize 0


Relevant info in the config file:
--CONFIG-- /etc/flexbackup.conf
$type = 'afio';
$device = '/dev/nst0';
$blksize = '0';
$mt_blksize = "0";
$pad_blocks = 'false';
$mt{'setblk'} = 'nop';

(I've tried all the blksize, mt_blksize settings, and also added/removed
the $mt{'setblk'} options and no change to the above error message).

As in the FAQ I tried out:
mt -f /dev/nst0 setblk 32768
  	(mt: /dev/nst0: Input/output error)
mt -f /dev/nst0 rewind
mt -f /dev/nst0 erase
find /etc -print | afio -o -z -v -b 32k - | dd ibs=32k obs=32k of=/dev/nst0

mt -f /dev/nst0 rewind
dd ibs=32k obs=32k if=/dev/nst0 | afio -t -z -v -b 32k -

Apart from the setblk command which failed, the rest worked perfectly, and
doing the listing showed the full list of files from /etc as it should, so
it appears the tape drive is working right.

Any suggestions on how to get flexbackup working in this situation??

Thanks for any help!

Andy <an...@ac...>

[flexbackup-help] where do I put config file?

[Jim M. <ja...@bt...>]

Annoying I know, have looked around and not found it! :-(

Where do I put config file and what should it be called?

Thanks a bunch

Jim

-- 


==============================================
Jim Maas Ph.D.
29 Amber Close
Earley, Reading, UK
RG6 7ED

0118-935-3283
0771-985-8698
ja...@bt...

[flexbackup-help] multiple volumes

[<sp...@da...>]

Hi
I tried a to backup to multiple volumes.
The backup part seemed to work okay (flexbackup -set test -volumes 2) even
though was a bit slow.

But when trying to restore it did not work.
I could not find any special arguments I tried to restore 1 file that was on the
sn2 tape.
It first had to read the whole tape (which is quite along time if you know you
need something of the 2nd tape.) and than it gave this error:

0 CONTROL_FILE/flexbackup.volume_header_info -- okay
afio: "-" [offset 3739m+884k+0]: Fatal error:
afio: "-": Premature input EOF
|------------------------------------------------------------
At block 957408.

I guess there has to bee some extra option passed to mbuffer or something?

I tried the same command on the sn2 tape:

flexbackup -extract

afio: "-" [offset 6]: Fatal error:
afio: "-": Unrecognizable archive

Any idea how to restore it tape by tape?

I know it is experimental but the restore does not seem to work at all. The
backup part ask nicely for the next tape. But I wonder what is writen on the
sn2 tape and how to get it back.

I prefer to beable to just get a few files from it but a full restore is also
fine.
My biggest problem is that I only have a dds 2 tape streamer and I really want
to backup my whole system which probably need 10 tapes if I use compression.
tar cannot handle it and I could make a script that calculates the file size
but that is the uncompressed filesize.

help :-)

Regards

Hans de Groot

[flexbackup-help] (no subject)

[<sp...@da...>]

Hello,

I just installed flexbackup 1.2.1-1

The reason I tried it was it has experimental support for multiple tapes. anyway
after configuring it I am not running a test but it seems super slow.
I am trying to backup mp3 files and I added them to the exclude list so they are
not compressed.
My cpu usage is super low to it's not wating for a slow cpu.
My tape led is flashing irregular like it is not doing much.
It's a dds 2 tape drive which should handle about 60 Mbyte per minute.
When I use tar it does reach arround those speeds.
With flexbackup it seems it is taking about 10 minutes or more to backup 60 mb.
This seems way to slow to me.
Now I use afio and mbuffer and the option -volumes 3

Does anyone have any advice why flexbackup goes so slow?

Thanks very much.

Regards

Hans de Groot


This is the contents of the /var/log/flexbackup/test.xxxx
-----
------------------------------------------------------------
| Doing level 0 backup of set test using afio
| ERROR: Tape doesn't have an index! (use -newtape?)
| Making sure tape is at end of data...
| Backup set "test" (/data5/music2)
| Tape #0
|------------------------------------------------------------
At block 0.
|------------------------------------------------------------
| Backup of: /data5/music2
| Date of this level 0 backup: Sat Mar 13 16:15:53 2004
| Date of last level 0 backup: the epoch
|------------------------------------------------------------
| printf "mp3 MP3 Z z gz gif zip ZIP lha jpeg jpg JPG taz tgz deb rpm bz2 \
|   lzo avi AVI mpg MPG mpeg MPEG arj rar RAR" > /tmp/nocompress.3611
| printf "Volume Label:\nlevel 0 /data5/music2 Sat Mar 13 16:15:53 2004 \
|   afio+gzip from koffie.dandy.nl\n\n" > /tmp/label.3611
| cd "/data5/music2" && (printf "//--/tmp/label.3611 \
|   flexbackup.volume_header_info\n" && find . -depth -xdev ! -type s \
|   -print ) | afio -o -E /tmp/nocompress.3611 -z -1 m -P gzip -Q -9 -Z -M \
|   6m -T 1k -v -b 10k - | mbuffer -q -m 10M -p 75 -s 10240 -f -o \
|   "/dev/nst0"
| rm -f /tmp/nocompress.3611 /tmp/label.3611
|------------------------------------------------------------
CONTROL_FILE/flexbackup.volume_header_info -- okay
---------------

[flexbackup-help] SCSI Performance Issue

[Steven L. <ste...@ho...>]

Hi all,

I use flexbackup with an Adaptec 2940UW and a HP Surestore DAT24 drive.  I 
have only recently started using flexbackup having found it the best tool I 
could find, but the first backup I tried, took a ridiculously long time (it 
was transferring at 50k per second, which for a SCSI DAT is not right).

I installed BUFFER to no avail.  Am I missing something here?  Is there a 
confiuration option I need to set?   I set it to use DUMP because I dont 
know anything about AFIO, but would it be better with AFIO?

Here is the .conf part for buffer:
$buffer = 'buffer';         # one of false/buffer/mbuffer
$buffer_megs = '20';        # buffer memory size (in megabytes)
$buffer_fill_pct = '75';    # start writing when buffer this percent full
$buffer_pause_usec = '100'; # pause after write (tape devices only)

Compression:
$compress = 'bzip2'; # one of false/gzip/bzip2/lzop/zip/compress/hardware
$compr_level = '4'; # compression level (1-9) (for gzip/bzip2/lzop/zip)

SCSI Part:
$device = '/dev/st0';
$blksize = '10';


Atimes and DUMP dates are off, maybe I am just being stupid here, if so 
someone please slap me... =)

If you need more info please let me know.

Regards
Steve

_________________________________________________________________
It's fast, it's easy and it's free. Get MSN Messenger today! 
http://www.msn.co.uk/messenger

Re: [flexbackup-help] Remote restore

[<bjs...@be...>]

Sebastien Prud'homme <pru...@la...> wrote:

> I use a server to backup itself and a remote server (with the
> help of ssh). Now i'd like to know how to restore files of the
> remote server directly on the remote server.

Sebastien

I also use Flexbackup in the same situation as you and there does not
appear to be a remote restore option.

If you have ssh working properly on both machines, that is you can ssh
to either machine from the other, then the following should work:

ssh into your remote machine and change to the remote directory that
you want to restore into.

Enter any mt commands to position the tape on your local machine by
using mt -f user@localmachine:/dev/tape <command> where user is the
user on your local machine, localmachine is your host name for that
machine and /dev/tape is the tape device name that you normally use.

Enter afio -i <any afio options here> user@localmachine:/dev/tape

You may get a 'broken pipe' error at the end of the process but it
appears to work for me.

You could try afio -t user@localmachine:/dev/tape first just to try it
out.

(Copy of this post to Flexbackup-Help mailing list)


Barry Samuels
http://www.beenthere-donethat.org.uk
The Unofficial Guide to Great Britain

[flexbackup-help] Extract problem

[Kelvin P. <kel...@ya...>]

With the help of flexbackup, I can backup files to the tape drive
I make Set 1 in /etc/flexbackup.conf
$set{'set1'} =3D "/bin /boot /dev /etc /home /initrd /lib /opt /root =
/sbin
/tmp /usr";
And start the backup using the command flexbackup -set all
The problem is, I don't know how to extract the file.
For example, I have a file in /root/bin/coWednesday.dmp.z
I use the command: flexbackup -num 9 -extract -onefile
bin/coWednesday.dmp.z
But no file extracted.
[root@intranet root]# flexbackup -num 9 -extract -onefile
bin/coWednesday.dmp.z
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE =3D (unset),
        LC_ALL =3D "default",
        LANG =3D "en_US"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
=20
flexbackup version 1.2.1 (http://flexbackup.sourceforge.net)
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE =3D (unset),
        LC_ALL =3D "default",
        LANG =3D "en_US"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
/etc/flexbackup.conf syntax OK
=20
|------------------------------------------------------------
| Checking 'mbuffer' on this machine... Ok
| Checking /bin/sh on this machine... bash2
|------------------------------------------------------------
| Logging output to "flexbackup.extract.200402271733.log"
| Positioning tape at file number 9
/dev/tape: Input/output error
|------------------------------------------------------------
At block 0.
|------------------------------------------------------------
| Extracting single file bin/coWednesday.dmp.z
| mbuffer -q -m 20M -p 75 -s 10240 -f -i "/dev/tape" | afio -i -w
|   /tmp/extract.20709 -z -x -D /usr/bin/flexbackup -P gzip -Q -d -Q -q
-Z
|   -v   -b 10k -
|------------------------------------------------------------
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE =3D (unset),
        LC_ALL =3D "default",
        LANG =3D "en_US"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
=20
Volume Label:
level 0 /usr Fri Feb 27 05:22:03 2004 afio+gzip from intranet
=20
CONTROL_FILE/flexbackup.volume_header_info - okay
=20
Sorry, I am new on linux. Thanks in advance
=20