Re: include problems with Amanda-2.4.4-20030428?

2003-11-26 Thread Paul Bijnens
Jean-Francois Malouin wrote:


The relevent sendbackup file for this DLE with include stanza:

sendbackup: debug 1 pid 10655570 ruid 666 euid 666: start at Tue Nov 25 04:41:51 2003
/opt/amanda/amanda1/libexec/sendbackup: version 2.4.4p1-20031120
  parsed request as: program `GNUTAR'
 disk `/data/nihpd/nihpd1/data/assembly_1'
 device `/data/nihpd/nihpd1/data/assembly'
 level 0
 since 1970:1:1:0:0:0
 options `|;auth=bsd;index;'


Not good! It should have an entry for the include/exclude like:

sendbackup: debug 1 pid 21690 ruid 1001 euid 1001: start at Wed Nov 26 
11:53:16 2003
/pkgs/amanda-2.4.4p1/libexec/sendbackup: version 2.4.4p1
  parsed request as: program `GNUTAR'
 disk `/space/scratch/pbi/brolpol'
 device `/space/scratch/pbi/brolpol'
 level 0
 since 1970:1:1:0:0:0
 options `|;auth=bsd;no-record;index;include-file=./[a-z]*;'



What does this say:
  $ amadmin YourConfig disklist
Do you see the include statement?

--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  Are you sure?  ...   YES   ...   Phew ...   I'm out  *
***



Re: include problems with Amanda-2.4.4-20030428?

2003-11-26 Thread Jean-Francois Malouin
* Paul Bijnens [EMAIL PROTECTED] [20031126 06:02]:
 Jean-Francois Malouin wrote:
 
 
 The relevent sendbackup file for this DLE with include stanza:
 
 sendbackup: debug 1 pid 10655570 ruid 666 euid 666: start at Tue Nov 25 
 04:41:51 2003
 /opt/amanda/amanda1/libexec/sendbackup: version 2.4.4p1-20031120
   parsed request as: program `GNUTAR'
  disk `/data/nihpd/nihpd1/data/assembly_1'
  device `/data/nihpd/nihpd1/data/assembly'
  level 0
  since 1970:1:1:0:0:0
  options `|;auth=bsd;index;'
 
 
 Not good! It should have an entry for the include/exclude like:
 
 sendbackup: debug 1 pid 21690 ruid 1001 euid 1001: start at Wed Nov 26 
 11:53:16 2003
 /pkgs/amanda-2.4.4p1/libexec/sendbackup: version 2.4.4p1
   parsed request as: program `GNUTAR'
  disk `/space/scratch/pbi/brolpol'
  device `/space/scratch/pbi/brolpol'
  level 0
  since 1970:1:1:0:0:0
  options `|;auth=bsd;no-record;index;include-file=./[a-z]*;'
 
 
 
 What does this say:
   $ amadmin YourConfig disklist
 
 Do you see the include statement?

no!
~# amadmin stk_40-conf1 disklist
line 13:
host bullcalf:
interface default
disk /data/nihpd/nihpd1/data/assembly_1:
device /data/nihpd/nihpd1/data/assembly
program GNUTAR
priority 2
dumpcycle 14
maxdumps 1
maxpromoteday 1
strategy STANDARD
compress NONE
auth BSD
kencrypt NO
holdingdisk YES
record YES
index YES
skip-incr NO
skip-full NO

jf

 
 -- 
 Paul Bijnens, XplanationTel  +32 16 397.511
 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
 http://www.xplanation.com/  email:  [EMAIL PROTECTED]
 ***
 * I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
 * quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
 * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
 * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
 * kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
 * ...  Are you sure?  ...   YES   ...   Phew ...   I'm out  *
 ***
 

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-26 Thread Martin
Maybe the line 'include ./[1]*' makes a difference? (instead of ./1*)

Martin



  The relevent sendbackup file for this DLE with include stanza:
  
  sendbackup: debug 1 pid 10655570 ruid 666 euid 666: start at Tue Nov 25
  04:41:51 2003
  /opt/amanda/amanda1/libexec/sendbackup: version 2.4.4p1-20031120
parsed request as: program `GNUTAR'
   disk `/data/nihpd/nihpd1/data/assembly_1'
   device `/data/nihpd/nihpd1/data/assembly'
   level 0
   since 1970:1:1:0:0:0
   options `|;auth=bsd;index;'
 
 
  Not good! It should have an entry for the include/exclude like:
 
  sendbackup: debug 1 pid 21690 ruid 1001 euid 1001: start at Wed Nov 26
  11:53:16 2003
  /pkgs/amanda-2.4.4p1/libexec/sendbackup: version 2.4.4p1
parsed request as: program `GNUTAR'
   disk `/space/scratch/pbi/brolpol'
   device `/space/scratch/pbi/brolpol'
   level 0
   since 1970:1:1:0:0:0
   options `|;auth=bsd;no-record;index;include-file=./[a-z]*;'
 
 
 
  What does this say:
$ amadmin YourConfig disklist
 
  Do you see the include statement?

 no!
 ~# amadmin stk_40-conf1 disklist
 line 13:
 host bullcalf:
 interface default
 disk /data/nihpd/nihpd1/data/assembly_1:
 device /data/nihpd/nihpd1/data/assembly
 program GNUTAR
 priority 2
 dumpcycle 14
 maxdumps 1
 maxpromoteday 1
 strategy STANDARD
 compress NONE
 auth BSD
 kencrypt NO
 holdingdisk YES
 record YES
 index YES
 skip-incr NO
 skip-full NO

 jf





Re: include problems with Amanda-2.4.4-20030428?

2003-11-26 Thread Jean-Francois Malouin
* Martin [EMAIL PROTECTED] [20031126 08:26]:
 Maybe the line 'include ./[1]*' makes a difference? (instead of ./1*)
 
 Martin

Turns out that the include stanza will show up if it is *after*
the dumptype line, ie:

bullcalf /data/nihpd/nihpd1/data/assembly_1 /data/nihpd/nihpd1/data/assembly {
high-tar
include ./1*
} 2

Then

~# amadmin stk_40-conf1 disklist
host bullcalf:
interface default
disk /data/nihpd/nihpd1/data/assembly_1:
device /data/nihpd/nihpd1/data/assembly
program GNUTAR
include file ./1*
priority 2
dumpcycle 14
maxdumps 1
maxpromoteday 1
strategy STANDARD
compress NONE
auth BSD
kencrypt NO
holdingdisk YES
record YES
index YES
skip-incr NO
skip-full NO

ah! A bug?

Thanks for the help to everybody who posted on this long thread.

jf

 
 
 
   The relevent sendbackup file for this DLE with include stanza:
   
   sendbackup: debug 1 pid 10655570 ruid 666 euid 666: start at Tue Nov 25
   04:41:51 2003
   /opt/amanda/amanda1/libexec/sendbackup: version 2.4.4p1-20031120
 parsed request as: program `GNUTAR'
disk `/data/nihpd/nihpd1/data/assembly_1'
device `/data/nihpd/nihpd1/data/assembly'
level 0
since 1970:1:1:0:0:0
options `|;auth=bsd;index;'
  
  
   Not good! It should have an entry for the include/exclude like:
  
   sendbackup: debug 1 pid 21690 ruid 1001 euid 1001: start at Wed Nov 26
   11:53:16 2003
   /pkgs/amanda-2.4.4p1/libexec/sendbackup: version 2.4.4p1
 parsed request as: program `GNUTAR'
disk `/space/scratch/pbi/brolpol'
device `/space/scratch/pbi/brolpol'
level 0
since 1970:1:1:0:0:0
options `|;auth=bsd;no-record;index;include-file=./[a-z]*;'
  
  
  
   What does this say:
 $ amadmin YourConfig disklist
  
   Do you see the include statement?
 
  no!
  ~# amadmin stk_40-conf1 disklist
  line 13:
  host bullcalf:
  interface default
  disk /data/nihpd/nihpd1/data/assembly_1:
  device /data/nihpd/nihpd1/data/assembly
  program GNUTAR
  priority 2
  dumpcycle 14
  maxdumps 1
  maxpromoteday 1
  strategy STANDARD
  compress NONE
  auth BSD
  kencrypt NO
  holdingdisk YES
  record YES
  index YES
  skip-incr NO
  skip-full NO
 
  jf
 
 

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-26 Thread Jean-Louis Martineau
On Wed, Nov 26, 2003 at 08:42:14AM -0500, Jean-Francois Malouin wrote:
 Turns out that the include stanza will show up if it is *after*
 the dumptype line, ie:
...  
 ah! A bug?

It would be the normal behavior if tar-high contain an include statement, but
it's a bug if it doesn't contain an include statement.

It will be fixed in today snapshot.

Jean-Louis
-- 
Jean-Louis Martineau email: [EMAIL PROTECTED] 
Département IRO, Université de Montréal
C.P. 6128, Succ. CENTRE-VILLETel: (514) 343-6111 ext. 3529
Montréal, Canada, H3C 3J7Fax: (514) 343-5834


Re: include problems with Amanda-2.4.4-20030428?

2003-11-26 Thread Jean-Francois Malouin
* Jean-Louis Martineau [EMAIL PROTECTED] [20031126 11:17]:
 On Wed, Nov 26, 2003 at 08:42:14AM -0500, Jean-Francois Malouin wrote:
  Turns out that the include stanza will show up if it is *after*
  the dumptype line, ie:
 ...  
  ah! A bug?
 
 It would be the normal behavior if tar-high contain an include statement, but
 it's a bug if it doesn't contain an include statement.

the dumptypes are specified as follow:

define dumptype high-tar {
root-tar
comment partitions dumped with tar
priority high
}

define dumptype root-tar {
global
program GNUTAR
comment root partitions dumped with tar
compress none
index
priority low
}

define dumptype global {
index yes
record yes
}


 
 It will be fixed in today snapshot.
 
 Jean-Louis
 -- 
 Jean-Louis Martineau email: [EMAIL PROTECTED] 
 Département IRO, Université de Montréal
 C.P. 6128, Succ. CENTRE-VILLETel: (514) 343-6111 ext. 3529
 Montréal, Canada, H3C 3J7Fax: (514) 343-5834

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-25 Thread Martin
FWIW, I am doing the same thing (splitting the stuff), my DLEs look
something like this:

host /home/all /home {
always-full
exclude ./hugemap ./evenhugermap ./*/.sambarecycle
}

host /home/hugemap /home {
always-full
include ./hugemap
}

host /home/evenhugermap /home {
always-full
include ./evenhugermap
}

Martin


 I'm having strange problems with the include thingy in an
 entry in my disklist:

 bullcalf /data/nihpd/nihpd1/data/assembly_0
/data/nihpd/nihpd1/data/assembly {
 include ./0*
 exclude ./[1-9]*
 high-tar
 } 1

 Looking at what got written to tape (a full backup) I see only
 directories (no files at all!), and even from the exclude
part...ie
 just the hierarchy structure of /data/nihpd/nihpd1/data/assembly
all
 the way down...

   
Don't know about your bad backup, but in your DLE you should not
have both the include and THAT exclude.  I.e., the include will
automatically drop any toplevel dir that does not match ./0*.
  
   That is want I want!
   All subdirs in there have a pattern [0-9]\{6\} , ie, 6 digits. There
   are other subdirs but there are not to be backed up. You see, my
   problem is that /data/nihpd/nihpd1/data/assembly can contains 100's of
   GB and I'm using 4 LTO tape drives, so I have to split that thing in
   small chuncks...
  
   Eventually I'll have 10 DLE's, those starting with ./0* ./1* ...
   you get the picture.
  
   
I think the only valid exclude usage would be for things under
the 0* directories.  For example 'exclude ./0*/tmp'.  Though
I must admit I've never tried that combination.
  




Re: include problems with Amanda-2.4.4-20030428?

2003-11-25 Thread Jean-Francois Malouin
* Martin [EMAIL PROTECTED] [20031125 02:48]:
 FWIW, I am doing the same thing (splitting the stuff), my DLEs look
 something like this:
 
 host /home/all /home {
 always-full
 exclude ./hugemap ./evenhugermap ./*/.sambarecycle
 }
 
 host /home/hugemap /home {
 always-full
 include ./hugemap
 }
 
 host /home/evenhugermap /home {
 always-full
 include ./evenhugermap
 }

Thanks for you input, Martin. But my situation is a little bit
different than yours. in /.../assembly I have a few thousands
of subdirs, 186 of them starts with a '1'. I just calculated
their total combined size: 11286948 KB or a little less than 11GB.

OK, now look at this amanda report from last night:

---
DUMP SUMMARY:
  DUMPER STATS  TAPER STATS  
HOSTNAME DISK LORIG-KB OUT-KB COMP% MMM:SSKB/s MMM:SSKB/s
--- -- --
bullcalf -/nihpd1 2  44220  44220   --   29:0425.4   0:10  4308.6
bullcalf -embly_0 1 120160 120160   --   22:1190.3   0:12 10093.6
bullcalf -embly_1 0  100460112  100460112   --  277:34  6032.2 277:34  6032.2

(brought to you by Amanda version 2.4.4p1-20031120)
---

I see that //assembly_1 was around 98GB which is actually the
entire content of /data/nihpd/nihpd1/data/assembly.

The disklist for this directory:

bullcalf /data/nihpd/nihpd1/data/assembly_1 /data/nihpd/nihpd1/data/assembly {
include ./1*
high-tar
} 2

hmmm, seems to me that the include thing does not behave as intended.
Maybe I'm just missing something though.

Any comments anyone? I really have to get this working asap!

thanks a million,
jf

 
 Martin
 
 
  I'm having strange problems with the include thingy in an
  entry in my disklist:
 
  bullcalf /data/nihpd/nihpd1/data/assembly_0
 /data/nihpd/nihpd1/data/assembly {
  include ./0*
  exclude ./[1-9]*
  high-tar
  } 1
 
  Looking at what got written to tape (a full backup) I see only
  directories (no files at all!), and even from the exclude
 part...ie
  just the hierarchy structure of /data/nihpd/nihpd1/data/assembly
 all
  the way down...
 

 Don't know about your bad backup, but in your DLE you should not
 have both the include and THAT exclude.  I.e., the include will
 automatically drop any toplevel dir that does not match ./0*.
   
That is want I want!
All subdirs in there have a pattern [0-9]\{6\} , ie, 6 digits. There
are other subdirs but there are not to be backed up. You see, my
problem is that /data/nihpd/nihpd1/data/assembly can contains 100's of
GB and I'm using 4 LTO tape drives, so I have to split that thing in
small chuncks...
   
Eventually I'll have 10 DLE's, those starting with ./0* ./1* ...
you get the picture.
   

 I think the only valid exclude usage would be for things under
 the 0* directories.  For example 'exclude ./0*/tmp'.  Though
 I must admit I've never tried that combination.
   
 

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-25 Thread Paul Bijnens
Jean-Francois Malouin wrote:

different than yours. in /.../assembly I have a few thousands
of subdirs, 186 of them starts with a '1'. I just calculated
their total combined size: 11286948 KB or a little less than 11GB.
OK, now look at this amanda report from last night:

---
DUMP SUMMARY:
  DUMPER STATS  TAPER STATS  
HOSTNAME DISK LORIG-KB OUT-KB COMP% MMM:SSKB/s MMM:SSKB/s
--- -- --
bullcalf -/nihpd1 2  44220  44220   --   29:0425.4   0:10  4308.6
bullcalf -embly_0 1 120160 120160   --   22:1190.3   0:12 10093.6
bullcalf -embly_1 0  100460112  100460112   --  277:34  6032.2 277:34  6032.2

(brought to you by Amanda version 2.4.4p1-20031120)
---
I see that //assembly_1 was around 98GB which is actually the
entire content of /data/nihpd/nihpd1/data/assembly.
The disklist for this directory:

bullcalf /data/nihpd/nihpd1/data/assembly_1 /data/nihpd/nihpd1/data/assembly {
include ./1*
high-tar
} 2
hmmm, seems to me that the include thing does not behave as intended.
Maybe I'm just missing something though.
Any comments anyone? I really have to get this working asap!


I would take a close look at the files in /tmp/amanda.
There is a sendbackup.timestamp.debug file with the exact commandline
used to perform the backup.  The gnutar command should also use
some file sendbackup._File_Sys.timestamp.{include|exclude} with the
the include or exclude list used.
Do they contain the correct command and include/exclude list?



--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  Are you sure?  ...   YES   ...   Phew ...   I'm out  *
***



Re: include problems with Amanda-2.4.4-20030428?

2003-11-25 Thread Jean-Francois Malouin
* Paul Bijnens [EMAIL PROTECTED] [20031125 11:04]:
 Jean-Francois Malouin wrote:
 
 different than yours. in /.../assembly I have a few thousands
 of subdirs, 186 of them starts with a '1'. I just calculated
 their total combined size: 11286948 KB or a little less than 11GB.
 
 OK, now look at this amanda report from last night:
 
 ---
 DUMP SUMMARY:
   DUMPER STATS  TAPER 
   STATS  HOSTNAME DISK LORIG-KB OUT-KB 
  COMP% MMM:SSKB/s 
 MMM:SSKB/s
 --- -- 
 --
 bullcalf -/nihpd1 2  44220  44220   --   29:0425.4   0:10  
 4308.6
 bullcalf -embly_0 1 120160 120160   --   22:1190.3   0:12 
 10093.6
 bullcalf -embly_1 0  100460112  100460112   --  277:34  6032.2 277:34  
 6032.2
 
 (brought to you by Amanda version 2.4.4p1-20031120)
 ---
 
 I see that //assembly_1 was around 98GB which is actually the
 entire content of /data/nihpd/nihpd1/data/assembly.
 
 The disklist for this directory:
 
 bullcalf /data/nihpd/nihpd1/data/assembly_1 
 /data/nihpd/nihpd1/data/assembly {
 include ./1*
 high-tar
 } 2
 
 hmmm, seems to me that the include thing does not behave as intended.
 Maybe I'm just missing something though.
 
 Any comments anyone? I really have to get this working asap!
 
 
 I would take a close look at the files in /tmp/amanda.
 There is a sendbackup.timestamp.debug file with the exact commandline
 used to perform the backup.  The gnutar command should also use
 some file sendbackup._File_Sys.timestamp.{include|exclude} with the
 the include or exclude list used.

I see the sendbackup._file_sys.timestamp.exclude for the other DLE
with excludes but no .include sendback file...

The relevent sendbackup file for this DLE with include stanza:

sendbackup: debug 1 pid 10655570 ruid 666 euid 666: start at Tue Nov 25 04:41:51 2003
/opt/amanda/amanda1/libexec/sendbackup: version 2.4.4p1-20031120
  parsed request as: program `GNUTAR'
 disk `/data/nihpd/nihpd1/data/assembly_1'
 device `/data/nihpd/nihpd1/data/assembly'
 level 0
 since 1970:1:1:0:0:0
 options `|;auth=bsd;index;'
sendbackup: try_socksize: send buffer size is 65536
sendbackup: time 0.003: stream_server: waiting for connection: 0.0.0.0.16689
sendbackup: time 0.003: stream_server: waiting for connection: 0.0.0.0.16690
sendbackup: time 0.004: stream_server: waiting for connection: 0.0.0.0.16691
sendbackup: time 0.004: waiting for connect on 16689, then 16690, then 16691
sendbackup: time 0.006: stream_accept: connection from 132.206.178.5.16515
sendbackup: time 0.007: stream_accept: connection from 132.206.178.5.16516
sendbackup: time 0.009: stream_accept: connection from 132.206.178.5.16517
sendbackup: time 0.009: got all connections
sendbackup-gnutar: time 0.010: doing level 0 dump as listed-incremental to 
/opt/amanda/amanda1/var/amanda/gnutar-lists/bullcalf_data_nihpd_nihpd1_data_assembly_1_0.new
sendbackup-gnutar: time 0.670: doing level 0 dump from date: 1970-01-01  0:00:00 GMT
sendbackup: time 0.745: spawning /opt/amanda/amanda1/libexec/runtar in pipeline
sendbackup: argument list: gtar --create --file - --directory 
/data/nihpd/nihpd1/data/assembly --one-file-system --listed-incremental 
/opt/amanda/amanda1/var/amanda/gnutar-lists/bullcalf_data_nihpd_nihpd1_data_assembly_1_0.new
 --sparse --ignore-failed-read --totals .
sendbackup: time 0.749: started index creator: /usr/freeware/bin/tar -tf - 
2/dev/null | sed -e 's/^\.//'
sendbackup-gnutar: time 0.749: /opt/amanda/amanda1/libexec/runtar: pid 10743130
sendbackup: time 16652.384: index created successfully
sendbackup: time 16652.971:  53:size(|): Total bytes written: 102871152640 (96GB, 
5.9MB/s)
sendbackup: time 16653.951: pid 10655570 finish time Tue Nov 25 09:19:25 2003

 
 Do they contain the correct command and include/exclude list?

as seen from the above, no.


regards,
jf

 
 
 
 -- 
 Paul Bijnens, XplanationTel  +32 16 397.511
 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
 http://www.xplanation.com/  email:  [EMAIL PROTECTED]
 ***
 * I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
 * quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
 * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
 * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
 * kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
 * ...  Are you sure?  ...   YES   ...   Phew ...   I'm out  *
 ***
 

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-24 Thread Jean-Francois Malouin
Hello again,

Oups, forgot to mention one important fact: the gnutar version:

/usr/freeware/bin/tar --version
tar (GNU tar) 1.13.25

* Jean-Francois Malouin [EMAIL PROTECTED] [20031124 12:00]:
 Hi,
 
 I'm having strange problems with the include thingy in an
 entry in my disklist:
 
 bullcalf /data/nihpd/nihpd1/data/assembly_0 /data/nihpd/nihpd1/data/assembly {
 include ./0*
 exclude ./[1-9]*
 high-tar
 } 1
 
 Looking at what got written to tape (a full backup) I see only
 directories (no files at all!), and even from the exclude part...ie
 just the hierarchy structure of /data/nihpd/nihpd1/data/assembly all
 the way down...
 
 I'm compiling 2.4.4p1-20031120 right now.
 But before putting that online, does anyone knows about this
 problem, and if so, will 2.4.4p1-20031120 do the trick?
 
 TIA,
 jf
 -- 
 We see in what we think we fear
 The clouding of our thoughts made clear

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-24 Thread Jean-Francois Malouin
* Paul Bijnens [EMAIL PROTECTED] [20031124 13:02]:
 Jean-Francois Malouin wrote:
 
 Hi,
 
 I'm having strange problems with the include thingy in an
 entry in my disklist:
 
 bullcalf /data/nihpd/nihpd1/data/assembly_0 
 /data/nihpd/nihpd1/data/assembly {
 include ./0*
 exclude ./[1-9]*
 high-tar
 } 1
 
 Looking at what got written to tape (a full backup) I see only
 directories (no files at all!), and even from the exclude part...ie
 just the hierarchy structure of /data/nihpd/nihpd1/data/assembly all
 the way down...
 
 
 Can you explain a little more? Show the relevant index.
 The index file itself starts with the directories, and then follows
 with the files in the directories.  But you shouldn't see anything
 from the excluded parts.  (Actually the exclude above can be omitted
 entirely.)
 

I didn't have a look at the index file, but here's an excerpt:

top:

/
/-/
/-/0/
/-/0/behavioural/
/-/0/behavioural/cantab/
/-/0/behavioural/carey/
/-/0/behavioural/cbcl/
/-/0/behavioural/cvlt2/
/-/0/behavioural/cvltc/

end:

/998601/1/mri/work/xfms/nihpd_998601_v1_obj1_t2w_1.t2w.clp.t1tal.xfm
/998601/1/mri/work/xfms/nihpd_998601_v1_obj1_t2w_1.t2w.clpt2t1native.xfm
/998724/1/behavioural/cbcl/998724v1_profexport.DBF
/998724/1/behavioural/cbcl/998724v1_profexport.txt
/998724/1/behavioural/cbcl/998724v1_rawexport.DBF
/998724/1/behavioural/cbcl/998724v1_rawexport.txt
/998724/1/behavioural/dps4/998724v1y.chr

So you're right, there are files in there. My mistake. I probably
looked at the wrong place. However the index entries start with a /
NOT ./ so is this the problem? That would explain why the
include/exclude pattern failed as the directories starting with
something else that ./0 should be excluded , right?

Looking at what is on the tape I see :

mt -f /hw/tape/tps12d1nrnsv fsf 1
dd if=/hw/tape/tps12d1nrnsv bs=32k skip=1 | /usr/freeware/bin/tar tf
./
./-/
./-/0/
./-/0/behavioural/
./-/0/behavioural/cantab/
./-/0/behavioural/carey/
./-/0/behavioural/cbcl/
./-/0/behavioural/cvlt2/
./-/0/behavioural/cvltc/

something's fishy...


 I'm compiling 2.4.4p1-20031120 right now.
 But before putting that online, does anyone knows about this
 problem, and if so, will 2.4.4p1-20031120 do the trick?
 
 There was a bug in 2.4.4p1 that if the include was not a pattern, it
 would not work.  That was only for include.  I believe the above
 should work find, even in a plain 2.4.4.
 Stupid question: the client has the same/correct version too?

client is the server, and the only one for this conf.

regards,
jf

 
 -- 
 Paul Bijnens, XplanationTel  +32 16 397.511
 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
 http://www.xplanation.com/  email:  [EMAIL PROTECTED]
 ***
 * I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
 * quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
 * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
 * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
 * kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
 * ...  Are you sure?  ...   YES   ...   Phew ...   I'm out  *
 ***
 

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-24 Thread Jon LaBadie
On Mon, Nov 24, 2003 at 11:59:41AM -0500, Jean-Francois Malouin wrote:
 Hi,
 
 I'm having strange problems with the include thingy in an
 entry in my disklist:
 
 bullcalf /data/nihpd/nihpd1/data/assembly_0 /data/nihpd/nihpd1/data/assembly {
 include ./0*
 exclude ./[1-9]*
 high-tar
 } 1
 
 Looking at what got written to tape (a full backup) I see only
 directories (no files at all!), and even from the exclude part...ie
 just the hierarchy structure of /data/nihpd/nihpd1/data/assembly all
 the way down...
 

Don't know about your bad backup, but in your DLE you should not
have both the include and THAT exclude.  I.e., the include will
automatically drop any toplevel dir that does not match ./0*.

I think the only valid exclude usage would be for things under
the 0* directories.  For example 'exclude ./0*/tmp'.  Though
I must admit I've never tried that combination.

-- 
Jon H. LaBadie  [EMAIL PROTECTED]
 JG Computing
 4455 Province Line Road(609) 252-0159
 Princeton, NJ  08540-4322  (609) 683-7220 (fax)


Re: include problems with Amanda-2.4.4-20030428?

2003-11-24 Thread Jean-Francois Malouin
* Jon LaBadie [EMAIL PROTECTED] [20031124 14:15]:
 On Mon, Nov 24, 2003 at 11:59:41AM -0500, Jean-Francois Malouin wrote:
  Hi,
  
  I'm having strange problems with the include thingy in an
  entry in my disklist:
  
  bullcalf /data/nihpd/nihpd1/data/assembly_0 /data/nihpd/nihpd1/data/assembly {
  include ./0*
  exclude ./[1-9]*
  high-tar
  } 1
  
  Looking at what got written to tape (a full backup) I see only
  directories (no files at all!), and even from the exclude part...ie
  just the hierarchy structure of /data/nihpd/nihpd1/data/assembly all
  the way down...
  
 
 Don't know about your bad backup, but in your DLE you should not
 have both the include and THAT exclude.  I.e., the include will
 automatically drop any toplevel dir that does not match ./0*.

That is want I want!
All subdirs in there have a pattern [0-9]\{6\} , ie, 6 digits. There
are other subdirs but there are not to be backed up. You see, my
problem is that /data/nihpd/nihpd1/data/assembly can contains 100's of
GB and I'm using 4 LTO tape drives, so I have to split that thing in
small chuncks...

Eventually I'll have 10 DLE's, those starting with ./0* ./1* ...
you get the picture.

 
 I think the only valid exclude usage would be for things under
 the 0* directories.  For example 'exclude ./0*/tmp'.  Though
 I must admit I've never tried that combination.

Have a look at my other post in this thread. The index file for that
DLE doesn't start with ./ but with /...Is my gnutar at fault here?
I thought that v1.13.25 had good karma...

thanks for the help,
jf

 
 -- 
 Jon H. LaBadie  [EMAIL PROTECTED]
  JG Computing
  4455 Province Line Road(609) 252-0159
  Princeton, NJ  08540-4322  (609) 683-7220 (fax)

-- 
We see in what we think we fear
The clouding of our thoughts made clear


Re: include problems with Amanda-2.4.4-20030428?

2003-11-24 Thread Jon LaBadie
On Mon, Nov 24, 2003 at 02:57:56PM -0500, Jean-Francois Malouin wrote:
 * Jon LaBadie [EMAIL PROTECTED] [20031124 14:15]:
  On Mon, Nov 24, 2003 at 11:59:41AM -0500, Jean-Francois Malouin wrote:
   Hi,
   
   I'm having strange problems with the include thingy in an
   entry in my disklist:
   
   bullcalf /data/nihpd/nihpd1/data/assembly_0 /data/nihpd/nihpd1/data/assembly {
   include ./0*
   exclude ./[1-9]*
   high-tar
   } 1
   
   Looking at what got written to tape (a full backup) I see only
   directories (no files at all!), and even from the exclude part...ie
   just the hierarchy structure of /data/nihpd/nihpd1/data/assembly all
   the way down...
   
  
  Don't know about your bad backup, but in your DLE you should not
  have both the include and THAT exclude.  I.e., the include will
  automatically drop any toplevel dir that does not match ./0*.
 
 That is want I want!
 All subdirs in there have a pattern [0-9]\{6\} , ie, 6 digits. There
 are other subdirs but there are not to be backed up. You see, my
 problem is that /data/nihpd/nihpd1/data/assembly can contains 100's of
 GB and I'm using 4 LTO tape drives, so I have to split that thing in
 small chuncks...
 
 Eventually I'll have 10 DLE's, those starting with ./0* ./1* ...
 you get the picture.
 
  
  I think the only valid exclude usage would be for things under
  the 0* directories.  For example 'exclude ./0*/tmp'.  Though
  I must admit I've never tried that combination.
 
 Have a look at my other post in this thread. The index file for that
 DLE doesn't start with ./ but with /...Is my gnutar at fault here?
 I thought that v1.13.25 had good karma...

No, the index should start with /,
but it should not start with /-/
unless you have a top-level directory named -.

-- 
Jon H. LaBadie  [EMAIL PROTECTED]
 JG Computing
 4455 Province Line Road(609) 252-0159
 Princeton, NJ  08540-4322  (609) 683-7220 (fax)


Re: include problems with Amanda-2.4.4-20030428?

2003-11-24 Thread Jean-Francois Malouin
* Jon LaBadie [EMAIL PROTECTED] [20031124 15:23]:
 On Mon, Nov 24, 2003 at 02:57:56PM -0500, Jean-Francois Malouin wrote:
  * Jon LaBadie [EMAIL PROTECTED] [20031124 14:15]:
   On Mon, Nov 24, 2003 at 11:59:41AM -0500, Jean-Francois Malouin wrote:
Hi,

I'm having strange problems with the include thingy in an
entry in my disklist:

bullcalf /data/nihpd/nihpd1/data/assembly_0 /data/nihpd/nihpd1/data/assembly {
include ./0*
exclude ./[1-9]*
high-tar
} 1

Looking at what got written to tape (a full backup) I see only
directories (no files at all!), and even from the exclude part...ie
just the hierarchy structure of /data/nihpd/nihpd1/data/assembly all
the way down...

   
   Don't know about your bad backup, but in your DLE you should not
   have both the include and THAT exclude.  I.e., the include will
   automatically drop any toplevel dir that does not match ./0*.
  
  That is want I want!
  All subdirs in there have a pattern [0-9]\{6\} , ie, 6 digits. There
  are other subdirs but there are not to be backed up. You see, my
  problem is that /data/nihpd/nihpd1/data/assembly can contains 100's of
  GB and I'm using 4 LTO tape drives, so I have to split that thing in
  small chuncks...
  
  Eventually I'll have 10 DLE's, those starting with ./0* ./1* ...
  you get the picture.
  
   
   I think the only valid exclude usage would be for things under
   the 0* directories.  For example 'exclude ./0*/tmp'.  Though
   I must admit I've never tried that combination.
  
  Have a look at my other post in this thread. The index file for that
  DLE doesn't start with ./ but with /...Is my gnutar at fault here?
  I thought that v1.13.25 had good karma...
 
 No, the index should start with /,
 but it should not start with /-/
 unless you have a top-level directory named -.

yes, there is a directory ./-
Most prob'ly a luser typo :)

jf

 
 -- 
 Jon H. LaBadie  [EMAIL PROTECTED]
  JG Computing
  4455 Province Line Road(609) 252-0159
  Princeton, NJ  08540-4322  (609) 683-7220 (fax)

-- 
We see in what we think we fear
The clouding of our thoughts made clear