Re: problem to use smbclient (Samba)

2002-01-15 Thread Pierre Volcke


answering my own question:

> 
> * Amanda version : 2.4.2p2, 
> compiled with --with-smb-client=/full/path/...
> 

I tried to recompile again, and checked the
"configure" output:

$ ./configure ... --with-smblient=/usr/local/samba/bin/smbclient ...

(...)
checking for smbclient... no
(...)

but, 

$ ls -al /usr/local/samba/bin/smbclient

-rwxr-xr-x1 root root   574817 Nov 22 16:01 smbclient

so, I guess configure doesn't parse the full path I give to him,
but tries a default one instead.
I then made a link to smbclient into /usr/bin (which is into the $PATH).

Then, "configure" works:

...
checking for smbclient... /usr/bin/smbclient
...

and "amcheck" works for the NT box after recompiling Amanda..

So, is there some mistake into the doc or into 
the configure command, or did I miss something ?

thanks,
Pierre





problem to use smbclient (Samba)

2002-01-15 Thread Pierre Volcke


hello,

I encounter a trouble when trying to configure
my Amanda server to backup NT hosts through
the Samba suite:

* Amanda index/tape server: Linux RH7.1 
(the server is also an Amanda client)

* Amanda version : 2.4.2p2, 
compiled with --with-smb-client=/full/path/...

* Samba version : 2.2

* amanda user : bin.disk

in the following dump for amcheck, 
the Amanda index/tape server is called "myServer", the NT box
is called "thePC". The amanda config is called "pc".

###

[root@myServer /usr]# su -c "/usr/local/amanda/sbin/amcheck -c -s pc" bin

Amanda Tape Server Host Check
-
... (OK)

Amanda Backup Client Hosts Check

ERROR: myServer: [The client is not configured for samba://thePC/sharename]
ERROR: myServer: [SMBCLIENT program not available]
Client check: 1 host checked in 0.016 seconds, 2 problems found

(brought to you by Amanda 2.4.2)

###

/etc/amandapass is:

//myPC/sharenameadministrateur%theRightPassword

###

disklist for "pc" is:

myServer //myPC/sharename   pc_settings

#

I must also note that the share is reachable
via the smbclient command used as a shell command,
and file permissions for Amanda user seems correct.

...but amcheck keeps sending this error!

May anyone give me some help? Did I forget something
when building Amanda?..

thanks,

Pierre





Re: Amrecover failure.

2001-05-07 Thread Pierre Volcke


> [root@ /root]# amrecover -C Normal
> AMRECOVER Version 2.4.2p1. Contacting server on  ...
> 220  AMANDA index server (2.4.2p1) ready.
> 500 Access not allowed: [access as amanda not allowed from root@]
> amandahostsauth failed
> [root@ /root]#

 In my ".amandahosts" file, I added the following line to make
 it possible for the "root" user to run "amrecover", thus connecting 
 to the tape/index server :

from_machine1.allow.com root
from_machine2.allow.com root
...

These lines come after the authorization line for the amanda user 
itself, "bin", in my case. e.g:

from_machine1.allow.com bin
...

I guess that it doesn't raise any security matter, provided that
the machine names are written as full qualified domain name,
and that amandahosts authentication is used , and last, that
amanda inetd services are tcp-wrapped.

Pierre.



Re: trouble with amrecover (gtar in fact)

2001-05-05 Thread Pierre Volcke


"John R. Jackson" wrote:
> 
> >here is a sample of the gnutar list for 1.13 (looks bad) :
> >
> >2555905 43008 ./kretzsch/mail/communication_OBS
> >2555905 40960 ./kretzsch/mail/SOHO-news
> 
> That's the wrong file.  I'm talking about the index files.  Those look
> like the incremental list file.  Look up the indexdir value from
> amanda.conf and check down in there.
> 

John, here is a sample of the right index files, with the big 
numbers you were talking about, I guess:

07274015541/./
07274015536/./_amanda/
07274015536/./_amanda/amanda-2.4.2/
07274015536/./_amanda/amanda-2.4.2/amplot/
07274015536/./_amanda/amanda-2.4.2/changer-src/ 

setting up gnutar 1.13.19 on both my server & clients seems to 
have solved the problem, that is, amrecover not showing all the 
backed up files.

Pierre.



Re: [amrestore] file too large

2001-05-04 Thread Pierre Volcke


> 
> On May  3, 2001, Pierre Volcke <[EMAIL PROTECTED]> wrote:
> 
> > amrestore:   4: restoring planeto._home_zone10.20010502.0
> > amrestore: write error: File too large
>

Alexandre Oliva wrote:
> 
> You seem to be bumping into file size limits of the host filesystem.
> You may want to pipe the output of amrestore -p directly to tar, or to
> bsplit.
> 

Bort Paul wrote:
> Perhaps your operating system cannot handle files over 2Gb? Most Linuxes
> can't. I don't know about *BSD or Solaris/SunOS.
>

 that seems to be the problem : the host is a Linux RH7, with kernel
2.2.16.
 
 thanks,

Pierre.



Re: trouble (?) with amrecover (gtar in fact)

2001-05-04 Thread Pierre Volcke


"John R. Jackson" wrote:
> 
> > Scanning /the/holdingdisk ...
> >  20010328: found Amanda directory. (*)
> >  20010416: found Amanda directory.
> >...
> > (*) these directories are empty.
> 
> Just for my own curiosity, why are they still there if they are empty?
> 

 I don't know why exactly, sorry. According to your question, I 
 shouldn't wait for them to be deleted automatically!
 They look rather old and I was maybe performing some tests that 
 failed at this time. 

>
> > the 'ls' command only shows me a very small part
> > of what is supposed to be backed up, I mean only a
> > few directories & files out of hundreds.
> 
> I take it you're using GNU tar?  What version?  If it's older than
> 1.13.17 (even better, 1.13.19), the index files may be corrupt.
> 

 that's a good guess, on my client machine (AIX) this is gtar 1.13 !

>
> Go to your index directory and look at one of the files.  If each line
> has a very large number on it, that's a symptom of a broken tar.
>

here is a sample of the gnutar list for 1.13 (looks bad) :

2555905 43008 ./kretzsch/mail/communication_OBS
2555905 40960 ./kretzsch/mail/SOHO-news

and a sample of gnutar list for 1.13.17 on my tape server (looks
better!):

2070 407572 ./doute/Missions_inst/NIMS/Data
2070 407569 ./doute/Missions_inst/NIMS/Tools

so the problem is now pointed out, thanks!

now it involves another question : 
Is there any document dealing with the advantages/disadvantages
of using gtar or the system's dump/restore?

I've read in the Amanda docs that 'gtar' provides me with 
exclude-lists,  but 'restore'  has an interactive restoration mode 
(restore -i) that may  be useful when using it with amrestore. 

I also believe that a gtar archive could be extracted from nearly 
any platform, but a classic dump archive won't because it's 
OS specific. (true/false?)

thanks again for helping,

Pierre.

* Laboratoire de Planetologie de Grenoble   
* http://lpg.obs.ujf-grenoble.fr
* Bat D de Physique, BP53, 38041 Grenoble Cedex 9 /
FRA
* tel. 04.76.51.41.55 , fax. 04.76.51.41.46
* e-mail. [EMAIL PROTECTED]



[amrestore] file too large

2001-05-03 Thread Pierre Volcke


 hello,

 I get the following message when restoring one of my
 full dumps with amrestore:

[tape-serveur]# /usr/local/amanda/sbin/amrestore /dev/nst0 planeto
amrestore:   0: skipping start of tape: date 20010502 label TAPE4
amrestore:   1: skipping jupiter._home_zone1.20010502.2
amrestore:   2: skipping jupiter._home_zone2.20010502.1
amrestore:   3: skipping jupiter._home_zone3.20010502.0
amrestore:   4: restoring planeto._home_zone10.20010502.0
amrestore: write error: File too large

The target disk is big enough to handle the whole taped archive.

According to the mail reports, the amount of datas saved 
on this tape is 17.8Gb. I use 20Gb/40Gb DLT Tapes.

Could it be a problem with my tapetype config or with
the tape hardware itself?

Of course, a "tar xvf" on this archive will fail at the
end of file:

[tape-server]#cat planeto._home_zone10.20010502.0 | tar xvf - 
...
tar: 511 garbage bytes ignored at end of archive
tar: Unexpected EOF in archive
tar: Error exit delayed from previous errors
...

any advice? 
thanks!

Pierre.



trouble (?) with amrecover

2001-05-03 Thread Pierre Volcke


 hello again,

 I'm trying to am-recover some datas, and I 
 get the following problem when I run amrecover :

 [tape-server]$ amrecover -C conf -s tape-server

 amrecover> sethost backup-client
 amrecover> setdisk /the/disk
 Scanning /the/holdingdisk ...
  20010328: found Amanda directory. (*)
  20010416: found Amanda directory.
 200 Disk set to /the/disk
 amrecover> ls

 (*) these directories are empty.

 problem:

 the 'ls' command only shows me a very small part 
 of what is supposed to be backed up, I mean only a 
 few directories & files out of hundreds.

 more details:

 - the directories & files that are shown do not 
 seem to have something special or common attributes.

 - amanda report didn't show anything wrong,
 it even shows a good full backup yesterday.

 - problem is the same if I run amrecover from 
 the backup client

 - the backup server also backs up some disks that it owns;
 I get no trouble when using amrecover on these disks.

 - I think that the 'hidden' files are indeed on the tape:
 the gnutar-lists on the client do show them fully,
 and the following command :

 [tape-server]$ ./amadmin conf info backup-client /the/disk 

  Dumps: lev datestmp  tape file   origK   compK secs
  0  20010502  TAPE4   4 2172930 2172960 1666
  1  20010503  TAPE5   3   24980   24992   20

 ... shows well-filled tape for level 0.
 

  So, why is amrecover hiding files?

 thanks,

Pierre.



setting up multi configurations

2001-05-03 Thread Pierre Volcke


 hello,

 I'm configuring my tape server to run different configurations
 (e.g., daily backups and periodic archival runs on the same
 filesystems, plus different configurations on separate filesystems) 

 The FAQ says to set the "record no" option for archival runs : 
 I guess it has no effect on the "amrecover" features and 
 capabilities, provided that the "index yes" option is set 
 in each configurations. True or false?!

 Would you also recommend to separate the "indexdir" & "infofile" 
 directories for each configurations, or should I leave them
 into the same subdirectories ?

thanks
Pierre.



Re: gnutar & client configuration

2001-05-01 Thread Pierre Volcke


Mitch Collinsworth wrote:
> 
> On Tue, 1 May 2001, Pierre Volcke wrote:
> 
> > checking for gnutar... (cached) tar
> > ...
> >
> > (what does this mean?)
> 
> This means you didn't rm config.cache before running configure, so
> configure used info saved in config.cache from the last time you
> ran it.  You should rm config.cache and run configure again.
> 
> -Mitch

 oops.. now it looks much more easier!
 thanks for the advice,

Pierre.



gnutar & client configuration

2001-05-01 Thread Pierre Volcke


 hello,

 I've just installed an Amanda backup client on an AIX 4.3 box.

 amanda version : 2.4.2

 # client side  #

 gnu-tar has been installed on this client in

 /usr/local/bin/gtar

 and to be quite sure, I've also made a link as

 /bin/gtar -> /usr/local/bin/gtar

 On this client, Amanda was compiled with the following options:

./configure --prefix=/usr/local/amanda
--without-server   
--with-gnutar=/usr/local/bin/gtar
...

the configure script then shows the following message:

...
checking for gnutar... (cached) tar
...

(what does this mean?)

 # tape server side 

 On the server (Redhat7), the dumptype section
 for the AIX client says :

define dumptype ... {
...
program "GNUTAR"
...
}
 
when trying a check witch amcheck, here is what I get:

sh$ ./amcheck -c daily1
 
Amanda Backup Client Hosts Check

ERROR: [backup-client-name]: [GNUTAR program not available]
...


##

 There must be something that I didn't understand!
 Does someone have an idea?

thanks,
Pierre.



Re: I was a new user in linux and Amanda.

2001-03-21 Thread Pierre Volcke

CHAN Sing Leung Cyril wrote:
> 
> Hi, I want to install the Amanda 2.4.2 in my new Mandrake 7.2 Linux
> server. But, I don't know how to start. When I reading the
> installation notes. sorry~ I don't understand the procedure. Can you
> recommand another document or web site for me to solve this problem.
> Please reply my message asap! Thanks! Regards Cyril -- Get your free
> email from www.linuxmail.org
> Powered by Outblaze

you may have a look to 

http://www.backupcentral.com/amanda.html

.. for more documentation,

or maybe use a pre-compiled package (such as RPM, but I don't know
if it's available on Mandrake).

hope it helps
Pierre.



amanda & NFS filesystems

2001-03-19 Thread Pierre Volcke


 hello,

 my Amanda backup server mounts some NFS filesystems
 exported from machines that are in fact
 Amanda backup clients.
 
 Should I better install the Amanda client software
 on the backup machines, or may I directly
 make a reference to the NFS filesystems into the 
 disklist as if it was a local filesystem?

thanks,
Pierre.



Re: about amRecover

2001-03-01 Thread Pierre Volcke


 hello John & all,

> 
> >my xinetd.conf for amandaidx & amidxtape is:
> >
> >service amandaidx
> >{
> >...
> >server = /usr/local/amanda/libexec/amindexd
> >server_args = amindexd
> >}
> 
> I don't think you want any server_args.  Note:
> 
> >xinetd state-dump gives:
> >
> >Service = amandaidx
> >...
> >Server = /usr/local/amanda/libexec/amindexd
> >Server argv = amindexd amindexd
> 
> See the two "amindexd" entries?  That looks to me like it's confusing
> amindexd into thinking you want it to work with a configuration named
> "amindexd".
> 

 absolutely! I sometime get blind, I guess ;)
 it now works fine without "server_args" keyword into xinetd.conf

thanks to Martin, Gerhard & you for help,
Pierre.



Re: about amRecover

2001-02-28 Thread Pierre Volcke


> >> did you check /tmp/amanda/amrecover.*.debug ?
> >
> > amrecover.debug shows only 1 line:
> > amrecover: debug 1 pid 16190 ruid 0 euid 0 start time Wed Feb 28
> > 12:13:55 2001
> 
> And are you running this on the server ?
> if not, you have to specify the server
> amrecover -C daily1 -s mytapeserver

 yes, I'm running all of this (tape server + client host) on
 the same machine.

> 
> If you are, you should also have an amindexd.*.debug file
> What's in that one ?
> 

 it looks like amrecover.debug:
   
amindexd: debug 1 pid 16662 ruid 1 euid 1 start time Wed Feb 28 14:12:40
2001
amindexd: version 2.4.2

I've just checked file/dir permissions on /usr/adm/amanda/index,
it looks ok. Log files do not show tcp-wrapper problems, or
failed connection attempts.

I "feel" I'm having trouble with xinetd.. I'm using xinetd
version 2.1.8.9pre11 on a Redhat7 box. Any trouble known
with that config?

regards,
Pierre.



Re: about amRecover + empy mail report

2001-02-28 Thread Pierre Volcke


> > When I try to run 'amrecover' to check file index &
> > recovery, here's what it says:
> 
> >> bash$ /usr/local/amanda/sbin/amrecover -C daily1
> >> AMRECOVER Version 2.4.2. Contacting server on *
> >> amrecover: Unexpected server end of file

>
> did you check /tmp/amanda/amrecover.*.debug ?
>

amrecover.debug shows only 1 line:
amrecover: debug 1 pid 16190 ruid 0 euid 0 start time Wed Feb 28
12:13:55 2001

>
> Do you have indexing turned on in your dumptype ?
>

 yes, my dumptype look like this:

define dumptype default_settings {
... 
index yes  
record yes 
...
}   

define dumptype daily1_settings {
default_settings
...
}  

 I've checked the /usr/adm/amanda/index//
 directory, I've found some .gz files that contains the correct 
 informations. So the indexes seems to be recorded.

 /etc/dumpdates has zero size. I don't know if it's normal.

 So it looks like some network config problem :

>
> Please check if 
> amandaidx stream  tcp nowait  root/usr/local/libexec/amindexd 
> amindexd
>
> amidxtape stream  tcp nowait  root /usr/local/libexec/amidxtaped 
> amidxtaped
>
> exist in /etc/inetd.conf and that inetd is restarted.
>

let's check my network config:

my /etc/services contains the following lines:

> amanda  10080/udp   
> kamanda 10081/tcp   
> kamanda 10081/udp   
> amandaidx   10082/tcp  
> amidxtape   10083/tcp   

my xinetd.conf for amandaidx & amidxtape is:

service amandaidx
{
socket_type = stream
protocol = tcp
wait = no
user = bin
group = disk
server = /usr/local/amanda/libexec/amindexd
server_args = amindexd
} 

service amidxtape
{
socket_type = stream
protocol = tcp
wait = no
user = bin
group = disk
server = /usr/local/amanda/libexec/amidxtaped
server_args = amidxtaped
} 

(bin is my amanda-user)

"netstat -a | grep am"  gives:

>tcp0  0 *:amandaidx *:* LISTEN
>tcp0  0 *:amidxtape *:* LISTEN
>udp0  0 *:amanda*:* 

those services are host-allowed from xinetd.conf & also 
from  /etc/hosts.allow.

xinetd state-dump gives:

Service = amidxtape
State = Active
Service configuration: amidxtape
id = amidxtape
socket_type = stream
Protocol (name,number) = (tcp,6)
Groups = 0
Server = /usr/local/amanda/libexec/amidxtaped
Server argv = amidxtaped amidxtaped
...
running servers = 0
retry servers = 0
attempts = 0
...

Service = amandaidx
State = Active
Service configuration: amandaidx
id = amandaidx
socket_type = stream
Protocol (name,number) = (tcp,6)
Groups = 0
Server = /usr/local/amanda/libexec/amindexd
Server argv = amindexd amindexd
...
running servers = 0
retry servers = 0
attempts = 0
...


Network config seems ok to me, but maybe I've got 
trapped somewhere?..

>
> amcheck sent me an e-mail report with correct
> subject header, but with empty body. 
>

 found the problem : I used a ";" instead of a "," for 
 email adresses separator. 

thanks for your help,

Pierre.



about amRecover + empy mail report

2001-02-28 Thread Pierre Volcke


hello,

this question was already posted yesterday, but I 
got no answer! ;)

1/

When I try to run 'amrecover' to check file index &
recovery, here's what it says:
 
> bash$ /usr/local/amanda/sbin/amrecover -C daily1
> AMRECOVER Version 2.4.2. Contacting server on *
> amrecover: Unexpected server end of file
 
(real server name replaced with *)
 
  what's the trouble?

2/ 

amcheck sent me an e-mail report with correct
subject header, but with empty body. 
Is it a known problem? 

thanks,
Pierre.



question about amRecover

2001-02-27 Thread Pierre Volcke


 hi there,

 I'm currently testing my Amanda conf.
 This night, a first 'amdump' built a first full dump
 of 1 filesystem.
 When I try to run 'amrecover' to check file index & 
 recovery, here's what it says:

> bash$ /usr/local/amanda/sbin/amrecover -C daily1
> AMRECOVER Version 2.4.2. Contacting server on *
> amrecover: Unexpected server end of file   

 (real server name replaced with *)

 what does this mean?

note.
an 'amoverview' run shows a correct calendar,
showing 1 full-dumped filesystem.

thanks,

Pierre.



Re: ./configure - error

2001-02-26 Thread Pierre Volcke


> 
> When I try to ./configure amanda I get an error with
> 
> *** --with-user=USER missing
> 
> whats that?
> 

you have to specify an user name and group for Amanda 
to compile and run, for example:

./configure --with-user=bin --with-group=disk [...]

have a look at Amanda installation notes, section 1.1.F

Pierre.



Re: setting compression 'off' on a DLT4000

2000-12-13 Thread Pierre Volcke

Matt McManus wrote:
> 
> I have now got amanda running very happily on my network. I have a query
> about how to turn hardware compression 'off' on my Quantum DLT4000 drive.
> At the moment I use the 'density select' button on the front of the drive,
> but I understand that I am able to do this with the 'tapedev'.
> 
> At the moment my tapedev is "/dev/nst0"
> 
> Will the following work to disable compression: "/dev/nst0 compression 0"
> 
> thanks for the help,
> 
> matt
> 

 hi Matt,

 if I remember well , I used the datcompression flag with my QT4000

 mt -f /dev/nst0 datcompression on

 (this can be checked with the compression led on the tape device)

 ..but this option works depending on other flags, like "compression"
 and "defcompression". Have a look at the "mt" man page to get some
 more details.

Pierre.



Re: No-rewind tape device on Linux Redhat...?

2000-12-05 Thread Pierre Volcke


 hello Isaac,

 this is an example, but if your auto-rewind device is /dev/st0,
 the non-rewind device is /dev/nst0.

 If you want to crontol hardware compression on/off , you may
 use the "mt" command (have a look at the man page)

for example:

mt -f /dev/st0 datcompression off

.. but compression control may depend on your tape device 
   SCSI implementation.

Pierre.


> 
>  Hello all,
> 
>  I am fairly new to Linux Administration, new enough to not know very much
> about tape drives at all. I have a Sony DAT SCSI tape drive that uses 4
> GIG/8 GIG with hardware compression tapes. I am trying to set it up with
> AMANDA but I have no idea which device is the no-rewind tape device and
> not even the faintest clue about how to tell if hardware compression is
> enabled or disabled.
> 
>  Is there anyone out there that can guide me to any resources that will
> help me out, or better yet tell me how to know what tape devices are what
> in Red Hat Linux release 6.1 (Cartman) Kernel 2.2.12-20 on an i586? Any
> help will be greatly appreciated.
> 
>  Thanks for your time
>  Isaac D. Flemming



Re: General question

2000-11-22 Thread Pierre Volcke


> 
> Can Amanda be used to backup up a Windows Machine, if so when the file is
> restored will you have to install Windows first?

 yes, Amanda can do that but you have to setup the Samba package first.
 See Samba docs about this.

 I guess that the files may be restored anywhere if the Windows client
 is not up anymore.

Pierre



Re: Windows client

2000-11-06 Thread Pierre Volcke


> 
> I need to back up a Windows pc with Amanda.
> I didn't find appropriate documentation about "how configure amanda to
> backup Windows client".
> Could someone help me?
> Tank you very much.
> 
> Alessandro Chiauzzi
> System Engineer
> Dialogos3 s.r.l.

 You may use both Samba & Amanda to backup your Win client.
 Have a look into the Amanda documentation, there's a
 file named "SAMBA".

Pierre.



Amanda + smbclient: about the archive bit

2000-11-03 Thread Pierre Volcke


 hello there,

 I'm backing up (full & incremental) my windows PCs 
 using Amanda & Samba's "smbclient".
 

 Smbclient unset the archive bit on all the backed up
 files onto the PC hardisk, so as to make incremental 
 backups possible.

 Does someone have heard about any conflicts with 
 some Windows programs that may use the archive bit
 for other matters? Any feedback helpful!

thanks,

Pierre.



amanda-user access problem

2000-10-31 Thread Pierre Volcke


 hello,

 I meet a problem while configuring my backup server
 (lets call it 'theserver')
 note : 'theserver' is also an amanda client.

 * amanda is compiled with the "--with-amandahosts" flag

 * my amanda user is 'bin' ('disk' group), 

 * the ~bin/.amandahosts file contains
   a line with "theserver"

 when I try an "amcheck -c ",
 the server will first try to check the client
 connection to himself, and returns the 
 following message:

Amanda Backup Client Hosts Check

ERROR: theserver: [access as bin not allowed from root@theserver]

 All permissions seems ok on my server, and rsh/rlogin
 connection is allowed.
 Does someone have an idea of what's happening?

thanks,

Pierre.