Re: [BackupPC-users] backuppc & quotas

2009-01-08 Thread Les Mikesell
cedric briner wrote:
> 
>> However, any per-client size checking is unrealistically simple because 
>> the server-side space that will be used depends very much on the 
>> uniqueness of the client's files.  That is, if you add a backup of a 
> Who cares that files could be already backup in the pool ? Not the 
> enduser, the enduser only cares about saving his data and being able to 
> get them back. And you as the sysadmin, are a bit more happy, because 
> you have a way to control how much data will be backed up. Which will 
> help you to evaluate your backup system.

I'd expect both of you to care - and for your goals not to be different. 
  It may affect the amount you can back up by orders of magnitude.

> The main problem with my situation is that people have huge amount of 
> data stored, and that we only want to provide backup for important ones. 
> So we let the user to tell backuppc which directory to backup, and each 
> a year we said, we want people beeing able to save 10Gb or 100Gb.
> 
>> large directory where the files are all copies of things already backed 
>> up from other machines it will take very little additional server space.
> yes, but as already said, such things is a nice feature for the sysadmin 
> and not for the enduser.

If you don't account for the space sharing factor (and compression if 
you use it) you can't realistically pick a size to permit.  And things 
can swing wildly either direction.  While files duplicated among users 
take no additional space per instance, files that change even slightly 
between each backup run cause a complete new copy to be stored for each 
  instance you keep on line and might end up taking 10x the space you 
expected on the server.  If you are already running backuppc you can 
probably make a reasonable estimate of the average size you can hold 
but a shift in the data type/usage could throw this off by 100x.

-- 
Les Mikesell
 lesmikes...@gmail.com


--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] BackupPC_link got error -4 when calling MakeFileLink

2009-01-08 Thread tagore

Hi!

Is there a result? We have similar problem.

Thanks

+--
|This was sent by hirleve...@gmail.com via Backup Central.
|Forward SPAM to ab...@backupcentral.com.
+--



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] Latest backup removed?

2009-01-08 Thread Dejan Batic
Hi,

 

I'm using Bakcuppc for around 30 machines running XP. The backup method used
is rsyncd.

 

Lately I noticed this type of messages in LOG file:

 

2009-01-08 08:34:47 incr backup 20 complete, 36 files, 15289353 bytes, 4
xferErrs (0 bad files, 0 bad shares, 4 other)

2009-01-08 08:34:47 removing incr backup 20

.

.

.

2009-01-06 08:00:17 incr backup started back to 2009-01-02 08:00:08 for
directory docs
2009-01-06 08:10:10 incr backup started back to 2009-01-02 08:00:08 for
directory desktop
2009-01-06 08:10:18 incr backup started back to 2009-01-02 08:00:08 for
directory fav
2009-01-06 08:10:18 incr backup started back to 2009-01-02 08:00:08 for
directory firefox
2009-01-06 08:10:24 incr backup started back to 2009-01-02 08:00:08 for
directory outlook
2009-01-06 08:41:57 incr backup 19 complete, 64 files, 229429471 bytes, 4
xferErrs (0 bad files, 0 bad shares, 4 other)
2009-01-06 08:41:57 removing incr backup 19
2009-01-06 11:35:17 incr backup started back to 2009-01-02 08:00:08 for
directory docs
2009-01-06 11:42:44 incr backup started back to 2009-01-02 08:00:08 for
directory desktop
2009-01-06 11:42:52 incr backup started back to 2009-01-02 08:00:08 for
directory fav
2009-01-06 11:42:53 incr backup started back to 2009-01-02 08:00:08 for
directory firefox
2009-01-06 11:42:57 incr backup started back to 2009-01-02 08:00:08 for
directory outlook
2009-01-06 12:12:57 incr backup 19 complete, 69 files, 233432195 bytes, 4
xferErrs (0 bad files, 0 bad shares, 4 other)
2009-01-06 12:12:57 removing incr backup 19

 

 

So why the latest backup gets removed immediately after it's done?

 

Part of configuration file that I think is relevant:

 

$Conf{FullPeriod} = '6.97';

$Conf{IncrPeriod} = '0.97';

$Conf{FullKeepCnt} = [

  '3'

];

$Conf{FullKeepCntMin} = '1';

$Conf{FullAgeMax} = '90';

$Conf{IncrKeepCnit} = '6';

$Conf{IncrKeepCntMin} = '1';

$Conf{IncrAgeMax} = '90';

$Conf{IncrLevels} = [

  '1'

];

$Conf{BackupsDisable} = '0';

$Conf{PartialAgeMax} = '3';

$Conf{IncrFill} = '0';

$Conf{RestoreInfoKeepCnt} = '10';

$Conf{ArchiveInfoKeepCnt} = '10';

 

Thanks

 

Dejan

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] Large Amounts of Data

2009-01-08 Thread Chris Baker

Have the backups run over their own separate network. This will give you
more bandwidth. The backups and other jobs on the network will be separate.
Install two network cards in your BackupPC and team them. You could even
team up network cards in all the machines you are backing up.

Chris


--
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] vista backup question

2009-01-08 Thread dan
in case you were unaware, junction points = hardlinks on ntfs.

Just exclude all of the WinXP directories like "Documents and Settings" to
solve this.  Vista users the "Users" directory instead..

On Wed, Jan 7, 2009 at 10:32 PM, Cody Dunne  wrote:

> Hi Mark,
>
> Mark Maciolek wrote:
> > hi,
> >
> > Running 3.1.0 on CentOS 5.2 with 40 clients so far mostly Linux systems.
> > Added my first Vista system today. Used Deltacopy to install rsync.
> >
> > The backup worked but still had 13446 error transfer, mainly file name
> > too long.
> >
> > Remote[1]: rsync: readlink_stat("All Users/Application Data/Application
> > Data/Application Data/Application Data/Application Data/Application
> > Data/Application Data/Application Data/Application Data/...
> >
> >
> > Basically it is repeating the Application Data over and over again.
> >
> > Has anyone seen this issue and know how to prevent it?
> >
> > Mark
>
> That's because rsync is following the Vista junction points for the old
> XP locations of things. See
> http://www.cs.umd.edu/~cdunne/projs/backuppc_guide.html#Xfer
> for the excludes you need to pass to rsync.
>
> Cody
>
>
> --
> Check out the new SourceForge.net Marketplace.
> It is the best place to buy or sell services for
> just about anything Open Source.
> http://p.sf.net/sfu/Xq1LFB
> ___
> BackupPC-users mailing list
> BackupPC-users@lists.sourceforge.net
> List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
> Wiki:http://backuppc.wiki.sourceforge.net
> Project: http://backuppc.sourceforge.net/
>
--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] vista backup question

2009-01-08 Thread Jeffrey J. Kosowsky
dan wrote at about 10:52:42 -0700 on Thursday, January 8, 2009:
 > in case you were unaware, junction points = hardlinks on ntfs.
Similar but not the same by any means.
Junction points are more "powerful" and hence more dangerous than
hardlinks.
  - Junction points can be on different file systems
  - Junction points can be directories
Also:
  - Junction points require special utility to delete properly (vs
standard "rm" for hard links)
  - With hard links, all links are identical -- i.e. there is no
notion of which is the original and which are the links. Not true
with junctions since the storage is associated with the original file
All of the above can cause problems if you are not careful.
(Note I am not an expert on junction points and they traditionally
have not been well documented so I may be missing some subtleties...)

 > 
 > Just exclude all of the WinXP directories like "Documents and Settings" to
 > solve this.  Vista users the "Users" directory instead..
 > 
 > On Wed, Jan 7, 2009 at 10:32 PM, Cody Dunne  wrote:
 > 
 > > Hi Mark,
 > >
 > > Mark Maciolek wrote:
 > > > hi,
 > > >
 > > > Running 3.1.0 on CentOS 5.2 with 40 clients so far mostly Linux systems.
 > > > Added my first Vista system today. Used Deltacopy to install rsync.
 > > >
 > > > The backup worked but still had 13446 error transfer, mainly file name
 > > > too long.
 > > >
 > > > Remote[1]: rsync: readlink_stat("All Users/Application Data/Application
 > > > Data/Application Data/Application Data/Application Data/Application
 > > > Data/Application Data/Application Data/Application Data/...
 > > >
 > > >
 > > > Basically it is repeating the Application Data over and over again.
 > > >
 > > > Has anyone seen this issue and know how to prevent it?
 > > >
 > > > Mark
 > >
 > > That's because rsync is following the Vista junction points for the old
 > > XP locations of things. See
 > > http://www.cs.umd.edu/~cdunne/projs/backuppc_guide.html#Xfer
 > > for the excludes you need to pass to rsync.
 > >
 > > Cody
 > >
 > >
 > > --
 > > Check out the new SourceForge.net Marketplace.
 > > It is the best place to buy or sell services for
 > > just about anything Open Source.
 > > http://p.sf.net/sfu/Xq1LFB
 > > ___
 > > BackupPC-users mailing list
 > > BackupPC-users@lists.sourceforge.net
 > > List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
 > > Wiki:http://backuppc.wiki.sourceforge.net
 > > Project: http://backuppc.sourceforge.net/
 > >
 > 
 > --
 > --
 > Check out the new SourceForge.net Marketplace.
 > It is the best place to buy or sell services for
 > just about anything Open Source.
 > http://p.sf.net/sfu/Xq1LFB
 > --
 > ___
 > BackupPC-users mailing list
 > BackupPC-users@lists.sourceforge.net
 > List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
 > Wiki:http://backuppc.wiki.sourceforge.net
 > Project: http://backuppc.sourceforge.net/

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] BackupPC misconfiguration & Rsync network usage

2009-01-08 Thread royden yates
On Wed, 2009-01-07 at 21:20 +1100, Adam Goryachev wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
> 
> William McKee wrote:
> > I use VMware on a co-lo server which has 3 guestts that all get backed
> > up by BackupPC. I could identify that the host was transmitting massive
> > amounts of data (130Gb) which appeared to be coming from one of the
> > three guests. However, I couldn't figure out which guest was pushing out
> > the excessive data.
> > 
> > I went through the usual log files without much luck. I then checked the
> > ifconfig output which all looked normal inside the hosts. Once I finally
> > looked at the BackupPC logs for the guest server, I realized what was
> > happening and corrected the issue by removing my bad entry. I also added
> > --bwlimit to the RsyncArgs setting in config.pl to control maxing out my
> > bandwidth.
> > 
> > However, this all took longer than I'd have liked. I'm stumped as to why
> > the data transmitted off of the guest did not show up in the ifconfig
> > output. I know that the guest is sending data via rsync based on the
> > logs. However it's not showing up in the ifconfig stats (see below). Is
> > this due to the way that rsync works? I was sending about 450Mb of data
> > every 1-2 hrs from 8pm - 6am (I can send the logs if that would be of
> > any help). I've included below the ifconfig outputs for the host
> > (massive TX bytes) and the guest (normal TX bytes). I would have
> > expected a corresponding amount of TX bytes for the guest. Thanks for
> > any insight.
> 
> I would suspect vmware has something to do with that. Try creating
> traffic with any other tool, and it likely won't be counted in the way
> you think it should as well.
> 
> Another option is perhaps the counters wrapped due to the amount of
> data... so if they wrapped recently, then the values will be very small,
> even though a huge amount of data has been transmitted.
> 
> There is nothing special that rsync does to cause it's bandwidth not to
> be counted normally (AFAIK).
> 
> Regards,
> Adam

I think Adam is spot on. AFAIK, ifconfig receives its byte data from the
proc subsystem and on 32 bit systems will wrap at 4 gigs.

Regards,

Royden


--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] host config file command line editor ?

2009-01-08 Thread Alex

Hi there, 
i'm not aware using Perl scripting, but ok with bash / php.
Then, i'd like to know if there's a way to set values for per pc config file 
using command line tool ?

By exemple, like to change only FullKeepCnt value, but, has it is by default 
written that way :
$Conf{FullKeepCnt} = [
  '400'
];

It's a bit hard for me to parse this file and write back a new value. 

Didn't find any way of doing this on the wiki. I'd like to be able to do that 
kind of thing without de cgi web interface, in order for me to be able to 
change values through scripts.

Thank you in advance for any advices :)

+--
|This was sent by alexan...@inforeseau.com via Backup Central.
|Forward SPAM to ab...@backupcentral.com.
+--



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] incremental tar xfer errors

2009-01-08 Thread Simone Marzona
Hi

Yes i'm sure...

r...@backup2:/etc/backuppc# cat config.pl |grep $Conf{TarIncrArgs}  
$Conf{TarIncrArgs} = '--newer=$incrDate+ $fileList+';

r...@stars:/etc/backuppc# cat ./*.pl | grep $Conf{TarIncrArgs}
$Conf{TarIncrArgs} = '--newer=$incrDate+ $fileList+';

I'm using debian and ubuntu and I got the same problem in both ...

On Wed, 2009-01-07 at 22:42 -0800, Craig Barratt wrote:
> Simone writes:
> 
> > I got a strange problem doing incrementals with tar over ssh using
> > --newer=$incrDate+. It seems an "escape problem" of part of the time
> > reference for the incremental.
> 
> Yes, the escaping isn't happening.  The "$incrDate+" form means
> to escape the value, so that is what you should use (since you
> are running through ssh).
> 
> Are you sure $Conf{TarIncrArgs} includes --newer=$incrDate+ rather
> than --newer=$incrDate?  Have you checked the per-client config too?
> 
> Craig


--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] incremental tar xfer errors

2009-01-08 Thread Simone Marzona
On Wed, 2009-01-07 at 22:42 -0800, Craig Barratt wrote:
> 
> Yes, the escaping isn't happening.  The "$incrDate+" form means
> to escape the value, so that is what you should use (since you
> are running through ssh).

the problem is:

- a configuration mistake (if so, where may I find it out?)
- a software bug (maybe in ubuntu/debian packaging?)

thanks


--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] host config file command line editor ?

2009-01-08 Thread Chris Robertson
Alex wrote:
> Hi there, 
> i'm not aware using Perl scripting, but ok with bash / php.
> Then, i'd like to know if there's a way to set values for per pc config file 
> using command line tool ?
>
> By exemple, like to change only FullKeepCnt value, but, has it is by default 
> written that way :
> $Conf{FullKeepCnt} = [
>   '400'
> ];
>   

That's HTML for...

$Conf{FullKeepCnt} = [
 '400'
];

It should not be HTML encoded on disk.

$Conf is a hash.  The hash entry "FullKeepCnt" is a list (I think) with, 
in this case, one value: 400.

> It's a bit hard for me to parse this file and write back a new value.

PHP has the htmlspecialchars_decode 
(http://us.php.net/manual/en/function.htmlspecialchars-decode.php), 
which will take the nasty looking string you started with, and transform 
it to the proper on-disk setup, which should also be easier to 
manipulate (I think PHP might even properly utilize Perl hashes).

>  
>
> Didn't find any way of doing this on the wiki. I'd like to be able to do that 
> kind of thing without de cgi web interface, in order for me to be able to 
> change values through scripts.
>
> Thank you in advance for any advices :)

Chris

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Kenneth L. Owen
This seems to be the one thing that I keep finding a way to mess up!

 

Last time, I made the mistake of using an ssh connection to the other
machine to avoid trotting back and forth between buildings.  YOU CAN'T DO
THIS since the ssh session keys get in the way of the backuppc key
generation process.  Once I figured this out, I was able to get my Fedora 8
installation working properly again.

 

Since Fedora 8 will no longer receive support and for some reason I can't
get Fedora 9 to install and work properly, I am moving my systems to Ubuntu.
I setup Ubuntu as a test system and have been playing with it to get
familiar before actually placing it into full (depend upon me) service.  I
thought I was ready, so that's how we got to here.

 

Everything has gone well up to the ssh keys generation process for BackupPC.
I can't get it to work.

 

There are differences between Fedora and Ubuntu:

 

1.  Ubuntu uses sudo and does not have root logon setup by default.  I
gave root a logon password.
2.  Fedora did not give backuppc user a password, but Ubuntu does.
3.  To perform work as user backuppc, in Fedora as root I had to use the
command 'su -s /bin/bash - backuppc', but on Ubuntu as root, I think it is
simply 'su backuppc'.

 

I worked through the key generation process (trotting back and forth between
machines) and all seemed to work exactly as it should all the way up to the
test of the result.  When I enter the command 

 

ssh -l root winserver whoamiorssh -l root
192.168.1.101 whoami

 

it asks for root password.

 

Maybe someone has been here before and knows what I may be missing.

 

I captured a transcript of the commands as run and excerpts follow.  - ken

 

k...@archiver:~$ sudo apt-get install rsync ssh openssh-server

-download details deleted.

Setting up ssh (1:4.7p1-8ubuntu1.2) ...

 

r...@archiver:/var/lib/backuppc/.ssh# su backuppc

$ whoami

backuppc

$ cd /var/lib/backuppc/.ssh

$ ls -al

total 8

drwx-- 2 backuppc backuppc 4096 2009-01-08 13:25 .

drwxr-xr-x 9 backuppc backuppc 4096 2009-01-08 12:46 ..

 

$ ssh-keygen -t rsa

Generating public/private rsa key pair.

Enter file in which to save the key (/var/lib/backuppc/.ssh/id_rsa): 

Enter passphrase (empty for no passphrase): 

Enter same passphrase again: 

Your identification has been saved in /var/lib/backuppc/.ssh/id_rsa.

Your public key has been saved in /var/lib/backuppc/.ssh/id_rsa.pub.

The key fingerprint is:

37:b9:95:9a:2a:1a:c0:f5:97:2a:ad:f8:3f:4d:66:69 backu...@archiver

 

$ cp id_rsa.pub BackupPC_id_rsa.pub

 

$ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/

The authenticity of host '192.168.1.101 (192.168.1.101)' can't be
established.

RSA key fingerprint is 71:a1:03:7d:fb:b9:87:1f:32:c7:a3:46:d0:81:2d:af.

Are you sure you want to continue connecting (yes/no)? yes

Warning: Permanently added '192.168.1.101' (RSA) to the list of known hosts.

r...@192.168.1.101's password: 

BackupPC_id_rsa.pub 100%  399
0.4KB/s   00:00

 

$ chmod -R go-rwx /var/lib/backuppc/.ssh 

 

$ ssh -l root 192.168.1.101 whoami

r...@192.168.1.101's password:  <== Shouldn't get this, but when
entered

root  <==
The correct response.

 

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Renke Brausse
Am Donnerstag, den 08.01.2009, 18:00 -0500 schrieb Kenneth L. Owen:
> $ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/

hmm, I think this should be something like "authorized_keys" on the
remote machine.

Renke


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil
--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Thomas Birnthaler
Hello,

we are using BackupPC and it works perfect on different places.
But we have 2 questions on it:

1. We wonder why neither the GUI nor the commands offer a possibility to 
delete backups (full and incremental). If e.g. the backup partition is filled 
up, this would be very helpful.

As far as we understood the concept of using hardlinks and different 
directories for each backup + a hash pool for each file it seems to be very 
easy to delete a backup NN.

Just delete recursively the backup directory NN and the nightly run will 
really destroy all orphaned files with a hardlink count of 1.

If a full backup is deleted that way, incremental backups depending on it then 
depend on the next earlyer full backup and work as before. Same is true for 
incremental backups. So the GUI works as before and the creation of new 
backups too.

Is that idea correct?

2. What is real benefit of incremental backups compared to full backups?

In both cases only changed/new files use disk space according to the hardlink 
concept of BackupPC.

We have also detected, that in some cases incremental backups need much more 
time than full backups (factor 3-5) This sounds odd to us.

Thanx
Thomas Birnthaler
-- 
OSTC Open Source Training and Consulting GmbH / HRB Nuernberg 20032
tel +49 911-3474544 / fax +49 911-1806277 / http://www.ostc.de
Delsenbachweg 32 / D-90425 Nuernberg / Geschaeftsfuehrung:
Thomas Birnthaler / +49 171-3047465 / t...@ostc.de / pgp 0xFEE7EB4C
Hermann Gottschalk / +49 173-3600680 / h...@ostc.de / pgp 0x0B2D8EEA

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Tino Schwarze
On Thu, Jan 08, 2009 at 06:00:10PM -0500, Kenneth L. Owen wrote:

> 1.Ubuntu uses sudo and does not have root logon setup by default.  I
> gave root a logon password.

Root does not need a password for key-based ssh access. You might need
to enable PermitRootLogin in /etc/ssh/sshd_config though - it might be
disabled.

> 2.Fedora did not give backuppc user a password, but Ubuntu does.
> 3.To perform work as user backuppc, in Fedora as root I had to use the
> command 'su -s /bin/bash - backuppc', but on Ubuntu as root, I think it is
> simply 'su backuppc'.

Depends on how the backuppc user is set up (whether it's got a shell in
/etc/passwd)

> I worked through the key generation process (trotting back and forth between
> machines) and all seemed to work exactly as it should all the way up to the
> test of the result.  When I enter the command 
> 
> ssh -l root winserver whoamiorssh -l root
> 192.168.1.101 whoami
> 
> it asks for root password.

That's not related to your server's root account! Not in any way! You
actually don't need to mess with the BackupPC's root account. Look at
your winserver's ssh log messages (they might even show up in the eventlog).


> I captured a transcript of the commands as run and excerpts follow.  - ken
> 
> k...@archiver:~$ sudo apt-get install rsync ssh openssh-server
> 
> -download details deleted.
> 
> Setting up ssh (1:4.7p1-8ubuntu1.2) ...

You shouldn't need an ssh server on the BackupPC server for backup
purposes. (It's usually handy for administration anyway, I just want to
make things clear for your setup.)
 
> r...@archiver:/var/lib/backuppc/.ssh# su backuppc
> 
> $ whoami
> backuppc
> 
> $ cd /var/lib/backuppc/.ssh
> $ ls -al
> total 8
> drwx-- 2 backuppc backuppc 4096 2009-01-08 13:25 .
> drwxr-xr-x 9 backuppc backuppc 4096 2009-01-08 12:46 ..
> 
> $ ssh-keygen -t rsa
> Generating public/private rsa key pair.
> Enter file in which to save the key (/var/lib/backuppc/.ssh/id_rsa): 
> Enter passphrase (empty for no passphrase): 
> Enter same passphrase again: 
> Your identification has been saved in /var/lib/backuppc/.ssh/id_rsa.
> Your public key has been saved in /var/lib/backuppc/.ssh/id_rsa.pub.
> The key fingerprint is:
> 37:b9:95:9a:2a:1a:c0:f5:97:2a:ad:f8:3f:4d:66:69 backu...@archiver

> $ cp id_rsa.pub BackupPC_id_rsa.pub
> 
> $ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/
> The authenticity of host '192.168.1.101 (192.168.1.101)' can't be
> established.
> RSA key fingerprint is 71:a1:03:7d:fb:b9:87:1f:32:c7:a3:46:d0:81:2d:af.
> Are you sure you want to continue connecting (yes/no)? yes
> Warning: Permanently added '192.168.1.101' (RSA) to the list of known hosts.
> r...@192.168.1.101's password: 
> BackupPC_id_rsa.pub 100%  399
> 0.4KB/s   00:00

Here you copied the generated public key file to your winserver's .ssh
directory. You did not tell the winserver to actually accept that key
automatically. To achieve this, you need to add the key's contents (it's
just one line of text) to the /root/.ssh/authorized_keys file.

> $ chmod -R go-rwx /var/lib/backuppc/.ssh 
 
The client side shouldn't matter.

> $ ssh -l root 192.168.1.101 whoami
> r...@192.168.1.101's password:  <== Shouldn't get this, but when entered

It's correct that you've got the prompt - see above.

To sum things up: For the ssh part of your setup, the winserver is the
ssh server and the backup server is the ssh client. The ssh server will
accept clients which authenticate by either password or an ssh key which
public part is found in the user's authorized_keys file. The ssh command
on the client will try to authenticate via a public key found in
.ssh/id_rsa.pub or .ssh/id_dsa.pub (details vary depending on ssh
config). So, what you need is

1. public/private key pair on BackupPC server in ~backuppc/.ssh/
2. public key on winserver in ~root/.ssh/authorized_keys

HTH,

Tino.

-- 
"What we nourish flourishes." - "Was wir nähren erblüht."

www.lichtkreis-chemnitz.de
www.craniosacralzentrum.de

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Kenneth L. Owen
Yes!  On the client machine I ran:
cat BackupPC_id_rsa.pub >> authorized_keys2

Thanks for the quick response!  -- ken

-Original Message-
From: Renke Brausse [mailto:rbrau...@gmx.net] 
Sent: Thursday, January 08, 2009 6:08 PM
To: General list for user discussion,questions and support
Subject: Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

Am Donnerstag, den 08.01.2009, 18:00 -0500 schrieb Kenneth L. Owen:
> $ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/

hmm, I think this should be something like "authorized_keys" on the
remote machine.

Renke


--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Tino Schwarze
On Fri, Jan 09, 2009 at 12:00:53AM +0100, Thomas Birnthaler wrote:

> we are using BackupPC and it works perfect on different places.
> But we have 2 questions on it:
> 
> 1. We wonder why neither the GUI nor the commands offer a possibility to 
> delete backups (full and incremental). If e.g. the backup partition is filled 
> up, this would be very helpful.
> 
> As far as we understood the concept of using hardlinks and different 
> directories for each backup + a hash pool for each file it seems to be very 
> easy to delete a backup NN.
> 
> Just delete recursively the backup directory NN and the nightly run will 
> really destroy all orphaned files with a hardlink count of 1.
> 
> If a full backup is deleted that way, incremental backups depending on it 
> then 
> depend on the next earlyer full backup and work as before. Same is true for 
> incremental backups. So the GUI works as before and the creation of new 
> backups too.
> 
> Is that idea correct?

Almost. ;-) You should not delete a full backup if there are some
incrementals depending on it. Rule: Only delete a level-n incremental if
there is no level-n+1 incremental (with n=0 for fulls).

Your web view would become inconsistent (e.g. deleted files becoming
visible again).

> In both cases only changed/new files use disk space according to the hardlink 
> concept of BackupPC.
 
Yes. Fulls need some more space because they are completely filled, that
is, all directories contain hardlinks to all the files, whereas
incrementals only contain hardlinks to new and changed files -
filesystem-wise they've got holes which are filled by the CGI.

> We have also detected, that in some cases incremental backups need much more 
> time than full backups (factor 3-5) This sounds odd to us.

What transfer method are you using?

HTH,

Tino.

-- 
"What we nourish flourishes." - "Was wir nähren erblüht."

www.lichtkreis-chemnitz.de
www.craniosacralzentrum.de

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Thomas Birnthaler
Wow Tino,

your faster anwering than I'm able to question ;-)

> Your web view would become inconsistent (e.g. deleted files becoming
> visible again).
Ok, this seems to be a case we forgot ;-(
But if this doesn't matter to us, is it then okay to delete any backup?
Do the GUI and the creation of new backups work and don't throw errors?

By the way: How are deleted files represented?
I argue they are just missing in the respective backup hardlink tree.
So if you delete the backup hardlink tree, the "holes" disappear?

> > In both cases only changed/new files use disk space according to the
> > hardlink concept of BackupPC.
>
> Yes. Fulls need some more space because they are completely filled, that
> is, all directories contain hardlinks to all the files, whereas
> incrementals only contain hardlinks to new and changed files -
> filesystem-wise they've got holes which are filled by the CGI.
But this is just management space (directories and their dentries) --- which 
uses of course some inode and data space. No additional data space and inode 
space is needed for the real files in full backups compared to a incremental 
backup. Is that correct?

If this is the case we think about doing full backups all the time.
So deletion of backups is ok then too.

> > We have also detected, that in some cases incremental backups need much
> > more time than full backups (factor 3-5) This sounds odd to us.
>
> What transfer method are you using?
rsync over GBit networks between Linux machines and also between MacOS 
machines. In both cases that effect happens.

Thomas Birnthaler
-- 
OSTC Open Source Training and Consulting GmbH / HRB Nuernberg 20032
tel +49 911-3474544 / fax +49 911-1806277 / http://www.ostc.de
Delsenbachweg 32 / D-90425 Nuernberg / Geschaeftsfuehrung:
Thomas Birnthaler / +49 171-3047465 / t...@ostc.de / pgp 0xFEE7EB4C
Hermann Gottschalk / +49 173-3600680 / h...@ostc.de / pgp 0x0B2D8EEA

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Les Mikesell
Thomas Birnthaler wrote:
>
> 2. What is real benefit of incremental backups compared to full backups?

Normally they are faster and with non-rsync methods transfer much less data.

> In both cases only changed/new files use disk space according to the hardlink 
> concept of BackupPC.

That's after they get to the server.

> We have also detected, that in some cases incremental backups need much more 
> time than full backups (factor 3-5) This sounds odd to us.

This sounds like you are using rsync and doing infrequent fulls. 
Normally rsync incrementals transfer everything that has changed since 
the last full which is the comparison base.  Files added after the full 
are copied over again in every subsequent incremental.  If you have 3.x 
you can change this with $Conf{IncrLevels}.

Hmmm, it still doesn't make sense that a full would be faster than the 
prior incremental, though.  Are you sure you are comparing apples to 
apples on this?  Rsync incrementals should quickly ignore any files 
where the timestamp and size matches the backuppc copy where fulls will 
do the rsync block checksum compare over everything.


-- 
   Les Mikesell
lesmikes...@gmail.com



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Ryan Manikowski
http://backuppc.wiki.sourceforge.net/How+to+delete+backups

Ryan Manikowski

System Administrator
2ergo Americas Inc.
 
:703.677.8499:
www.2ergo.com
Arlington, VA

 
This message (including attachments) is confidential and may be legally 
privileged. The content and views expressed are those of the sender and not 
necessarily the 2ergo Group. If you are not the intended recipient, you must 
not disclose, copy or use any part of it.  Please delete all copies immediately 
and notify the sender. 2ergo Americas Inc. was formerly known as Proteus Inc.



Thomas Birnthaler wrote:
> Hello,
>
> we are using BackupPC and it works perfect on different places.
> But we have 2 questions on it:
>
> 1. We wonder why neither the GUI nor the commands offer a possibility to 
> delete backups (full and incremental). If e.g. the backup partition is filled 
> up, this would be very helpful.
>
> As far as we understood the concept of using hardlinks and different 
> directories for each backup + a hash pool for each file it seems to be very 
> easy to delete a backup NN.
>
> Just delete recursively the backup directory NN and the nightly run will 
> really destroy all orphaned files with a hardlink count of 1.
>
> If a full backup is deleted that way, incremental backups depending on it 
> then 
> depend on the next earlyer full backup and work as before. Same is true for 
> incremental backups. So the GUI works as before and the creation of new 
> backups too.
>
> Is that idea correct?
>
> 2. What is real benefit of incremental backups compared to full backups?
>
> In both cases only changed/new files use disk space according to the hardlink 
> concept of BackupPC.
>
> We have also detected, that in some cases incremental backups need much more 
> time than full backups (factor 3-5) This sounds odd to us.
>
> Thanx
> Thomas Birnthaler
>   

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Tino Schwarze
Hi Thomas,

> > Your web view would become inconsistent (e.g. deleted files becoming
> > visible again).
> Ok, this seems to be a case we forgot ;-(
> But if this doesn't matter to us, is it then okay to delete any backup?
> Do the GUI and the creation of new backups work and don't throw errors?
> 
> By the way: How are deleted files represented?

They are marked as deleted in the fattrib files.

> I argue they are just missing in the respective backup hardlink tree.
> So if you delete the backup hardlink tree, the "holes" disappear?

There are a lot more cases to consider. I wouldn't risk getting an
inconsistend view of the backups... see below.

> > > In both cases only changed/new files use disk space according to the
> > > hardlink concept of BackupPC.
> >
> > Yes. Fulls need some more space because they are completely filled, that
> > is, all directories contain hardlinks to all the files, whereas
> > incrementals only contain hardlinks to new and changed files -
> > filesystem-wise they've got holes which are filled by the CGI.
> But this is just management space (directories and their dentries) --- which 
> uses of course some inode and data space. No additional data space and inode 
> space is needed for the real files in full backups compared to a incremental 
> backup. Is that correct?

Yes.

> If this is the case we think about doing full backups all the time.
> So deletion of backups is ok then too.

I wouldn't do that. You could use the $Conf{IncrFill} option. Then each
incremental is filled up and does not depend on previous incrementals
(except for doing actual backup).

What are you trying to achieve, after all? Please tell us about the
problem you're trying to solve - there might be easier approaches.
BackupPC automatically deletes too old backups and takes care for
keeping/deleting depending incrementals as long as neccessary.

> > > We have also detected, that in some cases incremental backups need much
> > > more time than full backups (factor 3-5) This sounds odd to us.
> >
> > What transfer method are you using?
>
> rsync over GBit networks between Linux machines and also between MacOS 
> machines. In both cases that effect happens.

That's strange. Please have a look at the server's and client's load
during incremental backup - maybe something is running on either of them
which takes a lot of CPU and/or network and/or memory...

HTH,

Tino.

-- 
"What we nourish flourishes." - "Was wir nähren erblüht."

www.lichtkreis-chemnitz.de
www.craniosacralzentrum.de

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] Change target directory

2009-01-08 Thread Holger Parplies
Hi,

Juergen Harms wrote on 2009-01-07 17:39:25 +0100 [Re: [BackupPC-users] Change 
target directory]:
> $Conf{TopDir} appears to be a "hardly coded variable".

is that supposed to be a pun?

> [...]
> I had found clear instructions on what to do in the wiki, but can't find
> them any more - the wiki appears to have been recently improved.

http://backuppc.wiki.sourceforge.net/change+archive+directory

- unchanged since 30th of August. First item under "Tips and Tricks". Maybe
your search method has been improved ? ;-)

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] vista backup question

2009-01-08 Thread Holger Parplies
Hi,

Jeffrey J. Kosowsky top-posted (please don't do that) on 2009-01-08 13:11:17 
-0500 [Re: [BackupPC-users] vista backup question]:
> [...]
> Similar but not the same by any means.
> Junction points are more "powerful" and hence more dangerous than
> hardlinks.
>   - Junction points can be on different file systems
>   - Junction points can be directories
> Also:
>   - Junction points require special utility to delete properly (vs
>   standard "rm" for hard links)
>   - With hard links, all links are identical -- i.e. there is no
>   notion of which is the original and which are the links. Not true
>   with junctions since the storage is associated with the original file

that sounds pretty much like bind mounts on Linux.

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] vista backup question

2009-01-08 Thread Holger Parplies
Hi,

Cody Dunne wrote on 2009-01-08 13:32:00 +0800 [Re: [BackupPC-users] vista 
backup question]:
> [...]
> See http://www.cs.umd.edu/~cdunne/projs/backuppc_guide.html#Xfer

you keep quoting your guide. How about contributing the information (or at
least the pointer, but better the information) to the wiki?

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Holger Parplies
Hi,

Thomas Birnthaler wrote on 2009-01-09 00:00:53 +0100 [[BackupPC-users] How to 
delete backups? + Benefit of incremental backups?]:
> [...]
> 1. We wonder why neither the GUI nor the commands offer a possibility to 
> delete backups (full and incremental). If e.g. the backup partition is filled 
> up, this would be very helpful.

it's usually not necessary (because BackupPC deletes backups when they expire)
or desired (until they expire, they should not be deleted). Furthermore, it's
a "dangerous" operation. I would argue against making that available with one
or two clicks on a web page.

> [...]
> 2. What is real benefit of incremental backups compared to full backups?

They are, in general (and if used correctly) significantly faster. They put
less strain on server and client disks. This has been answered many times.

Tino Schwarze wrote on 2009-01-09 01:17:21 +0100 [Re: [BackupPC-users] How to 
delete backups? + Benefit of incremental backups?]:
> > I argue they are just missing in the respective backup hardlink tree.
> > So if you delete the backup hardlink tree, the "holes" disappear?
> 
> There are a lot more cases to consider. I wouldn't risk getting an
> inconsistend view of the backups... see below.

The most important one, as a simple example:

First full backup contains only file 'a'.
Second full backup contains files 'a' and 'b'.
Incremental after second full contains only file 'c' (meaning the target
directory actually contained files 'a', 'b' and 'c').

Now remove the second full backup. Your incremental will now be interpreted as
containing files 'a' and 'c'. File 'b' has vanished. Note that this is a state
in which your original file system has been *at no point in time*. You not
only lose data, you get an inconsistent state.

Future rsync incrementals *should* fix that (or fail - not sure), because they
would be run against the contents of the reference backup. If BackupPC finds
the first full backup as a reference (again: not sure if that happens), then
all changes would be transferred. My guess is that if that does not happen,
the incremental would be done against an empty tree, thus transferring
*everything* (which is more than a full would do).
tar/smb incrementals are only based on a timestamp. Again, I'm not sure what
would be considered as the reference backup (since the real reference has
disappeared). If, somehow, the *time stamp* of the deleted backup were used,
the problem would *not* be fixed. File 'b' wasn't modified since the time
stamp - no reason to back it up.

In short: don't do it. What is the point of removing comparatively "young"
full backups (that have incrementals depending upon them) anyway? Wouldn't you
remove the *oldest* backups? Or *correctly* use the exponential strategy
BackupPC provides?

Someone in this thread wrote that BackupPC takes care of deleting dependant
incrementals with the expiring full. That is *not* correct. BackupPC actually
delays deleting the full backup until it is no longer needed, because there
are no dependant backups left. This is the sensible thing to do. Remove the
dependant incrementals rather than the full backup. Following your own
argumentation, removing an incremental will free just as much space as
removing a full would [well, that's not true, but you get the idea].

> > But this is just management space (directories and their dentries) ---
> > which uses of course some inode and data space. No additional data space
> > and inode space is needed for the real files in full backups compared to
> > a incremental backup. Is that correct?
> 
> Yes.

Almost. All *directories* are created for an incremental anyway (not only
directories containing changes). Without the entries for unchanged files, the
directories will be shorter, same number of inodes, though.
If the maximum hard link count is exceeded for a file, you will get a new
copy of it in the pool. While 32000 possible links is *a lot*, you still may
get overflows, and you'll get them sooner with only full backups. Just a
detail.

> What are you trying to achieve, after all? Please tell us about the
> problem you're trying to solve - there might be easier approaches.

This question is *important*. All of the points made here have been made *a
lot* of times. You are trying to do something that doesn't seem to make sense,
but, presumably, the problem you are trying to solve *does* exist. I would
rather help you solve a problem than criticize your questions.

> > > > We have also detected, that in some cases incremental backups need much
> > > > more time than full backups (factor 3-5) This sounds odd to us.
> > >
> > > What transfer method are you using?
> >
> > rsync over GBit networks between Linux machines and also between MacOS 
> > machines. In both cases that effect happens.
> 
> That's strange.

No, it's not. Not necessarily, anyway.

Les Mikesell wrote on 2009-01-08 17:52:10 -0600 [Re: [BackupPC-users] How to 
delete backups? + Benefit of incremental backups?]:
> [...

Re: [BackupPC-users] missing rdev on device...

2009-01-08 Thread Tim Connors
On Wed, 24 Dec 2008, Holger Parplies wrote:

> Hi,
>
> Glassfox wrote on 22.12.2008 at 16:50:24 [[BackupPC-users]  missing rdev on 
> device...]:
> > I tried a complete restore on my localhost (which also runs BackupPC) today 
> > and got a lot of error messages like this:
> >
> > "2008-12-22 21:54:23 localhost: File::RsyncP::FileList::encode: missing 
> > rdev on device file dev/initctl"
> >
> > Any idea what's wrong here?
>
> this is weird.

Bugger, I'm in need of restoring my filesystem and I just came across
this.  It issued a few of these messages, then seems to have paused, with
rsync or backuppc not showing any activity according to strace.  Dunno
whether the incorrect handling of the pipe caused the susequent pause.
I'm restoring to blank partitions, so it's not like it's blocking on
writing to a pipe on the remote filesystem, since the pipe doesnt exist
yet.  Note that in my debugging, I have blown away the temporary partition
I created, so I can't verify whether it had actually created the pipe yet
(one suspects not, if Rsync.pm couldn't handle it).

> 1.) /dev/initctl is a named pipe. There is no point in *restoring* a named
> pipe, because you can't "restore" the process listening "on the other
> end". You might still want to back up named pipes, for the sake of
> having an exact image of your file system for reference or auditing
> purposes.

Sure you want to restore named pipes (I think you're thinking of sockets).
Named pipes are pipes on the filesystem that haev a static name.  They can
be created temporarily by processes to communicate with another process
(but what's the point, that could have just passed over file descriptors
to a socket), or they can be pipes that you keep around that are
persistent over processes restarting (I first used pipes about 10 years
ago to have a music daemon that could block on a pipe until some reader
came along.  The first pipes backuppc came across when trying to
restore my current system was a pipe I created a few months ago in order
to work around limitations of a script I was using that wouldnt write to
stdout or /dev/stdout, but would write to a named pipe.

/dev/initctl for instance always exists in /dev.  If you want to restore
/dev, then you have to be able to handle named pipes.

Also, rsync is able to read and write named pipes.  It probably is -D, but
in order to restore a filesystem, you need to use -D.

> Did all of the error messages refer to named pipes?

2009-01-09 12:47:11 gamow: File::RsyncP::FileList::encode: missing rdev on 
device file tconnors/.config/gxine/socket
2009-01-09 12:47:11 gamow: File::RsyncP::FileList::encode: missing rdev on 
device file tconnors/.signature
2009-01-09 12:47:11 gamow: File::RsyncP::FileList::encode: missing rdev on 
device file 
tconnors/movies/a2k/handlebar_mounted_SSO_blackburnhill_return/intermediate1.mpg
2009-01-09 12:47:11 gamow: File::RsyncP::FileList::encode: missing rdev on 
device file 
tconnors/movies/a2k/handlebar_mounted_SSO_blackburnhill_return/intermediate2.mpg

They are named pipes on the system that was backed up and is trying to be
restored to.

> 2.) As far as I understand the complaining code (FileList/FileList.xs in the
> source of File::RsyncP), it is interpreting the file as a device, not
> a pipe. Might this be related to the "-D" option syntax change in rsync
> 2.6.7? Has anyone successfully tested backing up and restoring named
> pipes?

Backing up has never complained.  The file exists as an empty file in the
full backup.
Dunno how to parse the attrib file.

in FileList/rsync.h,
#define IS_DEVICE(mode) (S_ISCHR(mode) || S_ISBLK(mode) || S_ISSOCK(mode)
|| S_ISFIFO(mode))

> What rsync commands are run for backup and for restore? You can find them in
> the XferLOG and RestoreLOG files ...

I'm restoring from an incremental backup:
backup:
incr backup started back to 2008-12-25 15:00:01 (backup #14) for
directory /home
Running: /usr/bin/ssh -q -x -l backuppc gamow sudo /usr/bin/rsync
--server --sender --numeric-ids --perms --owner --group -D --links
--hard-links --times --block-size=2048 --recursive
--one-file-system --checksum-seed=32761 . /home/
restore:
Running: /usr/bin/ssh -q -x -l backuppc gamow sudo /usr/bin/rsync
--server --numeric-ids --perms --owner --group -D --links
--hard-links --times --block-size=2048 --relative
--ignore-times --recursive --checksum-seed=32761 . /mnt/


-- 
TimC
MacOSX: Sort of like a pedigree persian cat. Very sleek, very
sexy, but a little too prone to going cross-eyed, biting you on
your thumb and then throwing up on your trousers. -- Jim in ASR

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sour

Re: [BackupPC-users] I received the error "No files dumped for share"

2009-01-08 Thread Holger Parplies
Hi,

Craig Barratt wrote on 2009-01-07 22:46:09 -0800 [Re: [BackupPC-users] I 
received the error "No files dumped for share"]:
> Omar writes:
> 
> > $Conf{TarClientCmd} = ' env LC_ALL=C /usr/bin/sudo $tarPath -c -v -f -
> > -C $shareName+'
> > . ' --totals';
> > 
> > $Conf{TarClientRestoreCmd} = ' env LC_ALL=C /usr/bin/sudo $tarPath -x -p
> > --numeric-owner --same-owner'
> >. ' -v -f - -C $shareName+';
> 
> Both of these are wrong - they start with a space.  BackupPC doesn't
> know what program to exec.
> 
> You need something like:
> 
> $Conf{TarClientCmd} = '/usr/bin/sudo env LC_ALL=C $tarPath -c -v -f - -C 
> $shareName+ --totals';

though I would like to point out that configuring sudo to allow execution of
'env' is effectively allowing arbitrary commands as root. You need to have
something like

backuppcALL=NOPASSWD: /usr/bin/env LC_ALL=C /usr/bin/tar -c *

in /etc/sudoers to be fairly secure (you might want to specify more or less of
the command line, depending on what you want; more to limit it to a specific
directory, less to allow restores with the same sudoers entry).

If you use something like '/usr/bin/env LC_ALL=C /usr/bin/sudo ...' that
probably won't work, depending on which environment variables your sudo
propagates and which it doesn't.

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] Latest backup removed?

2009-01-08 Thread Holger Parplies
Hi,

Dejan Batic wrote on 2009-01-08 10:05:33 -0400 [[BackupPC-users] Latest backup 
removed?]:
> [...]
> I'm using Bakcuppc for around 30 machines running XP. The backup method used
> is rsyncd.
> 
> [...]
> 
> So why the latest backup gets removed immediately after it's done?

I can spot one typo - not sure if that is responsible :

> $Conf{IncrKeepCnit} = '6';

Should be IncrKeepCnt (without the 'i'). Is this from the main config file?

A quick glance at BackupPC_dump suggests that a *missing* IncrKeepCnt would
default to 0 and thus lead to what you are observing ...

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] vista backup question

2009-01-08 Thread Cody Dunne
Hi Holger,

Holger Parplies wrote:
> Hi,
> 
> Cody Dunne wrote on 2009-01-08 13:32:00 +0800 [Re: [BackupPC-users] vista 
> backup question]:
>> [...]
>> See http://www.cs.umd.edu/~cdunne/projs/backuppc_guide.html#Xfer
> 
> you keep quoting your guide. How about contributing the information (or at
> least the pointer, but better the information) to the wiki?
> 
> Regards,
> Holger
> 

I copied the excludes a while back to the Common_backup_excludes page on 
the wiki, and keep it updated. If anyone wants to put a link to the 
other parts of my guide on the wiki or even make a page for all of it, 
they're welcome to as long as they cite the source. I'm not interested 
in maintaining both copies, though.

Cody

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] BackupPC_link got error -4 when calling MakeFileLink

2009-01-08 Thread Holger Parplies
Hi mailing list,
hi Backup Central forum user,

tagore wrote on 2009-01-08 09:21:05 -0500 [[BackupPC-users] BackupPC_link got 
error -4 when calling MakeFileLink]:
> 
> Hi!
> 
> Is there a result? We have similar problem.
> 
> Thanks

from the complete details you are giving (no in-reply-to or references
headers, no quote), I'd say, "I have no idea what you are talking about". Your
subject suggests that the correct answer to your query is "yes", though.

It's often been asked, it's often been answered. You'll need to search.

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] host config file command line editor ?

2009-01-08 Thread Holger Parplies
Hi,

Alex wrote on 2009-01-08 10:41:35 -0500 [[BackupPC-users]  host config file 
command line editor ?]:
> i'm not aware using Perl scripting, but ok with bash / php.

learn Perl, it's much more fun :-).

> Then, i'd like to know if there's a way to set values for per pc config
> file using command line tool ?

'vi host.pl' ?

> By exemple, like to change only FullKeepCnt value, but, has it is by default
> written that way :
> $Conf{FullKeepCnt} = [
>   '400'
> ];
> 
> It's a bit hard for me to parse this file and write back a new value. 

I can see why that would be ...

> [...] I'd like to be able to do that kind of thing without de cgi web
> interface, in order for me to be able to change values through scripts.

Well, I don't know of any existing tool. It much depends on what you want to
do and be able to do.

- Can you guarantee that the host.pl files will never be modified through
  the web interface?
- Do you want to keep comments etc. in the host.pl files or just the settings?
- How do you suggest setting anything more complex than a simple string value
  (arrays, hashes, hashes of arrays)?
- BackupPC may accept different types of values for one parameter.
  BackupFilesExclude, for example, can be a string, an array of strings, or a
  hash of arrays of strings. Do you want to handle all cases correctly?

Depending on your answers, you could re-format your host.pl files to put each
assignment on one line (with some tending to get rather long) and then use
'sed' to modify them, do some "perl -e 'use Data::Dumper; ...'"-magic (I
started a small script but dropped it when the array question came up) ... you
could probably even use BackupPC::CGI::EditConfig, though I doubt emulating a
CGI POST request and supplying all the values is what you really had in mind.

Actually, I just added another few lines to the script for handling the first
hash/array cases that spring to mind, as well as removing values. It's not
extensively tested, but since I quick-hack(tm)ed it, I might as well include
it, just in case it fits anyone's needs. I'm not going to use it myself, as
I'd like to preserve comments in my config files.

Questions as to how to use it will be answered if posted to the mailing list
directly ;-). For the beginning, try

BackupPC_editconf.pl -f /etc/backuppc/testhost.pl FullKeepCnt=399 
'BackupFilesExclude{/}[0]=/sys'

Attachments can apparently not be accessed via Backup Central, which doesn't
really bother me. See the real mailing list archive instead:

http://sourceforge.net/mailarchive/forum.php?thread_name=1231429295.m2f.295958%40www.backupcentral.com&forum_name=backuppc-users

Regards,
Holger
#!/usr/bin/perl -w
#
# Author: Holger Parplies, 09.01.09, GPL, uncommented, -*-quick-hack-*-

use strict;
use Data::Dumper;
use Getopt::Std;

$Data::Dumper::Terse = 1;

my %opts = (
	f => undef,		# file name of config file to manipulate
   );

if (not getopts ('f:', \%opts) or @ARGV == 0 or not defined $opts {f}) {
  die "Usage: $0 -f host.pl VAR=VALUE ...\n";
}

our %Conf = ();
do $opts {f}
  or die "Can't do $opts{f}: $!\n";

foreach (@ARGV) {
  my ($var, $val) = m/^\s*(.*?)\s*(?:=\s*(.*?)\s*)?$/;
  my $confvar;
  if ($var =~ /^(.*)\{(.*)\}\[(.*)\]$/) {
defined $val ? $Conf {$1} {$2} [$3] = $val
		 : splice @{$Conf {$1} {$2}}, $3, 1;
  } elsif ($var =~ /^(.*)\[(.*)\]$/) {
defined $val ? $Conf {$1} [$2] = $val : splice @{$Conf {$1}}, $2, 1;
  } elsif ($var =~ /^(.*)\{(.*)\}$/) {
defined $val ? $Conf {$1} {$2} = $val : delete $Conf {$1} {$2};
  } else {
defined $val ? $Conf {$var} = $val : delete $Conf {$var};
  }
}

open CONF, ">$opts{f}.new"
  or die "Can't create new configuration file $opts{f}.new: $!\n";
foreach (sort keys %Conf) {
  my $val = Data::Dumper::Dumper ($Conf {$_});
  chomp ($val);
  print CONF "\$Conf {$_} = $val;\n";
}
close CONF;

rename $opts {f}, "$opts{f}.old";
rename "$opts{f}.new", $opts {f};

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] [OT] Documentation practices (was: Re: vista backup question)

2009-01-08 Thread Holger Parplies
Hi,

Cody Dunne wrote on 2009-01-09 10:58:27 +0800 [Re: [BackupPC-users] vista 
backup question]:
> Holger Parplies wrote:
> > you keep quoting your guide. How about contributing the information (or at
> > least the pointer, but better the information) to the wiki?
> 
> I copied the excludes a while back to the Common_backup_excludes page on 
> the wiki, and keep it updated. If anyone wants to put a link to the 
> other parts of my guide on the wiki or even make a page for all of it, 
> they're welcome to as long as they cite the source. I'm not interested 
> in maintaining both copies, though.

I don't really care one way or the other. Personally, I don't visit random web
sites. If I was looking for information on BackupPC, I'd look at the wiki.
When I give references, those will be to the wiki. If your guide was part of
the wiki, I'd look at it, reference it, and maybe make suggestions for
improvements (or add them to the wiki myself). As it isn't, I won't (I don't
reference what I haven't looked at).

The BackupPC software is open source and available through sourceforge. If you
prefer to offer documentation elsewhere, that is your choice (you may have
good reasons for doing so - I don't know). Providing documentation is
commendable. Making it easily available would be even more so. But that's just
my opinion.

There is a lot of information available on the mailing list. Much of it
doesn't get copied into the wiki. I'm not sure why, but I'm sure that's a
pity. And it's no coincidence I explicitly cc'ed Jeffrey ;-).

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] host config file command line editor ?

2009-01-08 Thread Jeffrey J. Kosowsky
Alex wrote at about 10:41:35 -0500 on Thursday, January 8, 2009:
 > 
 > Hi there, 
 > i'm not aware using Perl scripting, but ok with bash / php.
 > Then, i'd like to know if there's a way to set values for per pc config file 
 > using command line tool ?
 > 
I'm having a little trouble understanding your English (not ur fault)
but it seems like you are looking to use bash or php to do per-host
configuration.

Since the config files (and BackupPC itself) are written in Perl, it
would seem that Perl is the natural language to use.
If you only want to set variables on a per-machine basis, you just
need to copy/paste/edit the existing config file for each machine --
you don't really need to know any Perl except the most basic of
variable and array assignment syntax.

If you want to 'hack' the config script to run some actual perl code
(and not just assign static variables) then you may need to know a
little Perl code and I have posted last month some tricks you can use.

 > By exemple, like to change only FullKeepCnt value, but, has it is by default 
 > written that way :
 > $Conf{FullKeepCnt} = [
 >   '400'
 > ];
 > 
 > It's a bit hard for me to parse this file and write back a new value. 
 > 
 > Didn't find any way of doing this on the wiki. I'd like to be able to do 
 > that kind of thing without de cgi web interface, in order for me to be able 
 > to change values through scripts.
 > 
 > Thank you in advance for any advices :)
 > 
 > +--
 > |This was sent by alexan...@inforeseau.com via Backup Central.
 > |Forward SPAM to ab...@backupcentral.com.
 > +--
 > 
 > 
 > 
 > --
 > Check out the new SourceForge.net Marketplace.
 > It is the best place to buy or sell services for
 > just about anything Open Source.
 > http://p.sf.net/sfu/Xq1LFB
 > ___
 > BackupPC-users mailing list
 > BackupPC-users@lists.sourceforge.net
 > List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
 > Wiki:http://backuppc.wiki.sourceforge.net
 > Project: http://backuppc.sourceforge.net/
 > 

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Kenneth L. Owen
Tino,

Having gotten accustomed to Fedora, I prefer working as root when required
instead of having to enter lines twice -- once without sudo, error, then
retype it with sudo.  Also, root can navigate to some directories that
normal user can't go.  Once I get more used to using sudo, I may put root
back as it was.  This was the ONLY reason for setting a root password.

The backuppc password is only used initially to start BackupPC CGI for
setup, I think.  I haven't gotten that far as yet.  BackupPC is installed,
config.pl has been edited, but backuppc is currently halted until I get the
keys set.  I have my LVM drive to house the backup files mounted and ready.

The Archiver machine will also be my daytime workstation in the workshop.
Therefore, I want to be able to do more than just run BackupPC at night.
Some software not pertaining to BackupPC will be needed.

Tino, in the original transcript, I didn't show work done on Winserver.

Maybe I see the problem.  On the winserver end, I put the public key in
authorized_keys2.  Maybe Ubuntu version uses a different name from Fedora.
Maybe I should have put the key in authorized_keys.

I tried using authorized_keys and got the same result.  The transcript
follows.  (showing work on both ends) -- ken

>>> On Archiver:  ip: 192.168.1.106
k...@archiver:~$ su -
Password: 
r...@archiver:~# su backuppc
$ whoami
backuppc
$ ssh-keygen -t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/var/lib/backuppc/.ssh/id_rsa): 
Enter passphrase (empty for no passphrase): 
Enter same passphrase again: 
Your identification has been saved in /var/lib/backuppc/.ssh/id_rsa.
Your public key has been saved in /var/lib/backuppc/.ssh/id_rsa.pub.
The key fingerprint is:
08:2c:0f:a6:10:c6:6f:8a:0b:e9:1c:a9:1a:e3:30:39 backu...@archiver

$ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/
The authenticity of host '192.168.1.101 (192.168.1.101)' can't be
established.
RSA key fingerprint is 71:a1:03:7d:fb:b9:87:1f:32:c7:a3:46:d0:81:2d:af.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '192.168.1.101' (RSA) to the list of known hosts.
r...@192.168.1.101's password: 
BackupPC_id_rsa.pub
100%  399 0.4KB/s   00:00

>>> On Winserver:  ip:  192.168.1.101 (Also running Ubuntu)
k...@winserver:~$ su -
Password: 
r...@winserver:~# cd /root/.ssh
r...@winserver:~/.ssh# ls -al
total 12
drwx--  2 root root 4096 2009-01-08 22:58 .
drwxr-xr-x 10 root root 4096 2009-01-08 22:39 ..
-rw-r--r--  1 root root  399 2009-01-08 22:58 BackupPC_id_rsa.pub

r...@winserver:~/.ssh# exit
logout
>>>  root account can't run gedit in Ubuntu  <<<
k...@winserver:~$ sudo gedit /root/.ssh/BackupPC_id_rsa.pub
sudo: unable to resolve host Winserver
[sudo] password for ken: 
Added:  from="Archiver" at beginning of BackupPC_id_rsa.pub

k...@winserver:~$ su -
Password: 
r...@winserver:~# cd /root/.ssh
r...@winserver:~/.ssh# cat BackupPC_id_rsa.pub >> authorized_keys
r...@winserver:~/.ssh# cat authorized_keys
from="Archiver" ssh-rsa B...//...CzFk8lw== backu...@archiver
r...@winserver:~/.ssh# 

>>> On Archiver:
$ chmod -R go-rwx *
$ ls -al
total 24
drwx-- 2 backuppc backuppc 4096 2009-01-08 22:58 .
drwxr-xr-x 9 backuppc backuppc 4096 2009-01-08 12:46 ..
-rw--- 1 backuppc backuppc  399 2009-01-08 22:57 BackupPC_id_rsa.pub
-rw--- 1 backuppc backuppc 1675 2009-01-08 22:33 id_rsa
-rw--- 1 backuppc backuppc  399 2009-01-08 22:33 id_rsa.pub
-rw--- 1 backuppc backuppc  442 2009-01-08 22:58 known_hosts
$ ssh -l root 192.168.1.101 whoami
r...@192.168.1.101's password:<---Here it is asking for root password on
Winserver
root
$

-Original Message-
From: Tino Schwarze [mailto:backuppc.li...@tisc.de] 
Sent: Thursday, January 08, 2009 6:19 PM
To: backuppc-users@lists.sourceforge.net
Subject: Re: [BackupPC-users] It's me again, Margret! - ssh-keygen



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Jeffrey J. Kosowsky
Thomas Birnthaler wrote at about 00:00:53 +0100 on Friday, January 9, 2009:
 > Hello,
 > 
 > we are using BackupPC and it works perfect on different places.
 > But we have 2 questions on it:
 > 
 > 1. We wonder why neither the GUI nor the commands offer a possibility to 
 > delete backups (full and incremental). If e.g. the backup partition is 
 > filled 

Someone has written a bash routine called I think
BackupPC_deleteBackup (see:
http://backuppc.wiki.sourceforge.net/How+to+delete+backups) that
allows you to delete individual backups -- either full or incremental
-- along with any dependent higher level incremental backups. It is
necessary to also delete the dependent backups or else they are no
longer accurate as-is (note the program I wrote BackupPC_deleteFiles
goes through a *lot* of painful chains and cases to allow you to
delete individual files from individual backups but it would be very
slow to extend this to an entire backup).

 > up, this would be very helpful.
 > 
 > As far as we understood the concept of using hardlinks and different 
 > directories for each backup + a hash pool for each file it seems to be very 
 > easy to delete a backup NN.
 > 
 > Just delete recursively the backup directory NN and the nightly run will 
 > really destroy all orphaned files with a hardlink count of 1.
 > 
You need to also delete the higher level incremental backups or they
will be wrong/meaningless. Also you probably want to adjust the backup
info file -- this is done correctly by the program I reference above.

 > If a full backup is deleted that way, incremental backups depending on it 
 > then 
 > depend on the next earlyer full backup and work as before. Same is true for 
 > incremental backups. So the GUI works as before and the creation of new 
 > backups too.
 > 
 > Is that idea correct?
NOPE - incremental backups by definition are meaningless/invalid if
you delete lower level full or incremental backups on which they depend.
 > 
 > 2. What is real benefit of incremental backups compared to full backups?
Speed of backup.
 > 
 > In both cases only changed/new files use disk space according to the 
 > hardlink 
 > concept of BackupPC.
 > 
 > We have also detected, that in some cases incremental backups need much more 
 > time than full backups (factor 3-5) This sounds odd to us.
Seems odd...
 > 
 > Thanx
 > Thomas Birnthaler
 > -- 
 > OSTC Open Source Training and Consulting GmbH / HRB Nuernberg 20032
 > tel +49 911-3474544 / fax +49 911-1806277 / http://www.ostc.de
 > Delsenbachweg 32 / D-90425 Nuernberg / Geschaeftsfuehrung:
 > Thomas Birnthaler / +49 171-3047465 / t...@ostc.de / pgp 0xFEE7EB4C
 > Hermann Gottschalk / +49 173-3600680 / h...@ostc.de / pgp 0x0B2D8EEA
 > 
 > --
 > Check out the new SourceForge.net Marketplace.
 > It is the best place to buy or sell services for
 > just about anything Open Source.
 > http://p.sf.net/sfu/Xq1LFB
 > ___
 > BackupPC-users mailing list
 > BackupPC-users@lists.sourceforge.net
 > List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
 > Wiki:http://backuppc.wiki.sourceforge.net
 > Project: http://backuppc.sourceforge.net/
 > 

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] How to delete backups? + Benefit of incremental backups?

2009-01-08 Thread Jeffrey J. Kosowsky
Thomas Birnthaler wrote at about 00:41:26 +0100 on Friday, January 9, 2009:
 > Wow Tino,
 > 
 > your faster anwering than I'm able to question ;-)
 > 
 > > Your web view would become inconsistent (e.g. deleted files becoming
 > > visible again).
 > Ok, this seems to be a case we forgot ;-(
 > But if this doesn't matter to us, is it then okay to delete any backup?
You also have the problem that you are missing new changes so that you
will have an inconsistent system state.
 > Do the GUI and the creation of new backups work and don't throw errors?
 > 
 > By the way: How are deleted files represented?
They are given a type=10 deleted attribute in the corresponding
directory attrib file.
 > I argue they are just missing in the respective backup hardlink
 tree.
That would be wrong -- you need to explicitly notate that files are
deleted because otherwise the absence of a file is interpreted to mean
that you inherit the file from earlier versions.
 > So if you delete the backup hardlink tree, the "holes" disappear?
Nope -- see above.
 > 
 > > > In both cases only changed/new files use disk space according to the
 > > > hardlink concept of BackupPC.
 > >
 > > Yes. Fulls need some more space because they are completely filled, that
 > > is, all directories contain hardlinks to all the files, whereas
 > > incrementals only contain hardlinks to new and changed files -
 > > filesystem-wise they've got holes which are filled by the CGI.
 > But this is just management space (directories and their dentries) --- which 
 > uses of course some inode and data space. No additional data space and inode 
 > space is needed for the real files in full backups compared to a incremental 
 > backup. Is that correct?
Nope. Zero length files are given their own inodes. Also symbolic links are
represented as files (which are non-zero length). Now this might not
make a measurable difference but it is important to understand
what BackupPC is doing before you generalize or make assumptions.

 > 
 > If this is the case we think about doing full backups all the time.
 > So deletion of backups is ok then too.
 > 
 > > > We have also detected, that in some cases incremental backups need much
 > > > more time than full backups (factor 3-5) This sounds odd to us.
 > >
 > > What transfer method are you using?
 > rsync over GBit networks between Linux machines and also between MacOS 
 > machines. In both cases that effect happens.
 > 
 > Thomas Birnthaler
 > -- 
 > OSTC Open Source Training and Consulting GmbH / HRB Nuernberg 20032
 > tel +49 911-3474544 / fax +49 911-1806277 / http://www.ostc.de
 > Delsenbachweg 32 / D-90425 Nuernberg / Geschaeftsfuehrung:
 > Thomas Birnthaler / +49 171-3047465 / t...@ostc.de / pgp 0xFEE7EB4C
 > Hermann Gottschalk / +49 173-3600680 / h...@ostc.de / pgp 0x0B2D8EEA
 > 
 > --
 > Check out the new SourceForge.net Marketplace.
 > It is the best place to buy or sell services for
 > just about anything Open Source.
 > http://p.sf.net/sfu/Xq1LFB
 > ___
 > BackupPC-users mailing list
 > BackupPC-users@lists.sourceforge.net
 > List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
 > Wiki:http://backuppc.wiki.sourceforge.net
 > Project: http://backuppc.sourceforge.net/
 > 

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Adam Goryachev
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Kenneth L. Owen wrote:
> The Archiver machine will also be my daytime workstation in the workshop.
> Therefore, I want to be able to do more than just run BackupPC at night.
> Some software not pertaining to BackupPC will be needed.

Nothing stops you from running other programs, daemons, etc on the
backuppc server...

> Tino, in the original transcript, I didn't show work done on Winserver.
> 
> Maybe I see the problem.  On the winserver end, I put the public key in
> authorized_keys2.  Maybe Ubuntu version uses a different name from Fedora.
> Maybe I should have put the key in authorized_keys.

I've seen authorised_keys2 in some documentation, but have never needed
to use it. I use debian, ubuntu, and a long time ago used redhat (before
RH9) and slackware and have always used ~/.ssh/authorized_keys

> I tried using authorized_keys and got the same result.  The transcript
> follows.  (showing work on both ends) -- ken
> 
 On Archiver:  ip: 192.168.1.106
> k...@archiver:~$ su -
> Password: 
> r...@archiver:~# su backuppc
> $ whoami
> backuppc
> $ ssh-keygen -t rsa
> Generating public/private rsa key pair.
> Enter file in which to save the key (/var/lib/backuppc/.ssh/id_rsa): 
> Enter passphrase (empty for no passphrase): 
> Enter same passphrase again: 

Did you enter an empty passphrase?

> Your identification has been saved in /var/lib/backuppc/.ssh/id_rsa.
> Your public key has been saved in /var/lib/backuppc/.ssh/id_rsa.pub.
> The key fingerprint is:
> 08:2c:0f:a6:10:c6:6f:8a:0b:e9:1c:a9:1a:e3:30:39 backu...@archiver
> 
> $ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/

You are copying a different file... the above command created
/var/lib/backuppc/.ssh/id_rsa.pub and you are copying BackupPC_id_rsa.pub...

> The authenticity of host '192.168.1.101 (192.168.1.101)' can't be
> established.
> RSA key fingerprint is 71:a1:03:7d:fb:b9:87:1f:32:c7:a3:46:d0:81:2d:af.
> Are you sure you want to continue connecting (yes/no)? yes
> Warning: Permanently added '192.168.1.101' (RSA) to the list of known hosts.

This will save you from doing this later, if you configure backuppc to
backup 192.168.1.101 as opposed to the host "someserver" or whatever
name you use.

> r...@192.168.1.101's password: 
> BackupPC_id_rsa.pub
> 100%  399 0.4KB/s   00:00
> 
 On Winserver:  ip:  192.168.1.101 (Also running Ubuntu)
> k...@winserver:~$ su -
> Password: 
> r...@winserver:~# cd /root/.ssh
> r...@winserver:~/.ssh# ls -al
> total 12
> drwx--  2 root root 4096 2009-01-08 22:58 .
> drwxr-xr-x 10 root root 4096 2009-01-08 22:39 ..
> -rw-r--r--  1 root root  399 2009-01-08 22:58 BackupPC_id_rsa.pub
> 
> r...@winserver:~/.ssh# exit

You forgot to "mv BackupPC_id_rsa.pub authorized_keys"

> logout
  root account can't run gedit in Ubuntu  <<<
> k...@winserver:~$ sudo gedit /root/.ssh/BackupPC_id_rsa.pub
> sudo: unable to resolve host Winserver
> [sudo] password for ken: 
> Added:  from="Archiver" at beginning of BackupPC_id_rsa.pub

I would suggest leaving out this restriction until after you get the
basic version working... Just don't forget to add it later.

> k...@winserver:~$ su -
> Password: 
> r...@winserver:~# cd /root/.ssh
> r...@winserver:~/.ssh# cat BackupPC_id_rsa.pub >> authorized_keys
> r...@winserver:~/.ssh# cat authorized_keys
> from="Archiver" ssh-rsa B...//...CzFk8lw== backu...@archiver
> r...@winserver:~/.ssh# 

OK, so should work now assuming the from="Archiver" is satisfied...

BTW, what is "ls -la /root/.ssh" on winserver showing?

 On Archiver:
> $ chmod -R go-rwx *
> $ ls -al
> total 24
> drwx-- 2 backuppc backuppc 4096 2009-01-08 22:58 .
> drwxr-xr-x 9 backuppc backuppc 4096 2009-01-08 12:46 ..
> -rw--- 1 backuppc backuppc  399 2009-01-08 22:57 BackupPC_id_rsa.pub
> -rw--- 1 backuppc backuppc 1675 2009-01-08 22:33 id_rsa
> -rw--- 1 backuppc backuppc  399 2009-01-08 22:33 id_rsa.pub
> -rw--- 1 backuppc backuppc  442 2009-01-08 22:58 known_hosts
> $ ssh -l root 192.168.1.101 whoami
> r...@192.168.1.101's password:<---Here it is asking for root password on
> Winserver
> root
> $

Try ssh -v -l root 192.168.1.101 whoami

Also, on winserver, check /etc/ssh/sshd_conf, I am pretty sure no
changes are needed, but it might be worth checking it, or posting the
contents.

Regards,
Adam

- --
Adam Goryachev
Website Managers
www.websitemanagers.com.au
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAklm5OYACgkQGyoxogrTyiXsUwCfQijKef2PBtoLvxnoays0XlKc
nEMAn0WpjMMtwWFDxMdscMAFRBxj+689
=waj0
-END PGP SIGNATURE-

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing lis

[BackupPC-users] [OT] Documentation practices (was: Re: vista backup question)

2009-01-08 Thread Jeffrey J. Kosowsky
Holger Parplies wrote at about 06:10:48 +0100 on Friday, January 9, 2009:
 > Hi,
 > 
 > Cody Dunne wrote on 2009-01-09 10:58:27 +0800 [Re: [BackupPC-users] vista 
 > backup question]:
 > > Holger Parplies wrote:
 > > > you keep quoting your guide. How about contributing the information (or 
 > > > at
 > > > least the pointer, but better the information) to the wiki?
 > > 
 > > I copied the excludes a while back to the Common_backup_excludes page on 
 > > the wiki, and keep it updated. If anyone wants to put a link to the 
 > > other parts of my guide on the wiki or even make a page for all of it, 
 > > they're welcome to as long as they cite the source. I'm not interested 
 > > in maintaining both copies, though.
 > 
 > I don't really care one way or the other. Personally, I don't visit random 
 > web
 > sites. If I was looking for information on BackupPC, I'd look at the wiki.
 > When I give references, those will be to the wiki. If your guide was part of
 > the wiki, I'd look at it, reference it, and maybe make suggestions for
 > improvements (or add them to the wiki myself). As it isn't, I won't (I don't
 > reference what I haven't looked at).
 > 
 > The BackupPC software is open source and available through sourceforge. If 
 > you
 > prefer to offer documentation elsewhere, that is your choice (you may have
 > good reasons for doing so - I don't know). Providing documentation is
 > commendable. Making it easily available would be even more so. But that's 
 > just
 > my opinion.
 > 
 > There is a lot of information available on the mailing list. Much of it
 > doesn't get copied into the wiki. I'm not sure why, but I'm sure that's a
 > pity. And it's no coincidence I explicitly cc'ed Jeffrey ;-).
 > 

Hmmm... I had to think for a few moments to understand the
reference... I assume you are referring to my comments a couple of
days ago about how people keep asking the same questions and that maybe
it should be added to the FAQ.

My problem is that I am from the pre-Wiki and pre-HTML generation and
am a bit of a Luddite in never getting into them or figuring out
how/why to edit them. So, I was thinking in the context of the old
command-and-control ASCII-based FAQ list where it is owned by one
person and edited in emacs (or if you must vi). Maybe I should figure
out what this whole Wiki/HTML/WYSIWYG thing is all about one
day... (for full disclosure, I also never use the web interface for
BackupPC except as a way to pretty-print status and backup history
since I much prefer CLI's and editing config files by hand -- hence
why I was pushing for a fusefs representation of backups rather than
the slow and clunky web navigation).

Still as a general comment, it would be nice if new users would at
least tried to search the archive or even google before asking the
same questions. I'm not perfect, but I at least make an attempt ;)

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Holger Parplies
Hi Margret,

Kenneth L. Owen wrote on 2009-01-09 00:20:41 -0500 [Re: [BackupPC-users] It's 
me again, Margret! - ssh-keygen]:
> [...]
> Having gotten accustomed to Fedora, I prefer working as root when required
> instead of having to enter lines twice -- once without sudo, error, then
> retype it with sudo.


or 'sudo !!'

> Also, root can navigate to some directories that normal user can't go.

Right. 'sudo sh'.

Just some suggestions, because 'sudo' really is a good habit to get into.

> [...]
> Maybe I see the problem.  On the winserver end, I put the public key in
> authorized_keys2.  Maybe Ubuntu version uses a different name from Fedora.
> Maybe I should have put the key in authorized_keys.

I'd usually try that first. I'm not sure where the authorized_keys2 and
known_hosts2 files come from (presumably protocol version 2, but not always).
If in doubt, check your man pages, though the ssh man page organization is
admittedly confusing (sshd(8) describes the authorized_keys format!? Well, it
*does* make sense, but it's not exactly intuitive for the end user).

> [...]
> r...@winserver:~/.ssh# exit
> logout
> >>>  root account can't run gedit in Ubuntu  <<<

Yes, it can (but probably $DISPLAY was not set):

> k...@winserver:~$ sudo gedit /root/.ssh/BackupPC_id_rsa.pub
> sudo: unable to resolve host Winserver
> [sudo] password for ken: 
> Added:  from="Archiver" at beginning of BackupPC_id_rsa.pub

For the next time:

(echo -n 'from="Archiver" '; cat BackupPC_id_rsa.pub) >> authorized_keys

But before you do that, you could try *without* the from= restriction. Whether
or not your remote IP address resolves to the string "Archiver" or something
else (have you set a domain name?) depends on your name resolution setup.
The error message "sudo: unable to resolve host Winserver" suggests that there
are problems.

> [...]
> $ ssh -l root 192.168.1.101 whoami
> r...@192.168.1.101's password:<---Here it is asking for root password on
> Winserver

Where does that come from?

> root

Try

$ ssh -v -l root 192.168.1.101 whoami

and maybe something like

$ ssh -l root 192.168.1.101
Winserver# w

to find out what Winserver thinks your hostname is.

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Adam Goryachev
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Holger Parplies wrote:
>> $ ssh -l root 192.168.1.101 whoami
>> r...@192.168.1.101's password:<---Here it is asking for root password on
>> Winserver
> 
> Where does that come from?
> 
>> root
> 
> Try
> 
>   $ ssh -v -l root 192.168.1.101 whoami
> 
> and maybe something like
> 
>   $ ssh -l root 192.168.1.101
>   Winserver# w
> 
> to find out what Winserver thinks your hostname is.

BTW, before running the above command, you should be the user backuppc
on the backuppc server...
$ whoami
backuppc
$ ssh -v -l root 192.168.1.101 whoami

Regards,
Adam

- --
Adam Goryachev
Website Managers
www.websitemanagers.com.au
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAklm6LIACgkQGyoxogrTyiUVXwCgxD0scjmlCaDNgivlcRFCG+fT
vaEAnipjvvATf1wHI1vNP0O0yeAB7ole
=TWYN
-END PGP SIGNATURE-

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Jeffrey J. Kosowsky
Kenneth L. Owen wrote at about 00:20:41 -0500 on Friday, January 9, 2009:
 > Tino,
 > 
 > Having gotten accustomed to Fedora, I prefer working as root when required
 > instead of having to enter lines twice -- once without sudo, error, then
 > retype it with sudo.  
I consider that a *feature* rather than a bug -- many times it has
prevented me from doing some bone-headed damaging things (and by the
way I use sudo all the time on Fedora, only using root when I need a
full root shell environment or when I need to park myself in a
directory that is only root-readable)

Also, you don't need to retype -- use bash to recall the previous line
and just add sudo...
Type: sudo !!

Also, root can navigate to some directories that
 > normal user can't go.  Once I get more used to using sudo, I may put root
 > back as it was.  This was the ONLY reason for setting a root password.
 > 
 > The backuppc password is only used initially to start BackupPC CGI for
 > setup, I think.  I haven't gotten that far as yet.  BackupPC is installed,
 > config.pl has been edited, but backuppc is currently halted until I get the
 > keys set.  I have my LVM drive to house the backup files mounted and ready.
 > 
 > The Archiver machine will also be my daytime workstation in the workshop.
 > Therefore, I want to be able to do more than just run BackupPC at night.
 > Some software not pertaining to BackupPC will be needed.
 > 
 > Tino, in the original transcript, I didn't show work done on Winserver.
 > 
 > Maybe I see the problem.  On the winserver end, I put the public key in
 > authorized_keys2.  Maybe Ubuntu version uses a different name from Fedora.
 > Maybe I should have put the key in authorized_keys.
Shouldn't make a difference.
Listen, I think it would save you time and hassle if you invested some
time in reading up on ssh rather than trying to get spoon-fed
step-by-step. SSH is a pretty basic utility that has many uses and you
might as well lean how to use it right now since it can be very
helpful in many machine-to-machine communication and admin
activities. Plus, getting it right is critical to your machine
security, so you might as well understand what you are doing. It's not
rocket science -- it just requires a little stepping back and
understanding what public key encryption is all about.

 > 
 > I tried using authorized_keys and got the same result.  The transcript
 > follows.  (showing work on both ends) -- ken
 > 
 > >>> On Archiver:  ip: 192.168.1.106
 > k...@archiver:~$ su -
 > Password: 
 > r...@archiver:~# su backuppc
 > $ whoami
 > backuppc
 > $ ssh-keygen -t rsa
 > Generating public/private rsa key pair.
 > Enter file in which to save the key (/var/lib/backuppc/.ssh/id_rsa): 
 > Enter passphrase (empty for no passphrase): 
 > Enter same passphrase again: 
 > Your identification has been saved in /var/lib/backuppc/.ssh/id_rsa.
 > Your public key has been saved in /var/lib/backuppc/.ssh/id_rsa.pub.
 > The key fingerprint is:
 > 08:2c:0f:a6:10:c6:6f:8a:0b:e9:1c:a9:1a:e3:30:39 backu...@archiver
 > 
 > $ scp BackupPC_id_rsa.pub r...@192.168.1.101:/root/.ssh/
 > The authenticity of host '192.168.1.101 (192.168.1.101)' can't be
 > established.
 > RSA key fingerprint is 71:a1:03:7d:fb:b9:87:1f:32:c7:a3:46:d0:81:2d:af.
 > Are you sure you want to continue connecting (yes/no)? yes
 > Warning: Permanently added '192.168.1.101' (RSA) to the list of known hosts.
 > r...@192.168.1.101's password: 
 > BackupPC_id_rsa.pub
 > 100%  399 0.4KB/s   00:00
 > 
 > >>> On Winserver:  ip:  192.168.1.101 (Also running Ubuntu)
 > k...@winserver:~$ su -
 > Password: 
 > r...@winserver:~# cd /root/.ssh
 > r...@winserver:~/.ssh# ls -al
 > total 12
 > drwx--  2 root root 4096 2009-01-08 22:58 .
 > drwxr-xr-x 10 root root 4096 2009-01-08 22:39 ..
 > -rw-r--r--  1 root root  399 2009-01-08 22:58 BackupPC_id_rsa.pub
 > 
 > r...@winserver:~/.ssh# exit
 > logout
 > >>>  root account can't run gedit in Ubuntu  <<<
 > k...@winserver:~$ sudo gedit /root/.ssh/BackupPC_id_rsa.pub
 > sudo: unable to resolve host Winserver
 > [sudo] password for ken: 
 > Added:  from="Archiver" at beginning of BackupPC_id_rsa.pub
 > 
 > k...@winserver:~$ su -
 > Password: 
 > r...@winserver:~# cd /root/.ssh
 > r...@winserver:~/.ssh# cat BackupPC_id_rsa.pub >> authorized_keys
 > r...@winserver:~/.ssh# cat authorized_keys
 > from="Archiver" ssh-rsa B...//...CzFk8lw== backu...@archiver
 > r...@winserver:~/.ssh# 
 > 
 > >>> On Archiver:
 > $ chmod -R go-rwx *
 > $ ls -al
 > total 24
 > drwx-- 2 backuppc backuppc 4096 2009-01-08 22:58 .
 > drwxr-xr-x 9 backuppc backuppc 4096 2009-01-08 12:46 ..
 > -rw--- 1 backuppc backuppc  399 2009-01-08 22:57 BackupPC_id_rsa.pub
 > -rw--- 1 backuppc backuppc 1675 2009-01-08 22:33 id_rsa
 > -rw--- 1 backuppc backuppc  399 2009-01-08 22:33 id_rsa.pub
 > -rw--- 1 backuppc backuppc  442 2009-01-08 22:58 known_hosts
 > $ ssh -l root 192.168.1.101 whoami
 > r...@192.168.1.101's password:<---Here it is asking fo

[BackupPC-users] OT - for Holger Parplies

2009-01-08 Thread Jeffrey J. Kosowsky
Holger, I have at times tried to respond directly to your
communications but for some reason it gets returned saying 
"Your address is permanently blacklisted. Choose another provider."
Which by the way seems a bit presumptuous and rude of your server ;)
just kidding.

Any suggestions short of changing my ISP?
(I am obviously writing to the list because I have no other way of
contacting you personally)

Verizon Online Postmaster wrote at about 23:59:21 -0600 on Thursday, January 8, 
2009:
 > This report relates to a message you sent with the following header fields:
 > 
 >   Message-id: <18790.59150.218669.147...@consult.pretender>
 >   Date: Fri, 09 Jan 2009 00:56:30 -0500
 >   From: "Jeffrey J. Kosowsky" 
 >   To: Holger Parplies 
 >   Subject: [OT] Documentation practices (was: Re: [BackupPC-users] vista 
 > backup
 >question)
 > 
 > Your message cannot be delivered to the following recipients:
 > 
 >   Recipient address: wb...@parplies.de
 >   Reason: Rejection greeting returned by server.
 >   Diagnostic code: smtp;550 Your address is permanently blacklisted. Choose 
 > another provider.
 >   Remote system: dns;mail.parplies.de 
 > (TCP|206.46.173.3|53880|217.9.60.250|25) (Your address is permanently 
 > blacklisted. Choose another provider.)
 > 
 > 
 > --
 > Reporting-MTA: dns;vms173003.mailsrvcs.net (tcp-daemon)
 > 
 > Original-recipient: rfc822;wb...@parplies.de
 > Final-recipient: rfc822;wb...@parplies.de
 > Action: failed
 > Status: 5.0.0 (Rejection greeting returned by server.)
 > Remote-MTA: dns;mail.parplies.de (TCP|206.46.173.3|53880|217.9.60.250|25)
 >  (Your address is permanently blacklisted. Choose another provider.)
 > Diagnostic-code: smtp;550 Your address is permanently blacklisted. Choose
 >  another provider.
 > 
 > --
 > From: "Jeffrey J. Kosowsky" 
 > To: Holger Parplies 
 > Cc: Cody Dunne ,
 >  "Jeffrey J. Kosowsky" ,
 >  "General list for user discussion, questions and support"
 >  
 > Subject: [OT] Documentation practices (was: Re: [BackupPC-users] vista backup

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] It's me again, Margret! - ssh-keygen

2009-01-08 Thread Holger Parplies
Hi,

Adam Goryachev wrote on 2009-01-09 16:47:18 +1100 [Re: [BackupPC-users] It's me 
again, Margret! - ssh-keygen]:
> [...]
> > r...@winserver:~/.ssh# cat BackupPC_id_rsa.pub >> authorized_keys
> [...]
> 
> BTW, what is "ls -la /root/.ssh" on winserver showing?

this *could be* the problem. Ideally, authorized_keys should not be readable
for anyone except root, though I seem to get by with world-readable
authorized_keys in world accessible .ssh directories. I believe the
requirements of ssh are that the home directory(!), the .ssh directory and the
authorized_keys file must all be writeable for the user only (config setting
"StrictModes", sparcely documented).

Regards,
Holger

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] OK, question that I have no clue where to look for an answer...

2009-01-08 Thread James Beam
I have Backuppc working finally on my Ubuntu box - working like a charm for 
pc's and our servers.

Problem:

We have 7 users that are on laptops, they take them home every evening and 
sometimes travel so they aren't even in the office.

Is there a way to setup some kind of a test to see if a given laptop is 
connected to the local network (not when they are on travel obviously) and 
if the laptop is present, do a full or incremental backup at that time - 
every day?

Sometimes even I do not know when they are in the office.

Any advice on where to look for this info (if it exists) would be much 
appreciated. 



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


Re: [BackupPC-users] OK, question that I have no clue where to look for an answer...

2009-01-08 Thread Les Mikesell
James Beam wrote:
> I have Backuppc working finally on my Ubuntu box - working like a charm for 
> pc's and our servers.
> 
> Problem:
> 
> We have 7 users that are on laptops, they take them home every evening and 
> sometimes travel so they aren't even in the office.
> 
> Is there a way to setup some kind of a test to see if a given laptop is 
> connected to the local network (not when they are on travel obviously) and 
> if the laptop is present, do a full or incremental backup at that time - 
> every day?
> 
> Sometimes even I do not know when they are in the office.
> 
> Any advice on where to look for this info (if it exists) would be much 
> appreciated. 

That should just happen by default, unless they are connected and 
powered up regularly outside of the blackout time.  Once they have been 
seen some (configurable) number of times outside of the blackout, 
backuppc will stop checking for them at other times.   In any case it 
may be a better approach to set up web logins for the laptop users so 
they can start backups at convenient times (like when leaving for lunch) 
instead of having the backup run consume their resources randomly.

-- 
   Les Mikesell
lesmikes...@gmail.com

--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/


[BackupPC-users] host config file command line editor ?

2009-01-08 Thread Alex

Thank you guys for your fast answers.

Sorry for the html, as i posted it through the backupcentral forum, it modified 
the config part in html, didn't expect that.

Well, i'm gonna look the given solutions to get it work.

Sorry for my confuse question.

I know about the arrays etc. The goal is for me to change settings, even if 
it's a single value, or a complex array by replacing the all param in the 
config file.

I already know how to get a well formed array to reproduce perl's one from bash 
or php.

My first problem was erasing the specific area in the configuration file (only 
the modified value).

I'm gonna try to make :
1 : remove the desired value from config file (by any of the solutions you tell 
me above)
2 : add new configuration in the file (by ex in shell : echo 
"$Conf{FullKeepCnt} = ['400']; ">>/etc/BackupPC/pc/my_config_file.pl  (just to 
tell how i plan, i know i'll have to add \ before special chars.)

I think you see the way i wanna make it work.
Sorry for my bad english ;)

+--
|This was sent by alexan...@inforeseau.com via Backup Central.
|Forward SPAM to ab...@backupcentral.com.
+--



--
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
___
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List:https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/