Re: zmanda setup Issues: farkled permissions

[Gene Heskett]

On Sunday 26 May 2019 07:26:47 pm Nathan Stratton Treadway wrote:

> On Sun, May 26, 2019 at 14:02:18 -0400, Nathan Stratton Treadway wrote:
> > On Sun, May 26, 2019 at 11:47:14 -0600, Charles Curley wrote:
> > > Having create a set of zmanda debian packages on one debian 9.9
> > > machine, I have copied those to another freshly built machine, and
> > > installed them.
> > >
> > > I seem to hit a problem with permissions and ownership
> > > of /etc/amanda-security.conf.
> > >
> > > amcheck wants /etc/amanda-security.conf, but the file is in
> > > /etc/amanda, not /etc. That one cannot solve with a symlink. Copy
> > > the file instead.
> >
> > (This sounds like an Debian-package-build issue; I'll see if I can
> > spot an obvious problem in the scripts Chris has been working on.
> >
> > You installed the amanda-backup-server package on this new VM,
> > right?)
>
> (Chris,)
>
> Okay, the first part of the problem is that the Makefiles do not take
> the --with-security-file setting into account when installing the
> amanda-security.conf during the "installation" phase.
>
> (I haven't actually tried it myself, but I believe this would apply to
> all builds, i.e. including non-Debian-package builds.)
>
> I don't know enough about "configure" and friends to know the best way
> to make the setting propagate to the install phase, but looks like the
> issue is in .../common-src/Makefile, where there is the following
> sequence of lines:
>
>   CONFIG_DIR = /etc/amanda
>   [...]
>   etcamandadir = ${CONFIG_DIR}
>   etcamanda_FILE = amanda-security.conf
>   etcamanda_DATA = ${etcamanda_FILE}
>   [...]
>
>   install-etcamandaDATA: $(etcamanda_DATA)
> @$(NORMAL_INSTALL)
> @list='$(etcamanda_DATA)'; test -n "$(etcamandadir)" || list=;
> \ if test -n "$$list"; then \
>   echo " $(MKDIR_P) '$(DESTDIR)$(etcamandadir)'"; \
>   $(MKDIR_P) "$(DESTDIR)$(etcamandadir)" || exit 1; \
> fi; \
> for p in $$list; do \
>   if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
>   echo "$$d$$p"; \
> done | $(am__base_list) | \
> while read files; do \
>   echo " $(INSTALL_DATA) $$files '$(DESTDIR)$(etcamandadir)'";
> \ $(INSTALL_DATA) $$files "$(DESTDIR)$(etcamandadir)" || exit $$?; \
> done
>
> So, if I follow everything correctly, the problem is that the whole
> "install-etcamandaDATA" section of the script is devoted to installing
> each file in "${etcamanda_DATA}" files into "${etcamandadir}"... but
> actually the only file in that list is amanda-security.conf, the path
> to which you are supposed to be able to specify with the
> --with-security-file= option
>
> (Actually, given that .../config/amanda/defaults.m4 puts the file in
> SYSCONFDIR by default, it seems like it will be extremely rare that
> installing it in $CONFIG_DIR is the correct thing to do...)
>
> Perhaps the whole section can just be switched to install the
> "amanda-security.conf" directly to the "./configure"-calculated
> "security-file" path?
>
>
>
> Anyway, if that isn't going to be easy to fix at the Makefile level,
> you may want to implement a short-term fix the Zmanda .deb files to
> work around this problem
>
> The official Debian package works around it by using the line
>
>   etc/amanda/amanda-security.conf etc
>
> in amanda-client.install, rather than the
>
>   etc/amanda/*
>
> line currently found in both packaging/deb/amanda-backup-*.install
> files.
>
> I think making this change in your two files would work okay, too,
> since currently it seems the packages do not install any other files
> in /etc/amanda (so eliminating the wildcard would not cause any other
> files to be skipped):
>
>   $ dpkg-deb -c
> amanda-backup-server_3.5.1.git.19364c7b-1Debian99_amd64.deb | grep
> /etc/amanda drwxrwx--- amandabackup/disk 0 2019-05-15 16:45
> ./etc/amanda/ -rw-rw amandabackup/disk 2034 2019-05-15 16:45
> ./etc/amanda/amanda-security.conf $ dpkg-deb -c
> amanda-backup-client_3.5.1.git.19364c7b-1Debian99_amd64.deb | grep
> /etc/amanda drwxrwx--- amandabackup/disk 0 2019-05-15 16:45
> ./etc/amanda/ -rw-rw amandabackup/disk 2034 2019-05-15 16:45
> ./etc/amanda/amanda-security.conf
>
>
>
>   Nathan
Well I am down to 2 errors from amcheck

ERROR: program /usr/local/libexec/amanda/ambind: wrong permission, must 
be 'rwsr-x---'
But that is what it almost is:
-rwsr-x--x 1 root disk 26840 May 26 
21:59 /usr/local/libexec/amanda/ambind
It won't get past that point for anything if that final x isn't there

And:
ERROR: coyote: selfcheck request failed: Connection refused

According to the amanda file in /e/xinetd.d
=
# default: on
#
# description: Amanda services for Amanda server and client.
#

service amanda
{
disable = no
flags   = IPv4
socket_type = stream
protocol= tcp
wait= no

Re: zmanda setup Issues: farkled permissions

[Nathan Stratton Treadway]

On Sun, May 26, 2019 at 14:02:18 -0400, Nathan Stratton Treadway wrote:
> On Sun, May 26, 2019 at 11:47:14 -0600, Charles Curley wrote:
> > Having create a set of zmanda debian packages on one debian 9.9 machine,
> > I have copied those to another freshly built machine, and installed
> > them.
> > 
> > I seem to hit a problem with permissions and ownership
> > of /etc/amanda-security.conf.
> > 
> > amcheck wants /etc/amanda-security.conf, but the file is in /etc/amanda,
> > not /etc. That one cannot solve with a symlink. Copy the file instead.
> 
> (This sounds like an Debian-package-build issue; I'll see if I can spot
> an obvious problem in the scripts Chris has been working on.
> 
> You installed the amanda-backup-server package on this new VM, right?)

(Chris,)

Okay, the first part of the problem is that the Makefiles do not take
the --with-security-file setting into account when installing the
amanda-security.conf during the "installation" phase.  

(I haven't actually tried it myself, but I believe this would apply to
all builds, i.e. including non-Debian-package builds.)

I don't know enough about "configure" and friends to know the best way
to make the setting propagate to the install phase, but looks like the
issue is in .../common-src/Makefile, where there is the following
sequence of lines:

  CONFIG_DIR = /etc/amanda
  [...]
  etcamandadir = ${CONFIG_DIR}
  etcamanda_FILE = amanda-security.conf
  etcamanda_DATA = ${etcamanda_FILE}
  [...]

  install-etcamandaDATA: $(etcamanda_DATA)
@$(NORMAL_INSTALL)
@list='$(etcamanda_DATA)'; test -n "$(etcamandadir)" || list=; \
if test -n "$$list"; then \
  echo " $(MKDIR_P) '$(DESTDIR)$(etcamandadir)'"; \
  $(MKDIR_P) "$(DESTDIR)$(etcamandadir)" || exit 1; \
fi; \
for p in $$list; do \
  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
  echo "$$d$$p"; \
done | $(am__base_list) | \
while read files; do \
  echo " $(INSTALL_DATA) $$files '$(DESTDIR)$(etcamandadir)'"; \
  $(INSTALL_DATA) $$files "$(DESTDIR)$(etcamandadir)" || exit
  $$?; \
done

So, if I follow everything correctly, the problem is that the whole
"install-etcamandaDATA" section of the script is devoted to installing
each file in "${etcamanda_DATA}" files into "${etcamandadir}"... but
actually the only file in that list is amanda-security.conf, the path to
which you are supposed to be able to specify with the
--with-security-file= option

(Actually, given that .../config/amanda/defaults.m4 puts the file in
SYSCONFDIR by default, it seems like it will be extremely rare that
installing it in $CONFIG_DIR is the correct thing to do...)

Perhaps the whole section can just be switched to install the
"amanda-security.conf" directly to the "./configure"-calculated
"security-file" path?



Anyway, if that isn't going to be easy to fix at the Makefile level, you
may want to implement a short-term fix the Zmanda .deb files to work
around this problem

The official Debian package works around it by using the line

  etc/amanda/amanda-security.conf etc

in amanda-client.install, rather than the 

  etc/amanda/*

line currently found in both packaging/deb/amanda-backup-*.install
files.

I think making this change in your two files would work okay, too, since
currently it seems the packages do not install any other files in
/etc/amanda (so eliminating the wildcard would not cause any other files
to be skipped):

  $ dpkg-deb -c amanda-backup-server_3.5.1.git.19364c7b-1Debian99_amd64.deb | 
grep /etc/amanda
  drwxrwx--- amandabackup/disk 0 2019-05-15 16:45 ./etc/amanda/
  -rw-rw amandabackup/disk 2034 2019-05-15 16:45 
./etc/amanda/amanda-security.conf
  $ dpkg-deb -c amanda-backup-client_3.5.1.git.19364c7b-1Debian99_amd64.deb | 
grep /etc/amanda
  drwxrwx--- amandabackup/disk 0 2019-05-15 16:45 ./etc/amanda/
  -rw-rw amandabackup/disk 2034 2019-05-15 16:45 
./etc/amanda/amanda-security.conf



Nathan



Nathan Stratton Treadway  -  natha...@ontko.com  -  Mid-Atlantic region
Ray Ontko & Co.  -  Software consulting services  -   http://www.ontko.com/
 GPG Key: http://www.ontko.com/~nathanst/gpg_key.txt   ID: 1023D/ECFB6239
 Key fingerprint = 6AD8 485E 20B9 5C71 231C  0C32 15F3 ADCD ECFB 6239

Re: zmanda setup Issues: farkled permissions

[Gene Heskett]

On Sunday 26 May 2019 01:47:14 pm Charles Curley wrote:

> Having create a set of zmanda debian packages on one debian 9.9
> machine, I have copied those to another freshly built machine, and
> installed them.
>
> I seem to hit a problem with permissions and ownership
> of /etc/amanda-security.conf.
>
> amcheck wants /etc/amanda-security.conf, but the file is in
> /etc/amanda, not /etc. That one cannot solve with a symlink. Copy the
> file instead.
>
Or use my configure script and give it the full path so it defeats the 
defaults.

> amcheck also wants it to be owned by root, and not writable by the
> group.
>
> Having done that I get another error message complaining:
>
> Amanda Backup Client Hosts Check
> 
> HOST localhost ERROR: can not access '/etc/amanda-security.conf':
> Permission denied (ruid:63998 euid:63998) HOST localhost ERROR: can
> not access '/etc/amanda-security.conf': Permission denied (ruid:63998
> euid:63998) Client check: 1 host checked in 2.169 seconds.  2 problems
> found.
>
> Make up your mind: who do you want to own /etc/amanda-security.conf?
>
> xinetd runs the client as
>
> user= amandabackup
> group   = disk
Whereas I have user=amanda
and   group=backup

and backup is a member of group disk.

Copyright 2019 by Maurice E. Heskett
Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

Re: zmanda setup Issues: farkled permissions

[Charles Curley]

On Sun, 26 May 2019 14:02:18 -0400
Nathan Stratton Treadway  wrote:

> > amcheck wants /etc/amanda-security.conf, but the file is
> > in /etc/amanda, not /etc. That one cannot solve with a symlink.
> > Copy the file instead.  
> 
> (This sounds like an Debian-package-build issue; I'll see if I can
> spot an obvious problem in the scripts Chris has been working on.
> 
> You installed the amanda-backup-server package on this new VM, right?)

Correct. Brand new, freshly built debian 9.9.


> If I remember Jean-Louis's explaination from a couple years ago, the
> deal is that it must NOT be writable by anyone but root, but at the
> same time it must be *readable* by your amanda user (so that amanda
> programs can read it).
> 
> What does "ls -l /etc/amanda-security.conf" show now?

root@amanda2:/etc# ls -l /etc/amanda-security.conf
-rw--- 1 root root 2034 May 15 14:45 /etc/amanda-security.conf
root@amanda2:/etc# 

Based on Jean-Louis' comment above, I did:

root@amanda2:/etc# chmod og+r amanda-security.conf 
root@amanda2:/etc# ls -l /etc/amanda-security.conf
-rw-r--r-- 1 root root 2034 May 15 14:45 /etc/amanda-security.conf
root@amanda2:/etc#

and amcheck ran. Thank you.


-- 
"When we talk of civilization, we are too apt to limit the meaning of
the word to its mere embellishments, such as arts and sciences; but
the true distinction between it and barbarism is, that the one
presents a state of society under the protection of just and
well-administered law, and the other is left to the chance government
of brute force."
- The Rev. James White, Eighteen Christian Centuries, 1889
Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB
https://charlescurley.com

Re: zmanda setup Issues: farkled permissions

[Nathan Stratton Treadway]

On Sun, May 26, 2019 at 11:47:14 -0600, Charles Curley wrote:
> Having create a set of zmanda debian packages on one debian 9.9 machine,
> I have copied those to another freshly built machine, and installed
> them.
> 
> I seem to hit a problem with permissions and ownership
> of /etc/amanda-security.conf.
> 
> amcheck wants /etc/amanda-security.conf, but the file is in /etc/amanda,
> not /etc. That one cannot solve with a symlink. Copy the file instead.

(This sounds like an Debian-package-build issue; I'll see if I can spot
an obvious problem in the scripts Chris has been working on.

You installed the amanda-backup-server package on this new VM, right?)

 
> amcheck also wants it to be owned by root, and not writable by the
> group.
>
> 
> Having done that I get another error message complaining:
> 
> Amanda Backup Client Hosts Check
> 
> HOST localhost ERROR: can not access '/etc/amanda-security.conf': Permission 
> denied (ruid:63998 euid:63998)
> HOST localhost ERROR: can not access '/etc/amanda-security.conf': Permission 
> denied (ruid:63998 euid:63998)
> Client check: 1 host checked in 2.169 seconds.  2 problems found.
> 
> Make up your mind: who do you want to own /etc/amanda-security.conf?

If I remember Jean-Louis's explaination from a couple years ago, the
deal is that it must NOT be writable by anyone but root, but at the same
time it must be *readable* by your amanda user (so that amanda programs
can read it).

What does "ls -l /etc/amanda-security.conf" show now?

Nathan




Nathan Stratton Treadway  -  natha...@ontko.com  -  Mid-Atlantic region
Ray Ontko & Co.  -  Software consulting services  -   http://www.ontko.com/
 GPG Key: http://www.ontko.com/~nathanst/gpg_key.txt   ID: 1023D/ECFB6239
 Key fingerprint = 6AD8 485E 20B9 5C71 231C  0C32 15F3 ADCD ECFB 6239

zmanda setup Issues: farkled permissions

[Charles Curley]

Having create a set of zmanda debian packages on one debian 9.9 machine,
I have copied those to another freshly built machine, and installed
them.

I seem to hit a problem with permissions and ownership
of /etc/amanda-security.conf.

amcheck wants /etc/amanda-security.conf, but the file is in /etc/amanda,
not /etc. That one cannot solve with a symlink. Copy the file instead.

amcheck also wants it to be owned by root, and not writable by the
group.

Having done that I get another error message complaining:

Amanda Backup Client Hosts Check

HOST localhost ERROR: can not access '/etc/amanda-security.conf': Permission 
denied (ruid:63998 euid:63998)
HOST localhost ERROR: can not access '/etc/amanda-security.conf': Permission 
denied (ruid:63998 euid:63998)
Client check: 1 host checked in 2.169 seconds.  2 problems found.

Make up your mind: who do you want to own /etc/amanda-security.conf?

xinetd runs the client as

user= amandabackup
group   = disk



-- 
"When we talk of civilization, we are too apt to limit the meaning of
the word to its mere embellishments, such as arts and sciences; but
the true distinction between it and barbarism is, that the one
presents a state of society under the protection of just and
well-administered law, and the other is left to the chance government
of brute force."
- The Rev. James White, Eighteen Christian Centuries, 1889
Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB
https://charlescurley.com

Re: "virtual tapes" vs leave it on spool -temporary setup

[Debra S Baddorf]

> On Nov 2, 2016, at 5:16 PM, Stefan G. Weichinger  wrote:
> 
> Am 2016-11-02 um 22:21 schrieb Debra S Baddorf:
>> Good to know you’ve done it before.
>> 
>> It takes a long time for government to buy anything.  A month is 
>> actually an awfully
>> optimistic estimate, now that I think about it!   
>> 
>> But I’m told this temporary node has  20+ 2TB disks,  and yes I’ll just do 
>> one full and then incrementals. 
>> Probably of the user-data areas only,  not the OS’s on all 9 or more clients.
>> I’ll just keep adding more spool areas as the first setup gets full.   My 
>> current backups
>> of these clients (different software; now expired!)   is taking 3-4 LTO4 
>> tapes for a level 0 backup.
> 
> I would suggest to adjust the "reserve" parameter.
> 
> And maybe the cronjobs: "amdump --no-taper" ... to keep the noise low ;-)


I usually set the reserves to  02%

That’s a great idea that I hadn’t yet thought of for the amdump command.   I 
use that when manually
re-running a dump of a single node — but hadn’t thought of it for THIS instance.

Thanks-
Deb 

Re: "virtual tapes" vs leave it on spool -temporary setup

[Stefan G. Weichinger]

Am 2016-11-02 um 22:21 schrieb Debra S Baddorf:
> Good to know you’ve done it before.
> 
> It takes a long time for government to buy anything.  A month is actually 
> an awfully
> optimistic estimate, now that I think about it!   
> 
> But I’m told this temporary node has  20+ 2TB disks,  and yes I’ll just do 
> one full and then incrementals. 
> Probably of the user-data areas only,  not the OS’s on all 9 or more clients.
> I’ll just keep adding more spool areas as the first setup gets full.   My 
> current backups
> of these clients (different software; now expired!)   is taking 3-4 LTO4 
> tapes for a level 0 backup.

I would suggest to adjust the "reserve" parameter.

And maybe the cronjobs: "amdump --no-taper" ... to keep the noise low ;-)

Re: "virtual tapes" vs leave it on spool -temporary setup

[Debra S Baddorf]

Good to know you’ve done it before.

It takes a long time for government to buy anything.  A month is actually 
an awfully
optimistic estimate, now that I think about it!   

But I’m told this temporary node has  20+ 2TB disks,  and yes I’ll just do one 
full and then incrementals. 
Probably of the user-data areas only,  not the OS’s on all 9 or more clients.
I’ll just keep adding more spool areas as the first setup gets full.   My 
current backups
of these clients (different software; now expired!)   is taking 3-4 LTO4 tapes 
for a level 0 backup.

Deb Baddorf


> On Nov 2, 2016, at 4:10 PM, Chris Hoogendyk  wrote:
> 
> I did exactly that, and then flushed it a couple of times when I got the tape 
> drive replaced. However, I was only a week or so running that way, not a 
> month.
> 
> For my setup, I have an LTO6 tape library, and a Supermicro server running 
> Ubuntu 14.04. I have two 1TB enterprise SSDs for holding disk. During the 
> days that my tape drive failed, I set up extra disk space that was available 
> for a few more holding disks totaling something like 6TB more space. For a 
> couple of days, I was running incrementals only. That is backing up 4 servers 
> with a total of something like 20TB of actual data (as opposed to capacity) 
> divided up into just over 100 DLEs.
> 
> 
> On 11/2/16 4:31 PM, Debra S Baddorf wrote:
>> For a temporary setup,  until I get a new computer to attach to my tape 
>> drive ….
>> (perhaps a month)
>> is there any reason I can’t just have a LOT of spool space(s),   and let data
>> accumulate there?
>> 
>> The alternative is creating virtual-tapes on those same disks.  But that 
>> seems
>> like more work than just leaving the data in the spool area.   Is this a bad 
>> idea?
>> 
>> Deb Baddorf
>> Fermilab
>> 
> 
> -- 
> ---
> 
> Chris Hoogendyk
> 
> -
>   O__   Systems Administrator
>  c/ /'_ --- Biology & Geosciences Departments
> (*) \(*) -- 315 Morrill Science Center
> ~~ - University of Massachusetts, Amherst
> 
> 
> 
> ---
> 
> Erdös 4
> 

Re: "virtual tapes" vs leave it on spool -temporary setup

[Chris Hoogendyk]

I did exactly that, and then flushed it a couple of times when I got the tape drive replaced. 
However, I was only a week or so running that way, not a month.


For my setup, I have an LTO6 tape library, and a Supermicro server running Ubuntu 14.04. I have two 
1TB enterprise SSDs for holding disk. During the days that my tape drive failed, I set up extra disk 
space that was available for a few more holding disks totaling something like 6TB more space. For a 
couple of days, I was running incrementals only. That is backing up 4 servers with a total of 
something like 20TB of actual data (as opposed to capacity) divided up into just over 100 DLEs.



On 11/2/16 4:31 PM, Debra S Baddorf wrote:

For a temporary setup,  until I get a new computer to attach to my tape drive ….
(perhaps a month)
is there any reason I can’t just have a LOT of spool space(s),   and let data
accumulate there?

The alternative is creating virtual-tapes on those same disks.  But that seems
like more work than just leaving the data in the spool area.   Is this a bad 
idea?

Deb Baddorf
Fermilab



--
---

Chris Hoogendyk

-
   O__   Systems Administrator
  c/ /'_ --- Biology & Geosciences Departments
 (*) \(*) -- 315 Morrill Science Center
~~ - University of Massachusetts, Amherst



---

Erdös 4

Re: "virtual tapes" vs leave it on spool -temporary setup

[Stefan G. Weichinger]

Am 2016-11-02 um 21:31 schrieb Debra S Baddorf:
> For a temporary setup,  until I get a new computer to attach to my tape drive 
> ….
> (perhaps a month)
> is there any reason I can’t just have a LOT of spool space(s),   and let data
> accumulate there?
> 
> The alternative is creating virtual-tapes on those same disks.  But that seems
> like more work than just leaving the data in the spool area.   Is this a bad 
> idea?

No, I don't think so.

"virtual tapes" vs leave it on spool -temporary setup

[Debra S Baddorf]

For a temporary setup,  until I get a new computer to attach to my tape drive ….
(perhaps a month)
is there any reason I can’t just have a LOT of spool space(s),   and let data
accumulate there?

The alternative is creating virtual-tapes on those same disks.  But that seems
like more work than just leaving the data in the spool area.   Is this a bad 
idea?

Deb Baddorf
Fermilab

Re: Disk and Tape Setup Config

[Stefan G. Weichinger]

Am 2016-09-01 um 20:10 schrieb Debra S Baddorf:



On Sep 1, 2016, at 12:32 PM, Ryan Chase  wrote:

Hi,
 I'm new to Amanda and I'm finding the Documentation difficult to
navigate. I have a large nas disk to use for backup as well as a dell
tape library. I'd like to backup to disk and also to tape for offsite
backups. What is the best configuration scenario for this?

I've read about a holding disk and virtual tapes for disk, but haven't
seen anything that specifically addresses a disk and tape scenario.

Can someone provide guidance and/or a link to documentation for setting
up such a scenario?

--Ryan



You could either do a separate configuration,  and a whole separate backup to 
tape —
or,  lookup  “amarchive”  which copies a disk backup to a tape.


Additional note: I agree, some better howtos/docs/etc would really help 
users get started using amanda's features. Especially when it comes to 
the combined usage of tapes/disks/libraries (the scenarios change 
slightly from tape-only to more disk-based setups).


Some time ago (see thread "tapes and external disks: how to combine all 
this" above ...) I somehow talked about writing such a howto ;-)  Maybe 
this is a good reminder.


Stefan

Re: Disk and Tape Setup Config

[Debra S Baddorf]

> On Sep 1, 2016, at 12:32 PM, Ryan Chase  wrote:
> 
> Hi,
>  I'm new to Amanda and I'm finding the Documentation difficult to
> navigate. I have a large nas disk to use for backup as well as a dell
> tape library. I'd like to backup to disk and also to tape for offsite
> backups. What is the best configuration scenario for this?
> 
> I've read about a holding disk and virtual tapes for disk, but haven't
> seen anything that specifically addresses a disk and tape scenario.
> 
> Can someone provide guidance and/or a link to documentation for setting
> up such a scenario?
> 
> --Ryan


You could either do a separate configuration,  and a whole separate backup to 
tape —
or,  lookup  “amarchive”  which copies a disk backup to a tape.

Deb Baddorf
Fermilab

Disk and Tape Setup Config

[Ryan Chase]

Hi,
  I'm new to Amanda and I'm finding the Documentation difficult to
navigate. I have a large nas disk to use for backup as well as a dell
tape library. I'd like to backup to disk and also to tape for offsite
backups. What is the best configuration scenario for this?

I've read about a holding disk and virtual tapes for disk, but haven't
seen anything that specifically addresses a disk and tape scenario.

Can someone provide guidance and/or a link to documentation for setting
up such a scenario?

--Ryan

Re: setup 8 LTO-4 Tapes to backup 21 days or more

[Jon LaBadie]

Spam detection software, running on the system "slivovice.cz.omniscient.com", 
has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
The administrator of that system for details.

Content preview:  On Thu, Sep 19, 2013 at 10:07:47AM +0200, Stefan Nitz wrote:
   > Hello list, > > i want to to use 8 LTO-4 (400GiB) Tapes to backup 21 days.
   > Best with one full-backup every 7 days. > I've to backup about 1TiB 
fullbackup,
   150 GiB each increment. > > Could anyone gave some hints to setup amanda,
   pls. > I read all man-pages, i thought, but i dont understand how to do that.
   > [...] 

Content analysis details:   (5.0 points, 5.0 required)

 pts rule name  description
 -- --
-0.0 RCVD_IN_DNSWL_NONE RBL: Sender listed at http://www.dnswl.org/, no
trust
[206.46.173.7 listed in list.dnswl.org]
 4.0 HDRS_LCASE Odd capitalization of message header
 1.0 MANY_HDRS_LCASEOdd capitalization of multiple message headers


--- Begin Message ---
On Thu, Sep 19, 2013 at 10:07:47AM +0200, Stefan Nitz wrote:
> Hello list,
> 
> i want to to use 8 LTO-4 (400GiB) Tapes to backup 21 days.
> Best with one full-backup every 7 days.
> I've to backup about 1TiB fullbackup, 150 GiB each increment.
> 
> Could anyone gave some hints to setup amanda, pls. 
> I read all man-pages, i thought, but i dont understand how to do that.
> 

I'm all for consulting man pages, I do it regularly.

But for getting started, a good additional source of
information is the wiki at zmanda.com.

  http://wiki.zmanda.com

There are many "how-to" and "getting started" docs there.

After reading them you will probably still have some
questions.  We'll be glad to help, it is good software.

Jon
-- 
Jon H. LaBadie j...@jgcomp.com
 11226 South Shore Rd.  (703) 787-0688 (H)
 Reston, VA  20190  (609) 477-8330 (C)
--- End Message ---

setup 8 LTO-4 Tapes to backup 21 days or more

[Stefan Nitz]

Hello list,

i want to to use 8 LTO-4 (400GiB) Tapes to backup 21 days.
Best with one full-backup every 7 days.
I've to backup about 1TiB fullbackup, 150 GiB each increment.

Could anyone gave some hints to setup amanda, pls. 
I read all man-pages, i thought, but i dont understand how to do that.


With best regards from the sunny Lübeck 

Stefan Nitz

Advantic Systemhaus GmbH
Gutenbergstraße 15
23566 Lübeck
Telefon: 0451 60972-42
Fax: 0451 60972-20

Re: how to use runtapes >1 with single lto drive setup

[Jean-Louis Martineau]

Interactivity do not work with the default "traditional" taperscan, you 
must use "lexical" or "oldest".


define taperscan mytaperscan {
  plugin "lexical"
  #plugin "oldest"
}
taperscan "mytaperscan"

see: man amanda-taperscan

Jean-Louis

On 07/09/2013 07:28 AM, Karsten Fuhrmann wrote:

Hello,
i updated to amanda 3.3.4 now, but the problem still persists. I got no email 
from amanda and amstatus after an amflush run shows the same as before.
Any clues?

Here is a my amanda.conf:

# Global Settings

org "Cartoon-film"# your organization name for reports
displayunit "G"   # Display filesizes in megabytes
tpchanger "chg-single:tape:/dev/nsa0"
device-property "LEOM" "FALSE"
dumporder "BBB"   # specify the priority order of each dumper
taperalgo first
send-amreport-on all #always send an email report

dumpcycle 21# the number of days in the normal dump cycle
runspercycle 15 # the number of amdump runs in dumpcycle days
tapecycle 31 tapes  # the number of tapes in rotation

bumpsize 10 Gb  # minimum savings (threshold) to bump level 1 -> 2
bumpdays 1  # minimum days at each level
bumpmult 2  # threshold = bumpsize * bumpmult^(level-1)

etimeout 6000   # number of seconds per filesystem for estimates.
dtimeout 1800   # number of idle seconds before a dump is aborted.
ctimeout 300# maximum number of seconds that amcheck waits
  
runtapes 2		# number of tapes to be used in a single run of amdump

inparallel 4# maximum dumpers that will run in parallel (max 63)
maxdumps 1
maxdumpsize 2000 gb # Maximum number of bytes the planner will schedule
# for a run (default: runtapes * tape_length).

flush-threshold-dumped 100  # Start taping when 100% of tapesize  Capacity 
reside on HD
flush-threshold-scheduled 100

tapetype LTO4-ULTRIUM   # what kind of tape it is (see tapetypes below)
labelstr "^cf-daily-lto[0-9][0-9]*$"  # label constraint regex: all tapes must 
match

reserve 10 # This means save at least 30% of the holding disk space for 
degraded backups

autoflush yes # if autoflush is set to yes, then amdump will schedule all dump 
on holding disks to be flush to tape during the run.
columnspec 
"HostName=0:10,Disk=1:-1,Level=1:3,OrigKB=1:15,OutKB=1:15,DumpTime=1:8,TapeTime=1:8,DumpRate=1:8,TapeRate=1:8"

infofile "/var/amanda/cf-daily-lto/curinfo"   # database DIRECTORY
logdir   "/var/amanda/cf-daily-lto/logs"  # log directory
indexdir "/var/amanda/cf-daily-lto/index" # index directory


# Interactivity

define interactivity myinteractivity {
comment "interactivity email"
plugin "email"
property "mailto" "administrat...@cartoon-film.de"
property "resend_delay" "3600"
property "check-file" "/tmp/amanda_interactivity"
property "check-file-delay" "10"
}

interactivity "myinteractivity"

# Holding Disks

holdingdisk hd1 {
 comment "first holding disk"
 directory "/amanda"  # where the holding disk is
 use -5 Gb  # how much space can we use on it
# a non-positive value means:
#use all space but that value
 chunksize 50Gb # size of chunk if you want big dump to be
# dumped on multiple files on holding disks
#  N Kb/Mb/Gb split images in chunks of size N
# The maximum value should be
# (MAX_FILE_SIZE - 1Mb)
#  0  same as INT_MAX bytes
}

holdingdisk hd2 {
comment "second holding disk"
directory "/amanda2"
use -25Gb
chunksize 50Gb
}


# Tape types

define tapetype DLT8000 {
 comment "DLT8000"
 length 37 gbytes
 filemark 0 kbytes
 speed 1800 kps
}

define tapetype SDLT-320 {
comment "SDLT-320"
length 320 gbytes
filemark 0 kbytes
speed 36000 kbytes
}

define tapetype LTO4-ULTRIUM {
comment "LTO4"
length 850 gbytes #was 1200gbytes
filemark 0 kbytes
part_size 50 gbytes
}


# Scripts

define script-tool amzfs_snapshot {
comment "backup of zfs snapshot"
plugin  "amzfs-snapshot"
execute-on  pre-dle-amcheck, post-dle-amcheck, pre-dle-estimate, 
post-dle-estimate, pre-dle-backup, post-dle-backup
execute-where client
}


# Applications

Re: how to use runtapes >1 with single lto drive setup

[Karsten Fuhrmann]

Hello,
i updated to amanda 3.3.4 now, but the problem still persists. I got no email 
from amanda and amstatus after an amflush run shows the same as before.
Any clues?

Here is a my amanda.conf:

# Global Settings

org "Cartoon-film"  # your organization name for reports
displayunit "G" # Display filesizes in megabytes
tpchanger "chg-single:tape:/dev/nsa0"
device-property "LEOM" "FALSE"
dumporder "BBB" # specify the priority order of each dumper
taperalgo first
send-amreport-on all #always send an email report

dumpcycle 21# the number of days in the normal dump cycle
runspercycle 15 # the number of amdump runs in dumpcycle days
tapecycle 31 tapes  # the number of tapes in rotation

bumpsize 10 Gb  # minimum savings (threshold) to bump level 1 -> 2
bumpdays 1  # minimum days at each level
bumpmult 2  # threshold = bumpsize * bumpmult^(level-1)

etimeout 6000   # number of seconds per filesystem for estimates.
dtimeout 1800   # number of idle seconds before a dump is aborted.
ctimeout 300# maximum number of seconds that amcheck waits
 
runtapes 2  # number of tapes to be used in a single run of amdump
inparallel 4# maximum dumpers that will run in parallel (max 63)
maxdumps 1
maxdumpsize 2000 gb # Maximum number of bytes the planner will schedule
# for a run (default: runtapes * tape_length).

flush-threshold-dumped 100  # Start taping when 100% of tapesize  Capacity 
reside on HD
flush-threshold-scheduled 100 

tapetype LTO4-ULTRIUM   # what kind of tape it is (see tapetypes below)
labelstr "^cf-daily-lto[0-9][0-9]*$"# label constraint regex: all tapes 
must match

reserve 10 # This means save at least 30% of the holding disk space for 
degraded backups

autoflush yes # if autoflush is set to yes, then amdump will schedule all dump 
on holding disks to be flush to tape during the run.
columnspec 
"HostName=0:10,Disk=1:-1,Level=1:3,OrigKB=1:15,OutKB=1:15,DumpTime=1:8,TapeTime=1:8,DumpRate=1:8,TapeRate=1:8"

infofile "/var/amanda/cf-daily-lto/curinfo" # database DIRECTORY
logdir   "/var/amanda/cf-daily-lto/logs"# log directory
indexdir "/var/amanda/cf-daily-lto/index"   # index directory


# Interactivity

define interactivity myinteractivity {
comment "interactivity email"
plugin "email"
property "mailto" "administrat...@cartoon-film.de"
property "resend_delay" "3600"
property "check-file" "/tmp/amanda_interactivity"
property "check-file-delay" "10"
}

interactivity "myinteractivity" 

# Holding Disks

holdingdisk hd1 {
comment "first holding disk"
directory "/amanda" # where the holding disk is
use -5 Gb   # how much space can we use on it
# a non-positive value means:
#use all space but that value
chunksize 50Gb  # size of chunk if you want big dump to be
# dumped on multiple files on holding disks
#  N Kb/Mb/Gb split images in chunks of size N
# The maximum value should be
# (MAX_FILE_SIZE - 1Mb)
#  0  same as INT_MAX bytes
}

holdingdisk hd2 {
comment "second holding disk"
directory "/amanda2"
use -25Gb
chunksize 50Gb
}


# Tape types

define tapetype DLT8000 {
comment "DLT8000"
length 37 gbytes
filemark 0 kbytes
speed 1800 kps
}

define tapetype SDLT-320 {
comment "SDLT-320"
length 320 gbytes
filemark 0 kbytes
speed 36000 kbytes
}

define tapetype LTO4-ULTRIUM {
comment "LTO4"
length 850 gbytes #was 1200gbytes
filemark 0 kbytes
part_size 50 gbytes
}


# Scripts

define script-tool amzfs_snapshot {
   comment "backup of zfs snapshot"
   plugin  "amzfs-snapshot"
   execute-on  pre-dle-amcheck, post-dle-amcheck, pre-dle-estimate, 
post-dle-estimate, pre-dle-backup, post-dle-backup
   execute-where client
}


# Applications

define application-tool amzfs_sendrecv {
   comment "amzfs-sendrecv"
   property "KEEP-SNAPSHOT" "YES"
   plugin "amzfs-sendrecv"
}


# Dump Types

Re: how to use runtapes >1 with single lto drive setup

[Karsten Fuhrmann]

Hello,
i tried that, but didnt work. I ran an amflush and after the first tape was 
filled up the amflush finished without sending the email to change the tape. It 
did not even sent the normal email that it finished the amdump. It just quietly 
terminated amdump.
I am still using amanda 3.3.3 but with the patches you send my on 19.2.2013.

If i run an amstatus i still see the following:

Using /var/amanda/cf-daily-lto/logs/amflush.1
>From Thu Jul  4 14:54:12 UTC 2013

framestore.cartoon-film.de:/tank/koh/render0   286g process 
terminated while waiting to flush
framestore.cartoon-film.de:/tank/koh/user/./_rest_ 0 0g process 
terminated while waiting to flush
framestore.cartoon-film.de:/tank/koh/user/./k-o0   878g flushed 
(23:03:10)
framestore.cartoon-film.de:/tank/koh/user/./p-t054g flushed 
(23:36:42), PARTIAL
framestore.cartoon-film.de:/tank/koh/user/./u-z0 7g process 
terminated while waiting to flush
framestore.cartoon-film.de:/usr015g process 
terminated while waiting to flush
mail.cartoon-film.de:/ 0 0g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk1/pg_data0 0g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk1/sambawelt  033g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_buero  045g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_laura  015g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_laura2_buero   0 1g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_laura2_compo   011g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_laura2_keyscene0 1g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_laura2_presentation0 5g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/disk2/smb_zok2   0 5g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/extern1/smb_SaveOurNature097g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/extern1/smb_kein_ohr_hasen   019g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/extern1/smb_laura2_merchandising 021g process 
terminated while waiting to flush
mail.cartoon-film.de:/mnt/extern1/smb_laura_merchandise021g process 
terminated while waiting to flush

SUMMARY  part  real  estimated
   size   size
partition   :  20
estimated   :   00g
flush   :  20  1521g
failed  :   00g   (  0.00%)
wait for dumping:   00g   (  0.00%)
dumping to tape :   00g   (  0.00%)
dumping :   0 0g 0g (  0.00%) (  0.00%)
dumped  :   0 0g 0g (  0.00%) (  0.00%)
wait for writing:   0 0g 0g (  0.00%) (  0.00%)
wait to flush   :  18   588g   588g (100.00%) (  0.00%)
writing to tape :   0 0g 0g (  0.00%) (  0.00%)
failed to tape  :   0 0g 0g (  0.00%) (  0.00%)
taped   :   2   933g   933g (100.00%) ( 61.35%)
  tape 1:   2   933g   933g (109.84%) cf-daily-lto23 (20 chunks)
4 dumpers idle  : no-dumpers
taper status: Idle
taper qlen: 19
network free kps: 8
holding space   :  1168g (403.56%)
   taper busy   :  8:41:00  ( 99.46%)
 0 dumpers busy :  8:43:49  (100.00%)  no-dumpers:  8:43:41  ( 99.97%)
 not-idle:  0:00:08  (  0.03%)


The last lines of amflush.1 show this:
driver: result time 31429.525 from taper: REQUEST-NEW-TAPE 00-2
driver: send-cmd time 31429.525 to taper: START-SCAN worker0 00-2
driver: send-cmd time 31429.525 to taper: NEW-TAPE worker0 00-2
driver: state time 31429.527 free kps: 8 space: 1225029868 taper: writing 
idle-dumpers: 4 qlen tapeq: 18 runq: 0 roomq: 0 wakeup: 0 driver-idle: 
no-dumpers
driver: interface-state time 31429.527 if default: free 8
driver: hdisk-state time 31429.527 hdisk 0: free 1225029868 dumpers 0 hdisk 1: 
free 0 dumpers 0
driver: result time 31429.527 from taper: NO-NEW-TAPE 00-2
dump of driver schedule before start degraded mode:


dump of driver schedule after start degraded mode:


driver: state time 31429.528 free kps: 8 space: 1225029868 taper: DOWN 
idle-dumpers: 4 qlen tapeq: 18 runq: 0 roomq: 0 wakeup: 0 driver-idle: 
no-dumpers
driver: interface-state time 31429.528 if default: free 8
driver: hdisk-state time 31429.528 hdi

Re: how to use runtapes >1 with single lto drive setup

[Jean-Louis Martineau]

Karsten,

see: man amanda-interactivity

tpchanger "chg-single:tape:/dev/nst0"
define interactivity email {
  plugin "email"
  property "mailto" "m...@me.me"
}
interactivity email
runtapes 2


Jean-Louis


On 07/02/2013 11:04 AM, Karsten Fuhrmann wrote:

Hello,
i am trying to get this working for ages, but i seem to miss something 
all the time.
So i just want to use runtapes > 1 because some of my filesystems are 
bigger than my tapesize. I dont have a robot changer, just a lto 
drive, so i have to change tapes by hand. I would like to get an email 
when i need to change the tape. Can someone just post my the relevant 
lines of amanda.conf to do that?


Greetings,
Karsten Fuhrmann
System Administrator
Rothkirch Cartoon-Film GmbH
Hasenheide. 54
D-10967 Berlin
phone  +49 30 698084-109
fax  +49 30 698084-29
mobile +49 176 49118462
skype: parceval3000
AIM: in...@mac.com 
Jabber: parce...@jabber.org 

how to use runtapes >1 with single lto drive setup

[Karsten Fuhrmann]

Hello,
i am trying to get this working for ages, but i seem to miss something all the 
time.
So i just want to use runtapes > 1 because some of my filesystems are bigger 
than my tapesize. I dont have a robot changer, just a lto drive, so i have to 
change tapes by hand. I would like to get an email when i need to change the 
tape. Can someone just post my the relevant lines of amanda.conf to do that?

Greetings,
Karsten Fuhrmann
System Administrator
Rothkirch Cartoon-Film GmbH
Hasenheide. 54
D-10967 Berlin
phone  +49 30 698084-109
fax  +49 30 698084-29
mobile +49 176 49118462
skype: parceval3000
AIM: in...@mac.com
Jabber: parce...@jabber.org

15 min setup error

[Chris Arnold]

i just installed amanda backup server 3.3.1 on SLES11 and going through the 15 
min setup instructions. I get an error:
amandabackup@test:/space/vtapes/DailySet1/slots> for ((i=1; $i<=9;i++)); do 
amlabel DailySet1 DailySet1-0$i slot $i; done
If 'ammt' is not a typo you can run the following command to lookup the package 
that contains the binary:
command-not-found ammt  
 
-bash: ammt: command not found

I searched the computer for ammt and nothing was found. Amanda was installed 
via a rpm from the amanda site. How do i fix this?

new setup

[lxnf98mm]

I am about to set up a MSL6000
2 - LTO2 drives
1 - changer
2 - hba
Amanda 3.2.3
Before I reinvent the wheel, would anyone care to contribute a working 
amanda.conf?

Richard

Re: Need help with new architecture for NAS/NFS setup

[Leon Meßner]

On Thu, Nov 03, 2011 at 11:53:46AM -0400, Brendon Martino wrote:
> Thus, it was proposed that I mount some NFS space from another system (a
> NAS, for all intents and purposes, though it has a WORM architecture and
> is only used for archiving data with retention periods that delete old
> archives). So now I have 1TB of space mounted under /nfsbackup. How do I
> implement my architecture to only keep about a week (or even a day) of
> backups in the holding disk (locally on the system) but use the nfs
> storage space for archiving the rest of the old backups? The idea is
> that we keep 30 to 60 days worth of old backups on the NAS, but only the
> last day or few days locally on the backup server.
> 
> How do I do that? Is it possible? What would be the general idea/layout?
> What directives would I need to change? Would I need to use multiple
> DailySets?? I'm totally stumped. Any advice would be greatly appreciated.

Did you actually try yo setup a holdingdisk with a capacity of
7+something days of vtapes and putting your tapes on NFS ? You would
need to configure your holdingspace to flush after said time of course.
Or can't a holdingdisk have more than one tape worth of data?

cheers,
Leon

Re: Need help with new architecture for NAS/NFS setup

[John Hein]

Brendon Martino wrote at 11:53 -0400 on Nov  3, 2011:
 > I'm running Amanda version 2.6.0p2-14 on Fedora 10. My current
 > architecture for Amanda is as follows:
.
.
 >  How do I
 > implement my architecture to only keep about a week (or even a day) of
 > backups in the holding disk (locally on the system) but use the nfs
 > storage space for archiving the rest of the old backups? The idea is
 > that we keep 30 to 60 days worth of old backups on the NAS, but only the
 > last day or few days locally on the backup server.
 > 
 > How do I do that? Is it possible? What would be the general idea/layout?
 > What directives would I need to change? Would I need to use multiple
 > DailySets?? I'm totally stumped. Any advice would be greatly appreciated.

In addition to other suggestions on this thread, there's also amvault.
But not with amanda-2.6.

re: Need help with new architecture for NAS/NFS setup

[Olivier]

Hi Brendon,

> I'm having trouble figuring out what type of new architecture to go with
> for an NFS share dir on a NAS.

You could run two dumps in parallel, one with tapecycle of 7 onj your local 
disk and one with tape cycle of 30 or 60 on your NAS.

Best regards,

Olivier

> I'm running Amanda version 2.6.0p2-14 on Fedora 10. My current
> architecture for Amanda is as follows:
> 
> I'm running a set called DailySet1. My holding disk/vtapes are:
> /amanda/day0x/data/  where x is 1-8. So I have 8 vtapes in my holding
> disk. Here's my changer.conf:
> multieject 0
> gravity 0
> needeject 0
> ejectdelay 0
> statefile /var/lib/amanda/DailySet1/changer-status
> firstslot 1
> lastslot 8
> slot 1file:/amanda/day01
> slot 2file:/amanda/day02
> slot 3file:/amanda/day03
> slot 4file:/amanda/day04
> slot 5file:/amanda/day05
> slot 6file:/amanda/day06
> slot 7file:/amanda/day07
> slot 8file:/amanda/day08
> 
> If I do a dump each night, I get a backup for each client in my disklist
> per day. That gives me about a week's worth of backups. Here's some info
> from the amanda.conf file:
> 
> dumpcycle 7 days# the number of days in the normal dump cycle
> runspercycle 0  # the number of amdump runs in dumpcycle days
> tapecycle 8 tapes   # the number of tapes in rotation
> runtapes 1  # number of tapes to be used in a single run of
> amdump
> tpchanger "chg-multi"   # the tape-changer glue script
> changerfile "/etc/amanda/DailySet1/changer.conf"
> 
> 
> This works fine, however, it only does a week of backups. If I wanted to
> do a month of backups, I would have to multiply my current system X 4.
> So I would need a tapecycle of 30 (or so). My dumpcycle would remain 7
> since I still want to do full dumps weekly. But in this case, I would
> need to expand my array of vtapes from /amanda/day01/ to /amanda/day30/.
> This is not possible since my holding disk area (the size of the
> system's usable hard disk space) is not big enough for all of those
> vtapes (and the data they would end up having in them) as it only has
> about 100 GB free and my backups are about 50GB per week.
> 
> Thus, it was proposed that I mount some NFS space from another system (a
> NAS, for all intents and purposes, though it has a WORM architecture and
> is only used for archiving data with retention periods that delete old
> archives). So now I have 1TB of space mounted under /nfsbackup. How do I
> implement my architecture to only keep about a week (or even a day) of
> backups in the holding disk (locally on the system) but use the nfs
> storage space for archiving the rest of the old backups? The idea is
> that we keep 30 to 60 days worth of old backups on the NAS, but only the
> last day or few days locally on the backup server.
> 
> How do I do that? Is it possible? What would be the general idea/layout?
> What directives would I need to change? Would I need to use multiple
> DailySets?? I'm totally stumped. Any advice would be greatly 
appreciated.
> 
> Thanks
> -Brendon Martino

Re: Need help with new architecture for NAS/NFS setup

[Charles Curley]

On Thu, 3 Nov 2011 11:53:46 -0400
Brendon Martino  wrote:

> I'm having trouble figuring out what type of new architecture to go
> with for an NFS share dir on a NAS.

I'm inclined to think you can't, short of a messy script that plays
with symlinks and rsync. I wouldn't touch that with a ten meter pole.

I do a backup of similar size to a USB external hard drive, which I then
rsync to another external hard drive. The second I swap out weekly and
is my off-site backup. See:

http://www.charlescurley.com/blog/articles/off_site_backups_for_amanda/index.html
http://www.charlescurley.com/blog/articles/backing_up_amanda_metadata/index.html




-- 

Charles Curley  /"\ASCII Ribbon Campaign
Looking for fine software   \ /Respect for open standards
and/or writing?  X No HTML/RTF in email
http://www.charlescurley.com/ \No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

Need help with new architecture for NAS/NFS setup

[Brendon Martino]

I'm having trouble figuring out what type of new architecture to go with
for an NFS share dir on a NAS.

I'm running Amanda version 2.6.0p2-14 on Fedora 10. My current
architecture for Amanda is as follows:

I'm running a set called DailySet1. My holding disk/vtapes are:
/amanda/day0x/data/  where x is 1-8. So I have 8 vtapes in my holding
disk. Here's my changer.conf:
multieject 0
gravity 0
needeject 0
ejectdelay 0
statefile /var/lib/amanda/DailySet1/changer-status
firstslot 1
lastslot 8
slot 1file:/amanda/day01
slot 2file:/amanda/day02
slot 3file:/amanda/day03
slot 4file:/amanda/day04
slot 5file:/amanda/day05
slot 6file:/amanda/day06
slot 7file:/amanda/day07
slot 8file:/amanda/day08

If I do a dump each night, I get a backup for each client in my disklist
per day. That gives me about a week's worth of backups. Here's some info
from the amanda.conf file:

dumpcycle 7 days# the number of days in the normal dump cycle
runspercycle 0  # the number of amdump runs in dumpcycle days
tapecycle 8 tapes   # the number of tapes in rotation
runtapes 1  # number of tapes to be used in a single run of
amdump
tpchanger "chg-multi"   # the tape-changer glue script
changerfile "/etc/amanda/DailySet1/changer.conf"


This works fine, however, it only does a week of backups. If I wanted to
do a month of backups, I would have to multiply my current system X 4.
So I would need a tapecycle of 30 (or so). My dumpcycle would remain 7
since I still want to do full dumps weekly. But in this case, I would
need to expand my array of vtapes from /amanda/day01/ to /amanda/day30/.
This is not possible since my holding disk area (the size of the
system's usable hard disk space) is not big enough for all of those
vtapes (and the data they would end up having in them) as it only has
about 100 GB free and my backups are about 50GB per week.

Thus, it was proposed that I mount some NFS space from another system (a
NAS, for all intents and purposes, though it has a WORM architecture and
is only used for archiving data with retention periods that delete old
archives). So now I have 1TB of space mounted under /nfsbackup. How do I
implement my architecture to only keep about a week (or even a day) of
backups in the holding disk (locally on the system) but use the nfs
storage space for archiving the rest of the old backups? The idea is
that we keep 30 to 60 days worth of old backups on the NAS, but only the
last day or few days locally on the backup server.

How do I do that? Is it possible? What would be the general idea/layout?
What directives would I need to change? Would I need to use multiple
DailySets?? I'm totally stumped. Any advice would be greatly appreciated.

Thanks
-Brendon Martino

Re: Bad performance with split tape setup

[Dominik Schips]

Am Donnerstag, den 22.07.2010, 09:30 -0400 schrieb Dustin J. Mitchell:
> On Thu, Jul 22, 2010 at 4:04 AM, Dominik Schips  wrote:
> > With normal configuration the 500 GB full backup to tape took about
> > 10-11 hours.
> >
> > With the split tape config it took about 16 hours to backup 116 GB and
> > amanda ist still running to backup the other data.
> 
> When doing a split write without holding disk, Amanda must cache all
> of the data it writes.  That additional write can be time-consuming!
> Also, when not using holding disk, your write speed is constrained by
> the speed of the process actually dumping your filesystem (probably
> gnutar?), which for filesystems with a lot of small files can be very
> slow.

Yes it is gnutar with lots of small data (mailserver data).


> > Any advice or solution how to debug the problem (correct) and get normal
> > write performance with split tape dumps config?
> 
> * Add holding disk
> * Use a newer version of Amanda
> * Split in memory or use a faster disk for your split_diskbuffer

Thank you for the fast help. I'll try it with a holding disk.

-- 

Best regards

Dominik Schips

Re: Bad performance with split tape setup

[Dustin J. Mitchell]

On Thu, Jul 22, 2010 at 4:04 AM, Dominik Schips  wrote:
> With normal configuration the 500 GB full backup to tape took about
> 10-11 hours.
>
> With the split tape config it took about 16 hours to backup 116 GB and
> amanda ist still running to backup the other data.

When doing a split write without holding disk, Amanda must cache all
of the data it writes.  That additional write can be time-consuming!
Also, when not using holding disk, your write speed is constrained by
the speed of the process actually dumping your filesystem (probably
gnutar?), which for filesystems with a lot of small files can be very
slow.

> Any advice or solution how to debug the problem (correct) and get normal
> write performance with split tape dumps config?

* Add holding disk
* Use a newer version of Amanda
* Split in memory or use a faster disk for your split_diskbuffer

Dustin

-- 
Open Source Storage Engineer
http://www.zmanda.com

Bad performance with split tape setup

[Dominik Schips]

Hello,

I changed my setup like described in the wiki
(http://wiki.zmanda.com/index.php/How_To:Split_Dumps_Across_Tapes) to
split dumps accross tapes.

My problem is that I only get very low write performance with this
configuration now.

I am running amanda 2.5.2p1 on a Debian 5.0.5 (Lenny/i686) without a
holdingdisk and a LTO2 tape device with a 16 tape changer.

>From a log file before and after the change.

Normal backup run:


SUCCESS dumper backup.example.net  20100720135703 0 [sec
140.366 kb 966790 kps 6887.6 orig-kb 966790]
SUCCESS taper backup.example.net  20100720135703 0 [sec
140.684 kb 966816 kps 6872.2 {wr: writers 30214 rdwait 114.055 wrwait
24.330 filemark 2.053}]
SUCCESS dumper backup.example.net  20100720135703 0 [sec
108.058 kb 889870 kps 8235.1 orig-kb 889870]
SUCCESS taper backup.example.net  20100720135703 0 [sec
108.305 kb 889888 kps 8216.4 {wr: writers 27810 rdwait 84.511 wrwait
20.996 filemark 2.440}]
SUCCESS dumper backup.example.net  20100720135703 0 [sec
124.763 kb 763210 kps 6117.2 orig-kb 763210]
SUCCESS taper backup.example.net  20100720135703 0 [sec
125.878 kb 763232 kps 6063.2 {wr: writers 23852 rdwait 110.863 wrwait
10.540 filemark 4.280}]
SUCCESS dumper backup.example.net  20100720135703 0 [sec
141.383 kb 660400 kps 4671.0 orig-kb 660400]
SUCCESS taper backup.example.net  20100720135703 0 [sec
141.590 kb 660416 kps 4664.3 {wr: writers 20639 rdwait 129.368 wrwait
10.180 filemark 1.923}]


Split tape run:


CHUNK taper backup.example.net  20100721132703 2023 0 [sec
5.061 kb 10208 kps 2016.8 {wr: writers 320 rdwait 0.318 wrwait 0.149
filemark 4.591}]
CHUNK taper backup.example.net   20100721132703 2024 0 [sec
4.829 kb 10208 kps 2113.5 {wr: writers 320 rdwait 0.053 wrwait 0.151
filemark 4.622}]
CHUNK taper backup.example.net   20100721132703 2025 0 [sec
5.191 kb 10208 kps 1966.2 {wr: writers 320 rdwait 0.095 wrwait 0.152
filemark 4.941}]
CHUNK taper backup.example.net   20100721132703 2026 0 [sec
5.266 kb 10208 kps 1938.3 {wr: writers 320 rdwait 0.063 wrwait 0.153
filemark 5.047}]
CHUNK taper backup.example.net   20100721132703 2027 0 [sec
5.248 kb 10208 kps 1945.0 {wr: writers 320 rdwait 0.059 wrwait 0.151
filemark 5.034}]
CHUNK taper backup.example.net   20100721132703 2028 0 [sec
5.317 kb 10208 kps 1919.8 {wr: writers 320 rdwait 0.089 wrwait 0.152
filemark 5.072}]
CHUNK taper backup.example.net   20100721132703 2029 0 [sec
5.306 kb 10208 kps 1923.6 {wr: writers 320 rdwait 0.105 wrwait 0.154
filemark 5.043}]
CHUNK taper backup.example.net  20100721132703 2030 0 [sec
5.243 kb 10208 kps 1946.7 {wr: writers 320 rdwait 0.031 wrwait 0.155
filemark 5.053}]


With normal configuration the 500 GB full backup to tape took about
10-11 hours.

With the split tape config it took about 16 hours to backup 116 GB and
amanda ist still running to backup the other data.

Any advice or solution how to debug the problem (correct) and get normal
write performance with split tape dumps config?

Please let me know which configs can help to debug the problem.


-- 

Best regards

Dominik Schips

Re: holdingdisk-setup for Ultrium/LTO-2

[Stefan G. Weichinger]

Greg Troxel schrieb:

> Amanda reports a wide range of tape write speeds during actual backups
> and flushing, and I can't make sense of them - some of them seem too
> high by a factor of 8.

You mean the output of amstatus? No, there are no tape rates afaik, so
you mean the report email?

> Doing dd of a 8.5G file (2G ram, dumped several days ago, so could not
> possibly be in the cache) to /dev/null gets me 42.3 MB/s.  The raw drive
> gets me 67.8 MB/s.
> 
> This was all without another process writing dumps.
> 
> So for LTO-2, regular drives seem just barely ok.  For LTO-4 I wouldn't
> be so sure, as I think that needs 80 MB/s minimum.

My main problem seems to be the concurrent writes of the dumpers versus
the read-processes of the taper ... :-(

Right now I have a RAID0 over two LVM-LVs which still doesn't perform
good enough. I will repartition the drives to get two plain
Linux-raid-partitions without that additional LVM-layer I have right now
(it was simpler to try it this way first ...). Maybe that helps a bit.

An additional fact is that these two disks also contain DLEs to be
dumped so with amdump I get some traffic between LVs ... which also
decreases performance

Re: holdingdisk-setup for Ultrium/LTO-2

[Greg Troxel]

I am using an LTO2 drive with a holding disk which is a BSD UFS2 (==
FFSv2) filesystem on a normal Seagata ATA drive:

wd1 at atabus1 drive 1: 
wd1: quirks 2
wd1: drive supports 16-sector PIO transfers, LBA48 addressing
wd1: 372 GB, 775221 cyl, 16 head, 63 sec, 512 bytes/sect x 781422768 sectors
wd1: 32-bit data port
wd1: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 5 (Ultra/100)
wd1(piixide0:1:1): using PIO mode 4, Ultra-DMA mode 5 (Ultra/100) (using DMA)

Amanda reports a wide range of tape write speeds during actual backups
and flushing, and I can't make sense of them - some of them seem too
high by a factor of 8.

Doing dd of a 8.5G file (2G ram, dumped several days ago, so could not
possibly be in the cache) to /dev/null gets me 42.3 MB/s.  The raw drive
gets me 67.8 MB/s.

This was all without another process writing dumps.

So for LTO-2, regular drives seem just barely ok.  For LTO-4 I wouldn't
be so sure, as I think that needs 80 MB/s minimum.



pgpg1nFJVyzOQ.pgp
Description: PGP signature

Re: holdingdisk-setup for Ultrium/LTO-2

[Stefan G. Weichinger]

Chris Hoogendyk schrieb:

> raid0 for holding disk should be just fine.
> 
> any chance of going SAS or scsi?

hmm, I'd prefer to stay with the SATAs I already have here.

Currently I testdrive a RAID0 of two partitions on the two SATAs ... I
will also check for a third SATA-port on the motherboard to maybe add
one of my other SATAs (still have some 250GB and 500GB drives around).

It's more of workstation-hardware than server-hardware ... so it's a bit
limited.

> does the adaptec have another inteface?

Sure, yes, but the capacity of SCSI-drives won't be sufficient here or
at least the drives would be expensive.

Otoh HP recommends not to run tapedrive and harddisk on one
SCSI-controller (in a guide for optimizing performance with
Ultrium-1-drives ...) so I'd like to stay away from any more scsi-hdd.

> the minimum & maximum writing speeds can be drive dependent, but for
> LTO2 it should be 40 max and it might be 20 minimum. I would guess if
> you can keep it above 20 you won't get shoe shining. You're very close
> to that, but if you hit it, the drop in speed would be dramatic.

Afaik ultrium2 adjusts down to around 9 MB/sec or something like that,
so I assume I didn't get any shoeshining so far.

Thanks for your feedback,
Stefan

Re: holdingdisk-setup for Ultrium/LTO-2

[Stefan G. Weichinger]

Sven Rudolph schrieb:

> LTO drives usually should be able to adapt the write speed (data rate)
> depending on the incoming data rate.

Yes. I was only a bit surprised by the fact that it was adjusting that
quickly ... maybe it sounds more tape-stressing than it is.

> In order to feed LTO4 drives at full speed I have severeal holding
> disks, each one is a raid0 of three SAS drives. I use ext3, but I guess
> other file system types would work as well.

I agree.

Thanks, Stefan

Re: holdingdisk-setup for Ultrium/LTO-2

[Chris Hoogendyk]

Stefan G. Weichinger wrote:

Greets, amanda-users,

as I recently got myself an Ultrium 2 drive (at last ...) I wonder how
to optimize performance ...

Right now I only have the amanda holdingdisk on one physical SATA-drive
which seems to work out OK as long as there is no other I/O on that disk ...

The LTO is attached to a good old Adaptec 29160 controller (the only
scsi-device), the server is built around some Intel Prescott CPU (yeah,
kinda old ...) and runs gentoo linux (32 bit).

Right now there are 2 SATA-drives in there, both Seagate, one
ST3500320AS, one ST31000533CS.

The holdingdisk is an LVM2-LV formatted with xfs, the underlying
physical volume is a partition of the ST31000533CS.

I see speeds up to 22MB/s with amanda, but also consider the sound of
the tape-drive accelerating and decelerating could or should be avoided
... afaik it doesn't shoeshine but it doesn't always stream at constant
speed when other apps access the drives.

-->

What I would like to brainstorm:

I think about setting up a software-based RAID0 for the holdingdisk. I
have another ST31000533CS to substitute the smaller ST3500320AS, this
would give me the opportunity to do some shuffling and get a
holdingdisk-raid0 with 100GB or 200GB of size.

I assume the filesystem used for the holdingdisk won't matter as much as
the speed gained by doing the RAID0, so I would maybe just stay with XFS
for simplicity (good with big files ... no headaches ...)

Yes, I know, RAID0 means no redundancy but I would take the risk here ...

Additional q: Does anyone of you use some specific stinit.def for

Host: scsi0 Channel: 00 Id: 03 Lun: 00
  Vendor: HP   Model: Ultrium 2-SCSI   Rev: S63D
  Type:   Sequential-AccessANSI  SCSI revision: 03


raid0 for holding disk should be just fine.

any chance of going SAS or scsi?

does the adaptec have another inteface?

the minimum & maximum writing speeds can be drive dependent, but for 
LTO2 it should be 40 max and it might be 20 minimum. I would guess if 
you can keep it above 20 you won't get shoe shining. You're very close 
to that, but if you hit it, the drop in speed would be dramatic.



--
---

Chris Hoogendyk

-
  O__   Systems Administrator
 c/ /'_ --- Biology & Geology Departments
(*) \(*) -- 140 Morrill Science Center
~~ - University of Massachusetts, Amherst 




--- 


Erdös 4

Re: holdingdisk-setup for Ultrium/LTO-2

[Sven Rudolph]

"Stefan G. Weichinger"  writes:

> as I recently got myself an Ultrium 2 drive (at last ...) I wonder how
> to optimize performance ...
>
> Right now I only have the amanda holdingdisk on one physical SATA-drive
> which seems to work out OK as long as there is no other I/O on that disk ...

> I see speeds up to 22MB/s with amanda, but also consider the sound of
> the tape-drive accelerating and decelerating could or should be avoided
> ... afaik it doesn't shoeshine but it doesn't always stream at constant
> speed when other apps access the drives.

LTO drives usually should be able to adapt the write speed (data rate)
depending on the incoming data rate.

In order to feed LTO4 drives at full speed I have severeal holding
disks, each one is a raid0 of three SAS drives. I use ext3, but I guess
other file system types would work as well.

Sven

holdingdisk-setup for Ultrium/LTO-2

[Stefan G. Weichinger]

Greets, amanda-users,

as I recently got myself an Ultrium 2 drive (at last ...) I wonder how
to optimize performance ...

Right now I only have the amanda holdingdisk on one physical SATA-drive
which seems to work out OK as long as there is no other I/O on that disk ...

The LTO is attached to a good old Adaptec 29160 controller (the only
scsi-device), the server is built around some Intel Prescott CPU (yeah,
kinda old ...) and runs gentoo linux (32 bit).

Right now there are 2 SATA-drives in there, both Seagate, one
ST3500320AS, one ST31000533CS.

The holdingdisk is an LVM2-LV formatted with xfs, the underlying
physical volume is a partition of the ST31000533CS.

I see speeds up to 22MB/s with amanda, but also consider the sound of
the tape-drive accelerating and decelerating could or should be avoided
... afaik it doesn't shoeshine but it doesn't always stream at constant
speed when other apps access the drives.

-->

What I would like to brainstorm:

I think about setting up a software-based RAID0 for the holdingdisk. I
have another ST31000533CS to substitute the smaller ST3500320AS, this
would give me the opportunity to do some shuffling and get a
holdingdisk-raid0 with 100GB or 200GB of size.

I assume the filesystem used for the holdingdisk won't matter as much as
the speed gained by doing the RAID0, so I would maybe just stay with XFS
for simplicity (good with big files ... no headaches ...)

Yes, I know, RAID0 means no redundancy but I would take the risk here ...

Additional q: Does anyone of you use some specific stinit.def for

Host: scsi0 Channel: 00 Id: 03 Lun: 00
  Vendor: HP   Model: Ultrium 2-SCSI   Rev: S63D
  Type:   Sequential-AccessANSI  SCSI revision: 03

?

Thanks for any feedback,
greetings,
Stefan

Re: Best Amanda Setup for Restores

[Jean-Louis Martineau]

The amidxtaped.20080224080312.debug file show a problem.

amidxtaped: 1: restoring split dumpfile: date 20080114 host 
picard.gman.rox disk / part 2048/UNKNOWN lev 0 comp .gz program /bin/tar

amidxtaped:  Chunk out of order, will save to disk and append to output.

It says it create a temporary files.

amrecover should list all PART to amidxtaped, but it doesn't list the 
first 2047 part.
On which tape are the first 2047 part of that image? Have you already 
overwritten that tape with new a image?


What's the output of: amadmin  find picard.gman.rox /
It should tell on which tape the first part are.

Jean-Louis

Gordon J. Mills III wrote:

Jean-Louis,

Sorry for the late reply. I have been busy with some other things. I did
manage to set a different restore date and get some of the data back.
Apparently there was a problem with the full backup that I was trying to
retrieve from. I have included the relevant amidxtaped.*.debug files and the
amrecover.*.debug file as well from the recent attempt. Like I said, it was
partially successful. 


Also, I had to recover some other files just now and it worked seemingly
different from what I had experienced before. It did not need to restore the
entire DLE to /tmp/amanda (which is what everyone was saying it should not
do). The amrestore session went fairly quickly and the files were restored. 


This leads me to believe that there is something wrong with the full backup
of that dle for the date that I was trying to restore before.

Like I said, I have included the files in case you want to look at them. I
can understand if you don't since this appears to be a case of a bad backup
(from my humble guess).

Thanks again for all of your time and trouble. 


Regards,
Gordon

-Original Message-
From: Jean-Louis Martineau [mailto:[EMAIL PROTECTED] 
Sent: Friday, February 22, 2008 4:15 PM

To: [EMAIL PROTECTED]
Subject: Re: Best Amanda Setup for Restores

Gordon J. Mills III wrote:
  

I guess I am doing something wrong here. I would like to find out what it
is.

In /tmp/amanda, amrecover creates a file named
server.domain.name._.20080114.0.0002048. That file is 55G when I delete


it.
  

I have to delete it because at that point the disk is full. This is a


backup
  

of the / folder and it is compressed. The folder that is selected for
extraction is 4.3G.
  

Are you sure server.domain.name._.20080114.0.0002048 is not a file in 
the backup and you asked to restore it?
  

I have used the lcd command to select the folder where it want the restore
to be put.

The client and the server are the same machine in this scenario.

The amrecover.*.debug file is 2.5M. I will send it to you if that will


help.
  
  


It will help.
  

The amrecover session all goes fine. I run it. Select the disk, date,


folder
  

that I want, etc. run the extract. Where can I get the complete output of
the session?
  



cut&paste of the screen, I need to know exactly want you did.

Jean-Louis

Re: Best Amanda Setup for Restores

[Jeffrey D Anderson]

On Friday 22 February 2008, Gordon J. Mills III wrote:

> I am beginning to question my sanity here . I am using the amrecover
> interface to recover it but it seems that it will first extract the entire
> tar'd up dle to /tmp/Amanda before extracting the small folder that I
> requested. Am I wrong about that?
> If this is the case (and it seems to be what I have experienced) then my
> problem is that the entire tar'd up dle will not fit on the partition which
> contains /tmp/Amanda/. So to get at this particular folder for this restore
> I need to change that default to another drive which does have enough free
> space.
>

I'll let others answer the question of whether there is a way of changing this 
default, but I can recommend a simple solution -- make /tmp/amanda a link to 
a directory on some other disk that has enough space.

-- 
--
Jeffrey Anderson| [EMAIL PROTECTED]
Lawrence Berkeley National Laboratory   | 
Office: 50A-5104E   | Mailstop 50A-5101
Phone: 510 486-4208 | Fax: 510 486-6808

RE: Best Amanda Setup for Restores

[Gordon J. Mills III]

I guess I am doing something wrong here. I would like to find out what it
is.

In /tmp/amanda, amrecover creates a file named
server.domain.name._.20080114.0.0002048. That file is 55G when I delete it.
I have to delete it because at that point the disk is full. This is a backup
of the / folder and it is compressed. The folder that is selected for
extraction is 4.3G.

I have used the lcd command to select the folder where it want the restore
to be put.

The client and the server are the same machine in this scenario.

The amrecover.*.debug file is 2.5M. I will send it to you if that will help.

The amrecover session all goes fine. I run it. Select the disk, date, folder
that I want, etc. run the extract. Where can I get the complete output of
the session?

Thanks,
-Gordon

-Original Message-
From: Jean-Louis Martineau [mailto:[EMAIL PROTECTED] 
Sent: Friday, February 22, 2008 3:32 PM
To: [EMAIL PROTECTED]
Cc: amanda-users@amanda.org
Subject: Re: Best Amanda Setup for Restores

amrecover only write to disk the files you asked for extraction.
amrecover extract in the current directory when we executed amrecover 
(not on /tmp/Amanda), you can use the lcd command to change the cwd.

Can you post the complete output of the amrecover session and the 
complete amrecover.*.debug file.

Jean-Louis

Gordon J. Mills III wrote:
>> Is this simply a case of not realizing that 'amrestore' is generally used
>> for complete dle restoration while 'amrecover' is used for individual
>> file/directory recovery?
>> 
>
> I am beginning to question my sanity here . I am using the amrecover
> interface to recover it but it seems that it will first extract the entire
> tar'd up dle to /tmp/Amanda before extracting the small folder that I
> requested. Am I wrong about that? 
> If this is the case (and it seems to be what I have experienced) then my
> problem is that the entire tar'd up dle will not fit on the partition
which
> contains /tmp/Amanda/. So to get at this particular folder for this
restore
> I need to change that default to another drive which does have enough free
> space.
>
> To further clarify, I am running Debian and using the packages in the
debian
> archive for Amanda. I am using vtapes also.
>
> Thanks for the help,
> Gordon
>
>
>   

RE: Best Amanda Setup for Restores

[Gordon J. Mills III]

>Let me assure, AFAIK Amanda does not do that.
>How did you come to that conclusion?
>What is the name of the file that is created in "/tmp/Amanda" (uppercase 
>A in the name???)

The A is lowercase. My mail client put the uppercase. The file is
server.domain.name._.20080114.0.0002048. It is 55G when I am forced to
delete it because the disk is full. 

Scratching head now.
-Gordon

> If this is the case (and it seems to be what I have experienced) then my
> problem is that the entire tar'd up dle will not fit on the partition
which
> contains /tmp/Amanda/. So to get at this particular folder for this
restore
> I need to change that default to another drive which does have enough free
> space.
> 
> To further clarify, I am running Debian and using the packages in the
debian
> archive for Amanda. I am using vtapes also.



-- 
Paul Bijnens, xplanation Technology ServicesTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, ^^, *
* F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... *
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: Best Amanda Setup for Restores

[Jean-Louis Martineau]

amrecover only write to disk the files you asked for extraction.
amrecover extract in the current directory when we executed amrecover 
(not on /tmp/Amanda), you can use the lcd command to change the cwd.


Can you post the complete output of the amrecover session and the 
complete amrecover.*.debug file.


Jean-Louis

Gordon J. Mills III wrote:

Is this simply a case of not realizing that 'amrestore' is generally used
for complete dle restoration while 'amrecover' is used for individual
file/directory recovery?



I am beginning to question my sanity here . I am using the amrecover
interface to recover it but it seems that it will first extract the entire
tar'd up dle to /tmp/Amanda before extracting the small folder that I
requested. Am I wrong about that? 
If this is the case (and it seems to be what I have experienced) then my

problem is that the entire tar'd up dle will not fit on the partition which
contains /tmp/Amanda/. So to get at this particular folder for this restore
I need to change that default to another drive which does have enough free
space.

To further clarify, I am running Debian and using the packages in the debian
archive for Amanda. I am using vtapes also.

Thanks for the help,
Gordon


  

Re: Best Amanda Setup for Restores

[Brian Cuttler]

On Fri, Feb 22, 2008 at 01:17:26PM -0500, Jon LaBadie wrote:
> On Fri, Feb 22, 2008 at 11:48:02AM -0600, Gordon J. Mills III wrote:
> > I have recently run into a problem with restoring. I want to restore a
> > folder on a DLE that is about 100G. The folder I want to restore is small
> > but apparently Amanda has to restore the entire DLE to get at the 1 folder.
> > It is restoring the large file to /tmp/Amanda/. Unfortunately, I do not have
> > enough room there to extract the entire DLE.
> > 
> > My first question is: How can I have Amanda use a different path for this
> > task. I have an NFS mounted external drive that has much more room on it.
> > 
> > 2nd question: Is there a way to setup Amanda so that it does not have to
> > retrieve the entire DLE file? Is it just removing compression from the DLE
> > def? Or chopping up the DLE to smaller chunks (this will be hard since most
> > of the data is in 1 place on that dle).
> > 
> > I would like for Amanda to behave in a similar way that I have had other
> > backup software behave. That is, if I just want a single file or folder,
> > then it scans the tape (in this case vtapes) and retrieves only what I want.
> > When I have successfully retrieved files with Amanda before it took a very
> > long time. Is there a way to set it up so that it will do this?
> > 
> 
> Is this simply a case of not realizing that 'amrestore' is generally used
> for complete dle restoration while 'amrecover' is used for individual
> file/directory recovery?

We never really configured amrecover and use amrestore for everything.

Typically the command sequence looks like this

# cd /amanda/restore (on the amanda work partition)
# amrestore -p /dev/rmt/0n host /partition | restore -if - .

-p for pipe the amrestore 
-i for interactive, I can walk down the directory structure.

Similar method exists for tar "dump" type.

My milage varies, yours will too. Depending on the OS, dumptype
(ufsrestore and xfsrestore take somewhat different syntax and I'm
certain linux restores, or MAC OS X, are in the near future since
we do have amanda clients of those types ), and there are some
cross platform restores that are trickier, Solaris amanda host
and IRIX amanda client, but its all doable without having to bring
back complete and large DLEs.







> -- 
> Jon H. LaBadie  [EMAIL PROTECTED]
>  JG Computing
>  12027 Creekbend Drive(703) 787-0884
>  Reston, VA  20194(703) 787-0922 (fax)
---
   Brian R Cuttler [EMAIL PROTECTED]
   Computer Systems Support(v) 518 486-1697
   Wadsworth Center(f) 518 473-6384
   NYS Department of HealthHelp Desk 518 473-0773



IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.

Re: Best Amanda Setup for Restores

[Paul Bijnens]

Gordon J. Mills III wrote:

Is this simply a case of not realizing that 'amrestore' is generally used
for complete dle restoration while 'amrecover' is used for individual
file/directory recovery?


I am beginning to question my sanity here . I am using the amrecover
interface to recover it but it seems that it will first extract the entire
tar'd up dle to /tmp/Amanda before extracting the small folder that I
requested. Am I wrong about that? 


Let me assure, AFAIK Amanda does not do that.
How did you come to that conclusion?
What is the name of the file that is created in "/tmp/Amanda" (uppercase 
A in the name???)





If this is the case (and it seems to be what I have experienced) then my
problem is that the entire tar'd up dle will not fit on the partition which
contains /tmp/Amanda/. So to get at this particular folder for this restore
I need to change that default to another drive which does have enough free
space.

To further clarify, I am running Debian and using the packages in the debian
archive for Amanda. I am using vtapes also.




--
Paul Bijnens, xplanation Technology ServicesTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, ^^, *
* F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... *
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

RE: Best Amanda Setup for Restores

[Gordon J. Mills III]

>Is this simply a case of not realizing that 'amrestore' is generally used
>for complete dle restoration while 'amrecover' is used for individual
>file/directory recovery?

I am beginning to question my sanity here . I am using the amrecover
interface to recover it but it seems that it will first extract the entire
tar'd up dle to /tmp/Amanda before extracting the small folder that I
requested. Am I wrong about that? 
If this is the case (and it seems to be what I have experienced) then my
problem is that the entire tar'd up dle will not fit on the partition which
contains /tmp/Amanda/. So to get at this particular folder for this restore
I need to change that default to another drive which does have enough free
space.

To further clarify, I am running Debian and using the packages in the debian
archive for Amanda. I am using vtapes also.

Thanks for the help,
Gordon

Re: Best Amanda Setup for Restores

[Jean-Louis Martineau]

Dustin J. Mitchell wrote:

On Fri, Feb 22, 2008 at 12:48 PM, Gordon J. Mills III
<[EMAIL PROTECTED]> wrote:
  

 My first question is: How can I have Amanda use a different path for this
 task. I have an NFS mounted external drive that has much more room on it.



If you're using amrecover, you should 'cd' into the directory into
which you want to recover your files before starting amrecover.
  
For safety, you should always start amrecover on an empty directory. 
Amrecover can erase files from the working directory.


Jean-Louis

RE: Best Amanda Setup for Restores

[Gordon J. Mills III]

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dustin J.
Mitchell
Sent: Friday, February 22, 2008 12:24 PM

>Are you using amrestore or amrecover?  Amrecover will only actually
>write the files you have requested to disk, although your observation
>that it reads the entire dump from tape is correct.

Dustin, thanks for the response. I am using amrecover but it will extract
the entire dle first. Is there a way I can redirect it to use a different
directory for this so that I can extract this 100G dle? After that I will
try to break it up.

>>  My first question is: How can I have Amanda use a different path for
this
>>  task. I have an NFS mounted external drive that has much more room on
it.

>If you're using amrecover, you should 'cd' into the directory into
>which you want to recover your files before starting amrecover.

This is not really what I'm asking. Apparently Amanda has a default to
restore the dle into /tmp/Amanda before actually extracting what you
requested. I did cd to the proper directory first, but this is irrelevant
here. I need to be able to change the location that it restores the dle
(before doing the actual restore) because the drive that /tmp is on is not
large enough to handle it.

>Currently, that's the only solution -- smaller DLEs.
That is a bummer.

>Not yet, but that is a major aim of the Application API and transfer
>architecture, which are under active development.  For a hint as to
>how complex this is, see
>http://wiki.zmanda.com/index.php/User:Dustin/Data_Handling_Model.
>There's no specific ETA on this development at the moment.

Thanks for the link. I looked at it but I think most of it is beyond my
understanding of things here.

Gordon

Re: Best Amanda Setup for Restores

[Jon LaBadie]

On Fri, Feb 22, 2008 at 11:48:02AM -0600, Gordon J. Mills III wrote:
> I have recently run into a problem with restoring. I want to restore a
> folder on a DLE that is about 100G. The folder I want to restore is small
> but apparently Amanda has to restore the entire DLE to get at the 1 folder.
> It is restoring the large file to /tmp/Amanda/. Unfortunately, I do not have
> enough room there to extract the entire DLE.
> 
> My first question is: How can I have Amanda use a different path for this
> task. I have an NFS mounted external drive that has much more room on it.
> 
> 2nd question: Is there a way to setup Amanda so that it does not have to
> retrieve the entire DLE file? Is it just removing compression from the DLE
> def? Or chopping up the DLE to smaller chunks (this will be hard since most
> of the data is in 1 place on that dle).
> 
> I would like for Amanda to behave in a similar way that I have had other
> backup software behave. That is, if I just want a single file or folder,
> then it scans the tape (in this case vtapes) and retrieves only what I want.
> When I have successfully retrieved files with Amanda before it took a very
> long time. Is there a way to set it up so that it will do this?
> 

Is this simply a case of not realizing that 'amrestore' is generally used
for complete dle restoration while 'amrecover' is used for individual
file/directory recovery?


-- 
Jon H. LaBadie  [EMAIL PROTECTED]
 JG Computing
 12027 Creekbend Drive  (703) 787-0884
 Reston, VA  20194  (703) 787-0922 (fax)

Re: Best Amanda Setup for Restores

[Dustin J. Mitchell]

On Fri, Feb 22, 2008 at 12:48 PM, Gordon J. Mills III
<[EMAIL PROTECTED]> wrote:
> I have recently run into a problem with restoring. I want to restore a
>  folder on a DLE that is about 100G. The folder I want to restore is small
>  but apparently Amanda has to restore the entire DLE to get at the 1 folder.
>  It is restoring the large file to /tmp/Amanda/. Unfortunately, I do not have
>  enough room there to extract the entire DLE.

Are you using amrestore or amrecover?  Amrecover will only actually
write the files you have requested to disk, although your observation
that it reads the entire dump from tape is correct.

>  My first question is: How can I have Amanda use a different path for this
>  task. I have an NFS mounted external drive that has much more room on it.

If you're using amrecover, you should 'cd' into the directory into
which you want to recover your files before starting amrecover.

>  2nd question: Is there a way to setup Amanda so that it does not have to
>  retrieve the entire DLE file? Is it just removing compression from the DLE
>  def? Or chopping up the DLE to smaller chunks (this will be hard since most
>  of the data is in 1 place on that dle).

Currently, that's the only solution -- smaller DLEs.

>  I would like for Amanda to behave in a similar way that I have had other
>  backup software behave. That is, if I just want a single file or folder,
>  then it scans the tape (in this case vtapes) and retrieves only what I want.
>  When I have successfully retrieved files with Amanda before it took a very
>  long time. Is there a way to set it up so that it will do this?

Not yet, but that is a major aim of the Application API and transfer
architecture, which are under active development.  For a hint as to
how complex this is, see
http://wiki.zmanda.com/index.php/User:Dustin/Data_Handling_Model.
There's no specific ETA on this development at the moment.

Dustin

-- 
Storage Software Engineer
http://www.zmanda.com

Best Amanda Setup for Restores

[Gordon J. Mills III]

I have recently run into a problem with restoring. I want to restore a
folder on a DLE that is about 100G. The folder I want to restore is small
but apparently Amanda has to restore the entire DLE to get at the 1 folder.
It is restoring the large file to /tmp/Amanda/. Unfortunately, I do not have
enough room there to extract the entire DLE.

My first question is: How can I have Amanda use a different path for this
task. I have an NFS mounted external drive that has much more room on it.

2nd question: Is there a way to setup Amanda so that it does not have to
retrieve the entire DLE file? Is it just removing compression from the DLE
def? Or chopping up the DLE to smaller chunks (this will be hard since most
of the data is in 1 place on that dle).

I would like for Amanda to behave in a similar way that I have had other
backup software behave. That is, if I just want a single file or folder,
then it scans the tape (in this case vtapes) and retrieves only what I want.
When I have successfully retrieved files with Amanda before it took a very
long time. Is there a way to set it up so that it will do this?

Thanks for the help.

Regards,
Gordon

Re: another setup problem

[Matt Hyclak]

On Thu, Aug 16, 2007 at 02:38:03PM -0700, Brian Maxwell enlightened us:
> I'm sorry to be so bothersome, but I'm having another problem, and after 
> looking through all of the documentation I could find, I can't solve it. 
> 
> In my disklist file, I've defined a new client, named lucy. I set up the 
> amanda client package on lucy according to the directions in the 
> documentation. 
> 
> Here is the contents of the .amandahosts file on lucy:
> odus amanda
> 
> That is the only configuration file I set up on lucy. Should there be 
> others?
> 
> When I run amcheck back on odus (the backup server), I get this:
> 
> odus 31 /home3/local/etc/amanda/normal> amcheck normal
> Amanda Tape Server Host Check
> -
> Holding disk /home3/amanda/holding: 7297619 KB disk space available, using 
> 7195219 KB
> slot 3: read label `normal104', date `X'
> NOTE: skipping tape-writable test
> Tape normal104 label ok
> NOTE: conf info dir /usr/local/var/amanda/normal/curinfo does not exist
> NOTE: it will be created on the next run.
> NOTE: index dir /usr/local/var/amanda/normal/index does not exist
> NOTE: it will be created on the next run.
> Server check took 0.528 seconds
> 
> Amanda Backup Client Hosts Check
> 
> WARNING: lucy: selfcheck request failed: timeout waiting for ACK
> Client check: 2 hosts checked in 30.162 seconds, 1 problem found
> 
> (brought to you by Amanda 2.5.1)
> 
> Odus and lucy appear in the disklist file like this:
> 
> odus /dev/dsk/c0t1d0s5 hard-disk-dump # tools (/opt)
> lucy /dev/dsk/c0t0d0s7 hard-disk-dump # gmls (/export/home)
> 
> Does anyone have any idea why amanda on odus can't talk to amanda on lucy?
> 
> Thanks in advance,
> 

Did you place the correct entries in (x)inetd and restart (x)inetd?

Matt

-- 
Matt Hyclak
Department of Mathematics 
Department of Social Work
Ohio University
(740) 593-1263

Re: setup problem

[Brian Maxwell]

Thanks all for getting me on the right track. Adding amanda to the sys 
group fixed the problem.

 


Brian Maxwell
Sr. Software Engineer I
SWCM RAM Block 1 & 2 GMLS and Phalanx/SeaRAM
520.794.9076
520.498.5962 Pager
520.975.4725 Cell
[EMAIL PROTECTED] 

 
<><>

another setup problem

[Brian Maxwell]

Hi all,

I'm sorry to be so bothersome, but I'm having another problem, and after 
looking through all of the documentation I could find, I can't solve it. 

In my disklist file, I've defined a new client, named lucy. I set up the 
amanda client package on lucy according to the directions in the 
documentation. 

Here is the contents of the .amandahosts file on lucy:
odus amanda

That is the only configuration file I set up on lucy. Should there be 
others?

When I run amcheck back on odus (the backup server), I get this:

odus 31 /home3/local/etc/amanda/normal> amcheck normal
Amanda Tape Server Host Check
-
Holding disk /home3/amanda/holding: 7297619 KB disk space available, using 
7195219 KB
slot 3: read label `normal104', date `X'
NOTE: skipping tape-writable test
Tape normal104 label ok
NOTE: conf info dir /usr/local/var/amanda/normal/curinfo does not exist
NOTE: it will be created on the next run.
NOTE: index dir /usr/local/var/amanda/normal/index does not exist
NOTE: it will be created on the next run.
Server check took 0.528 seconds

Amanda Backup Client Hosts Check

WARNING: lucy: selfcheck request failed: timeout waiting for ACK
Client check: 2 hosts checked in 30.162 seconds, 1 problem found

(brought to you by Amanda 2.5.1)

Odus and lucy appear in the disklist file like this:

odus /dev/dsk/c0t1d0s5 hard-disk-dump # tools (/opt)
lucy /dev/dsk/c0t0d0s7 hard-disk-dump # gmls (/export/home)

Does anyone have any idea why amanda on odus can't talk to amanda on lucy?

Thanks in advance,

Brian Maxwell

RE: setup problem

[donald.ritchey]

The various dump programs on Solaris need access to the raw device for
the file system.  The rundump command is set-uid root so it can make the
backups, but the amcheck portion is not.  The solution I found for my
Solaris systems was to change the group membership to the 'amanda'
account to the group that owns the devices and make sure the devices are
read-only to that group.
 
Does that help?

Don Ritchey 
IT ED RTS Tech Services, Senior IT Analyst (UNIX), Exelon Corporation

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Brian Maxwell
Sent: Thursday, August 16, 2007 12:05 PM
To: amanda-users@amanda.org
    Subject: setup problem



Hi all, 

I hope you can help! I'm just setting up amanda on a solaris 5.7
server, and I'm getting this error: 

odus 99 /home3/local/etc> amcheck normal 
Amanda Tape Server Host Check 
- 
Holding disk /home3/amanda/holding: 7302172 KB disk space
available, using 7199772 KB 
slot 1: read label `normal101', date `X' 
NOTE: skipping tape-writable test 
Tape normal101 label ok 
NOTE: conf info dir /usr/local/var/amanda/normal/curinfo does
not exist 
NOTE: it will be created on the next run. 
NOTE: index dir /usr/local/var/amanda/normal/index does not
exist 
NOTE: it will be created on the next run. 
Server check took 0.525 seconds 

Amanda Backup Client Hosts Check 
 
ERROR: odus: [could not access /dev/rdsk/c0t1d0s5
(/export/home0): Permission denied] 
Client check: 1 host checked in 0.141 seconds, 1 problem found 

(brought to you by Amanda 2.5.1) 

Any idea what might be causing the error in the client hosts
check? I don't see anything wrong with the permissions for
/export/home0. 

Thanks, 

Brian Maxwell




-
**
This e-mail and any of its attachments may contain Exelon
Corporation proprietary information, which is privileged,
confidential, or subject to copyright belonging to the Exelon
Corporation family of Companies.
This e-mail is intended solely for the use of the individual or
entity to which it is addressed.  If you are not the intended
recipient of this e-mail, you are hereby notified that any
dissemination, distribution, copying, or action taken in relation
to the contents of and attachments to this e-mail is strictly
prohibited and may be unlawful.  If you have received this e-mail
in error, please notify the sender immediately and permanently
delete the original and any copy of this e-mail and any printout.
Thank You.
**

Re: setup problem

[Brian Cuttler]

What group is your amanda user in ?

Follow the links from the device /dev/rdsk/c0t1d0s5 to the
/devices/... and check the protections on that.

you may want amanda in a different group or something.

On Thu, Aug 16, 2007 at 10:04:42AM -0700, Brian Maxwell wrote:
> Hi all,
> 
> I hope you can help! I'm just setting up amanda on a solaris 5.7 server, 
> and I'm getting this error:
> 
> odus 99 /home3/local/etc> amcheck normal
> Amanda Tape Server Host Check
> -
> Holding disk /home3/amanda/holding: 7302172 KB disk space available, using 
> 7199772 KB
> slot 1: read label `normal101', date `X'
> NOTE: skipping tape-writable test
> Tape normal101 label ok
> NOTE: conf info dir /usr/local/var/amanda/normal/curinfo does not exist
> NOTE: it will be created on the next run.
> NOTE: index dir /usr/local/var/amanda/normal/index does not exist
> NOTE: it will be created on the next run.
> Server check took 0.525 seconds
> 
> Amanda Backup Client Hosts Check
> 
> ERROR: odus: [could not access /dev/rdsk/c0t1d0s5 (/export/home0): 
> Permission denied]
> Client check: 1 host checked in 0.141 seconds, 1 problem found
> 
> (brought to you by Amanda 2.5.1)
> 
> Any idea what might be causing the error in the client hosts check? I 
> don't see anything wrong with the permissions for /export/home0.
> 
> Thanks,
> 
> Brian Maxwell---
   Brian R Cuttler [EMAIL PROTECTED]
   Computer Systems Support(v) 518 486-1697
   Wadsworth Center(f) 518 473-6384
   NYS Department of HealthHelp Desk 518 473-0773



IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.

setup problem

[Brian Maxwell]

Hi all,

I hope you can help! I'm just setting up amanda on a solaris 5.7 server, 
and I'm getting this error:

odus 99 /home3/local/etc> amcheck normal
Amanda Tape Server Host Check
-
Holding disk /home3/amanda/holding: 7302172 KB disk space available, using 
7199772 KB
slot 1: read label `normal101', date `X'
NOTE: skipping tape-writable test
Tape normal101 label ok
NOTE: conf info dir /usr/local/var/amanda/normal/curinfo does not exist
NOTE: it will be created on the next run.
NOTE: index dir /usr/local/var/amanda/normal/index does not exist
NOTE: it will be created on the next run.
Server check took 0.525 seconds

Amanda Backup Client Hosts Check

ERROR: odus: [could not access /dev/rdsk/c0t1d0s5 (/export/home0): 
Permission denied]
Client check: 1 host checked in 0.141 seconds, 1 problem found

(brought to you by Amanda 2.5.1)

Any idea what might be causing the error in the client hosts check? I 
don't see anything wrong with the permissions for /export/home0.

Thanks,

Brian Maxwell

Question on setup

[Donofrio, Lewis]

I'd like to backup 474gib of data, and hold this and increment for one
week.  Would I just adjust the dumpcycle to 7 instead of default of 10.

I need enough holding space to properly stream to vtapes or to a jukebox
in the future.

--thanks for you time.
__
Lewis [EMAIL PROTECTED]
Cell: (734) 323-8776 
 

Re: amrecover setup issues -- 2.5.1p3 on Solaris 9

[Jean-Louis Martineau]

ON="Amanda-2.5.1p3"
amandad:BUILT_DATE="Fri Feb 16 14:56:40 EST 2007"
amandad:BUILT_MACH="SunOS mormyrid 5.9 Generic_118558-35 sun4u
sparc SUNW,Ultra-250"
amandad:CC="gcc"
amandad:CONFIGURE_COMMAND="'./configure' '--with-user=amanda'
'--with-group=amanda' '--with-gnutar=/
usr/local/etc/amanda/tools/gtar' '--with-ssh-security'"
amandad: paths: bindir="/usr/local/bin" sbindir="/usr/local/sbin"
amandad:libexecdir="/usr/local/libexec" mandir="/usr/local/man"
amandad:AMANDA_TMPDIR="/tmp/amanda" AMANDA_DBGDIR="/tmp/amanda"
amandad:CONFIG_DIR="/usr/local/etc/amanda" DEV_PREFIX="/dev/dsk/"
amandad:RDEV_PREFIX="/dev/rdsk/"
amandad:DUMP="/usr/local/etc/amanda/tools/ufsdump"
amandad:RESTORE="/usr/sbin/ufsrestore" VDUMP=UNDEF VRESTORE=UNDEF
amandad:XFSDUMP=UNDEF XFSRESTORE=UNDEF VXDUMP=UNDEF VXRESTORE=UNDEF
amandad:SAMBA_CLIENT=UNDEF GNUTAR="/usr/local/etc/amanda/tools/gtar"
amandad:COMPRESS_PATH="/usr/bin/gzip"
amandad:UNCOMPRESS_PATH="/usr/bin/gzip" LPRCMD="/usr/bin/lp"
amandad:MAILER="/usr/bin/mailx"
amandad:listed_incr_dir="/usr/local/var/amanda/gnutar-lists"
amandad: defs:  DEFAULT_SERVER="mormyrid" DEFAULT_CONFIG="DailySet1"
amandad:DEFAULT_TAPE_SERVER="mormyrid" HAVE_MMAP HAVE_SYSVSHM
amandad:LOCKING=POSIX_FCNTL SETPGRP_VOID DEBUG_CODE
amandad:AMANDA_DEBUG_DAYS=4 BSD_SECURITY RSH_SECURITY
USE_AMANDAHOSTS
amandad:CLIENT_LOGIN="amanda" FORCE_USERID HAVE_GZIP
amandad:COMPRESS_SUFFIX=".gz" COMPRESS_FAST_OPT="--fast"
amandad:COMPRESS_BEST_OPT="--best" UNCOMPRESS_OPT="-dc"
amandad: time 0.000: dgram_recv(dgram=ff343d70, timeout=0,
fromaddr=ff353d5c)
amandad: time 0.001: (sockaddr_in *)ff353d5c = { 2, 732, 172.30.52.128 }
security_handleinit(handle=285a0, driver=ff343980 (BSD))
amandad: time 0.004: accept recv REQ pkt:
<<<<<
SERVICE amindexd
OPTIONS features=feff9ffeff7f;auth=bsd;
  
amandad: time 0.004: amindexd: invalid service

amandad: time 0.004: sending NAK pkt:
<<<<<
ERROR amindexd: invalid service
  
amandad: dgram_send_addr(addr=ffbff8c8, dgram=ff343d70)

amandad: time 0.004: (sockaddr_in *)ffbff8c8 = { 2, 732, 172.30.52.128 }
amandad: dgram_send_addr: ff343d70->socket = 0
security_close(handle=285a0, driver=ff343980 (BSD))
amandad: time 30.005: pid 28453 finish time Tue Mar 13 11:31:34 2007

# cd ../client
# ls
daily
# cd daily
# pwd
/tmp/amanda/client/daily
# ls
amrecover.20070313113104.debug
# more amrecover.20070313113104.debug

amrecover: debug 1 pid 28452 ruid 0 euid 0: start at Tue Mar 13 11:31:04
2007
Could not open conf file "/usr/local/etc/amanda/amanda-client.conf": No
such file or directory
Could not open conf file
"/usr/local/etc/amanda/daily/amanda-client.conf": No such file or directory
amrecover: debug 1 pid 28452 ruid 0 euid 0: rename at Tue Mar 13
11:31:04 2007
security_getdriver(name=bsd) returns ff343980
security_handleinit(handle=3d900, driver=ff343980 (BSD))
amrecover: bind_portrange2: Try  port 732: Available   - Success
amrecover: dgram_bind: socket bound to 0.0.0.0.732
amrecover: dgram_send_addr(addr=ffbff978, dgram=ff343d70)
amrecover: (sockaddr_in *)ffbff978 = { 2, 10080, 172.30.52.128 }
amrecover: dgram_send_addr: ff343d70->socket = 4
amrecover: dgram_recv(dgram=ff343d70, timeout=0, fromaddr=ff353d5c)
amrecover: (sockaddr_in *)ff353d5c = { 2, 10080, 172.30.52.128 }
security_close(handle=3d900, driver=ff343980 (BSD))

#


I'm assuming that, since I am on the backup server itself, and it can
access the full amanda.conf in /usr/local/etc/amanda/daily, I therefore
don't need the amanda-client.conf. I do see that error in the amrecover
debug, but I figured it is just letting me know and will then use the
amanda.conf. Anyway, it doesn't seem like that would cause the invalid
service, especially since it does then proceed with what looks like an
attempt to make a connection.

I used the quick setup directions on the zmanda wiki, and have prowled
and searched the rest of the wiki and the web for additional information
as needed.

TIA


---

Chris Hoogendyk

-
   O__   Systems Administrator
  c/ /'_ --- Biology & Geology Departments
 (*) \(*) -- 140 Morrill Science Center
~~ - University of Massachusetts, Amherst 


<[EMAIL PROTECTED]>

--- 


Erdös 4


  

amrecover setup issues -- 2.5.1p3 on Solaris 9

[Chris Hoogendyk]

27; '--with-gnutar=/
usr/local/etc/amanda/tools/gtar' '--with-ssh-security'"
amandad: paths: bindir="/usr/local/bin" sbindir="/usr/local/sbin"
amandad:libexecdir="/usr/local/libexec" mandir="/usr/local/man"
amandad:AMANDA_TMPDIR="/tmp/amanda" AMANDA_DBGDIR="/tmp/amanda"
amandad:CONFIG_DIR="/usr/local/etc/amanda" DEV_PREFIX="/dev/dsk/"
amandad:RDEV_PREFIX="/dev/rdsk/"
amandad:DUMP="/usr/local/etc/amanda/tools/ufsdump"
amandad:RESTORE="/usr/sbin/ufsrestore" VDUMP=UNDEF VRESTORE=UNDEF
amandad:XFSDUMP=UNDEF XFSRESTORE=UNDEF VXDUMP=UNDEF VXRESTORE=UNDEF
amandad:SAMBA_CLIENT=UNDEF GNUTAR="/usr/local/etc/amanda/tools/gtar"
amandad:COMPRESS_PATH="/usr/bin/gzip"
amandad:UNCOMPRESS_PATH="/usr/bin/gzip" LPRCMD="/usr/bin/lp"
amandad:MAILER="/usr/bin/mailx"
amandad:listed_incr_dir="/usr/local/var/amanda/gnutar-lists"
amandad: defs:  DEFAULT_SERVER="mormyrid" DEFAULT_CONFIG="DailySet1"
amandad:DEFAULT_TAPE_SERVER="mormyrid" HAVE_MMAP HAVE_SYSVSHM
amandad:LOCKING=POSIX_FCNTL SETPGRP_VOID DEBUG_CODE
amandad:AMANDA_DEBUG_DAYS=4 BSD_SECURITY RSH_SECURITY
USE_AMANDAHOSTS
amandad:CLIENT_LOGIN="amanda" FORCE_USERID HAVE_GZIP
amandad:COMPRESS_SUFFIX=".gz" COMPRESS_FAST_OPT="--fast"
amandad:COMPRESS_BEST_OPT="--best" UNCOMPRESS_OPT="-dc"
amandad: time 0.000: dgram_recv(dgram=ff343d70, timeout=0,
fromaddr=ff353d5c)
amandad: time 0.001: (sockaddr_in *)ff353d5c = { 2, 732, 172.30.52.128 }
security_handleinit(handle=285a0, driver=ff343980 (BSD))
amandad: time 0.004: accept recv REQ pkt:
<<<<<
SERVICE amindexd
OPTIONS features=feff9ffeff7f;auth=bsd;
>>>>>
amandad: time 0.004: amindexd: invalid service
amandad: time 0.004: sending NAK pkt:
<<<<<
ERROR amindexd: invalid service
>>>>>
amandad: dgram_send_addr(addr=ffbff8c8, dgram=ff343d70)
amandad: time 0.004: (sockaddr_in *)ffbff8c8 = { 2, 732, 172.30.52.128 }
amandad: dgram_send_addr: ff343d70->socket = 0
security_close(handle=285a0, driver=ff343980 (BSD))
amandad: time 30.005: pid 28453 finish time Tue Mar 13 11:31:34 2007

# cd ../client
# ls
daily
# cd daily
# pwd
/tmp/amanda/client/daily
# ls
amrecover.20070313113104.debug
# more amrecover.20070313113104.debug

amrecover: debug 1 pid 28452 ruid 0 euid 0: start at Tue Mar 13 11:31:04
2007
Could not open conf file "/usr/local/etc/amanda/amanda-client.conf": No
such file or directory
Could not open conf file
"/usr/local/etc/amanda/daily/amanda-client.conf": No such file or directory
amrecover: debug 1 pid 28452 ruid 0 euid 0: rename at Tue Mar 13
11:31:04 2007
security_getdriver(name=bsd) returns ff343980
security_handleinit(handle=3d900, driver=ff343980 (BSD))
amrecover: bind_portrange2: Try  port 732: Available   - Success
amrecover: dgram_bind: socket bound to 0.0.0.0.732
amrecover: dgram_send_addr(addr=ffbff978, dgram=ff343d70)
amrecover: (sockaddr_in *)ffbff978 = { 2, 10080, 172.30.52.128 }
amrecover: dgram_send_addr: ff343d70->socket = 4
amrecover: dgram_recv(dgram=ff343d70, timeout=0, fromaddr=ff353d5c)
amrecover: (sockaddr_in *)ff353d5c = { 2, 10080, 172.30.52.128 }
security_close(handle=3d900, driver=ff343980 (BSD))

#


I'm assuming that, since I am on the backup server itself, and it can
access the full amanda.conf in /usr/local/etc/amanda/daily, I therefore
don't need the amanda-client.conf. I do see that error in the amrecover
debug, but I figured it is just letting me know and will then use the
amanda.conf. Anyway, it doesn't seem like that would cause the invalid
service, especially since it does then proceed with what looks like an
attempt to make a connection.

I used the quick setup directions on the zmanda wiki, and have prowled
and searched the rest of the wiki and the web for additional information
as needed.

TIA


---

Chris Hoogendyk

-
   O__   Systems Administrator
  c/ /'_ --- Biology & Geology Departments
 (*) \(*) -- 140 Morrill Science Center
~~ - University of Massachusetts, Amherst 

<[EMAIL PROTECTED]>

--- 

Erdös 4

Advice for setup and backup strategy needed

[Markus Koppenberger]

Hi,

I'm quite new to amanda but as far as I have read about it and after
some first tries, it seems that it could do the job for our needs, but I
would need some advice for a backup strategy and for how to setup up amanda

I will describe you our current situation and our needs as well as my
thoughts of how to do it.

Every help is very apreciated!


We are planing to change from hard-disk backup (mirroring) to use amanda
to backup our RAID storage(s) with a total of 9 TB available on a 8x
LTO-2 (200GB) tape changer.

Currently about 4.8 TB of total storage space on three partitions is used:
  - 3.0T on a 3.3T RAID,
  - 1.4T on a 1.9T RAID and
  - 0.4T on another 1.9TB RAID.

Most of the stored data will not change over time, but the total amount
of data will (slowly) grow.

So what we would need is an initial full backup followed by regular
(maybe daily) incremental backups (which will be quite small)

Of course it is possible to split up the backup into smaller chunks
(single directories, not 3.3 T at once) but most of these smaller chunks
still would be greater then the 200G tapes we have and therefor I would
use tape-spanning.

Most of these directories have the following structure:

- directory1
   +- bigdata (> 200 GB, but does not change [often])
   +- otherdata (smaller [<< 200 GB], but changes more often)
- directory2
   +- bigdata
   +- otherdata
- ...


This could maybe used to define different backup-set:
- set(s): bigdata:
contains the 'bigdata' subdirectory of one (ore more,
depending on total size) directories be run maybe manually when
new files are added.
- set(s): otherdata:
contains the 'otherdata' sub-directories of various directories and
which will be run on daily basis with maybe monthly or two-weekly
full backup

Is this possible and does it make sense to split up the directories on
sub-directory basis into two (well actually more) backup sets ('big' and
'other')?

And if this would be the right way to do it, how would I configure the
different sets?

Maybe you could give me a general hint on how to use tape-spanning,
especially how to configure the 'runtapes' parameter.
Am I right that I have to configure it that 'runtapes' * 'size-of-tape'
= 'total_size_to_backup'? So, that when I have to backup let's say 1TB I
would need a minimum of runtapes=5 that the whole (full) backup fits on
the tapes ( 5 x 200GB)?

Thanks in advance,
koppi

-- 
Grupo de Tecnologia Musical (MTG)
Universidad Pompeu Fabra (UPF), Barcelona, Espana
http://www.iua.upf.es/mtg/



signature.asc
Description: OpenPGP digital signature

Re: setup help with archive type backups

[Joshua Baker-LePain]

On Tue, 7 Nov 2006 at 5:41pm, Bgs wrote

I'd like to as some experienced Amanda user what approach would be best for 
us (or is amanda the best solution for this at all...)


We do archiving and not classic share backups. That is we assemble some raw 
material from time to time and archive it to tape. They always have new file, 
no file change involved. Consequently we do not erase and rotate tapes 
either.


On the source side, a source directory with date named directories looks to 
be a good solution, but how should I set up the tape part?


Just setup a config with 'dumpcycle 0', 'runspercycle 1', and a very large 
tapecycle.  That will force amanda to do full dumps every amdump and never 
recycle tapes.  For the source directory, you could re-use the same DLE 
every time.  Or, what I do for my 'archive random bits' config, add a DLE 
(or DLEs), amdump it/them, and then comment out those DLEs.  That way your 
disklist contains a record of everything you've archvied with that config.


--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University

setup help with archive type backups

[Bgs]

 Greetings,

I'd like to as some experienced Amanda user what approach would be best 
for us (or is amanda the best solution for this at all...)


We do archiving and not classic share backups. That is we assemble some 
raw material from time to time and archive it to tape. They always have 
new file, no file change involved. Consequently we do not erase and 
rotate tapes either.


On the source side, a source directory with date named directories looks 
to be a good solution, but how should I set up the tape part?


Thanks
Bgs

PS: Btw: We have a Tandberg LTO3 changer. Does anyone have an official 
definition for it or should I use the one I assembled?

Re: Disk to disk backup setup

[Paul Bijnens]

On 2006-10-20 13:59, Yogesh Hasabnis wrote:
> 
> We have decided to use Amanda for our backup requirements. We are in the
> process of procuring our backup device. By the time the device arrives,
> I wish to do some hands-on and get more comfortable with Amanda. For
> this, I plan to configure a disk-to-disk backup setup with one machine
> acting as a backup server and one machine acting as backup client. I
> tried to find out the docs for disk-to-disk backup setup in the Amanda
> documentation, but I couldn't find any. Would be grateful if I get some
> pointers for the same.


http://wiki.zmanda.com/index.php/File_driver

And, because you said you're experimenting with Amanda, also have
a look at:

http://wiki.zmanda.com/index.php/Quick_start
http://wiki.zmanda.com/index.php/Test_environment_with_virtual_tapes


-- 
Paul Bijnens, xplanation Technology ServicesTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, ^^, *
* F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... *
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: Disk to disk backup setup

[Gene Heskett]

On Friday 20 October 2006 07:59, Yogesh Hasabnis wrote:
>Hi All,
>
> We have decided to use Amanda for our backup requirements. We are in the
> process of procuring our backup device. By the time the device arrives,
> I wish to do some hands-on and get more comfortable with Amanda. For
> this, I plan to configure a disk-to-disk backup setup with one machine
> acting as a backup server and one machine acting as backup client. I
> tried to find out the docs for disk-to-disk backup setup in the Amanda
> documentation, but I couldn't find any. Would be grateful if I get some
> pointers for the same.
>
Look up the "FILE:" device.  And I've found that its much more dependable 
than a real tape in the form of a dds2 drive ever was.  Of course dds2 is 
considered to be one of those 4 letter words you don't use in polite 
company too...  You get what you pay for, or TANSTAAFL. :)

> Thanks In Advance
>
> Yogesh H
>
>
>-
>Stay in the know. Pulse on the new Yahoo.com.  Check it out.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2006 by Maurice Eugene Heskett, all rights reserved.

Disk to disk backup setup

[Yogesh Hasabnis]

Hi All,  We have decided to use Amanda for our backup requirements. We are in the process of procuring our backup device. By the time the device arrives, I wish to do some hands-on and get more comfortable with Amanda. For this, I plan to configure a disk-to-disk backup setup with one machine acting as a backup server and one machine acting as backup client. I tried to find out the docs for disk-to-disk backup setup in the Amanda documentation, but I couldn't find any. Would be grateful if I get some pointers for the same.  Thanks In Advance  Yogesh H  
		Stay in the know. Pulse on the new Yahoo.com.  Check it out. 

Re: Another setup question...

[Paul Bijnens]

On 2006-04-05 17:34, Bruce Thompson wrote:

Hi Geert,

Thanks for the idea. At the moment I'm using my WRT54G access point as 
my DHCP server and HyperWRT doesn't provide that option.


If my memory is still good, that box does support it.
Hun the webconfiguration for DHCP and there you find some
form where you can add mac-adresses -> fixed-ip numbers.
(I believe that such a box even runs a Linux kernel.)

--
Paul Bijnens, xplanation Technology ServicesTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, ^^, *
* F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... *
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: Another setup question...

[Bruce Thompson]

Hi Geert,

Thanks for the idea. At the moment I'm using my WRT54G access point  
as my DHCP server and HyperWRT doesn't provide that option.


If switching my PowerBook becomes an issue then I'll switch over to  
using my Linux box as the DHCP server which will allow me to set up  
dhcpd as you suggest.


Thanks again!
Bruce.

On Apr 5, 2006, at 8:17 AM, Geert Uytterhoeven wrote:


Just give your backup clients a fixed address in dhcpd.conf, like

host myhost {
hardware ethernet 00:11:22:33:44:55;
fixed-address myhost.at.mydomain;
}

and make sure myhost.at.mydomain is in your DNS config, too.

Then your clients will always receive the same IP address.

Gr{oetje,eeting}s,


--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

Re: Another setup question...

[Jon LaBadie]

On Wed, Apr 05, 2006 at 08:00:23AM -0700, Bruce Thompson wrote:
> Hi all,
> 
> First of all, thanks Paul for the help with my server's setup. I've  
> completed a full backup, a test restore, and I'm now burning the  
> vtapes to DVD (thanks for the capacity note too, I've adjusted my  
> tapedev accordingly).
> 
> Next step is to set up our two PowerBooks. Currently, both PowerBooks  
> get their IP address via DHCP. While switching them to static  
> addresses is not impossible, it's a bit inconvenient for me moving  
> between home and work.
> 
> From what I can tell, Amanda wants a hostname for the clients. Is  
> there an easy way anyone knows of to set it up with dynamic client  
> addresses?

Most systems allow a single network interface to respond to multiple
IP addresses.  If, for example, your primary interface is eth0, you
could also setup eth0:1 , eth0:2 , etc.

Each of those alternate interfaces can have distinct setups.  Perhaps
you could have your primary interface, eth0, use dhcp and that would
be suitable for home and office use as you now do.  But also setup a
eth0:1 with a fixed, static IP that you list in your host files or DNS.
Haven't tried it, but it might work.


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

Re: Another setup question...

[Geert Uytterhoeven]

On Wed, 5 Apr 2006, Bruce Thompson wrote:
> Next step is to set up our two PowerBooks. Currently, both PowerBooks get
> their IP address via DHCP. While switching them to static addresses is not
> impossible, it's a bit inconvenient for me moving between home and work.
> 
> From what I can tell, Amanda wants a hostname for the clients. Is there an
> easy way anyone knows of to set it up with dynamic client addresses?

Just give your backup clients a fixed address in dhcpd.conf, like

host myhost {
hardware ethernet 00:11:22:33:44:55;
fixed-address myhost.at.mydomain;
}

and make sure myhost.at.mydomain is in your DNS config, too.

Then your clients will always receive the same IP address.

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [EMAIL PROTECTED]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Another setup question...

[Bruce Thompson]

Hi all,

First of all, thanks Paul for the help with my server's setup. I've  
completed a full backup, a test restore, and I'm now burning the  
vtapes to DVD (thanks for the capacity note too, I've adjusted my  
tapedev accordingly).


Next step is to set up our two PowerBooks. Currently, both PowerBooks  
get their IP address via DHCP. While switching them to static  
addresses is not impossible, it's a bit inconvenient for me moving  
between home and work.


From what I can tell, Amanda wants a hostname for the clients. Is  
there an easy way anyone knows of to set it up with dynamic client  
addresses?


Thanks!
Bruce.


--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

Re: AMANDA+Samba setup...

[boztr0n]

hey paul,

On 6/24/05, Paul Bijnens <[EMAIL PROTECTED]> wrote:
> because the logs that contain the above lines is in /tmp/amanda/
> unless you make special arrangements during compilation of the
> client programs.
> 

the log files are indeed in /var with the stock Fedora build.
> Obvious question:  the /etc/amandapass is createdon the right
> computer, i.e. the server executing smbclient, that does not
> necessarily is the amandaserver itself.

yes, it is setup on the right machine.

> Maybe set debugging level on samba very high somehow (e.g.
> by replacing smblient with a wrapper like:
> 
>exec /usr/bin/smbclient-real -d 9  ${1+"$@"}

this is a good idea... I'll give it a try when I get back to the
office next week.

Re: AMANDA+Samba setup...

[Jay Fenlason]

On Fri, Jun 24, 2005 at 10:04:49AM -0400, Matt Hyclak wrote:
> On Fri, Jun 24, 2005 at 03:51:13PM +0200, Paul Bijnens enlightened us:
> > What is also strange is your statement just before the above
> > paragraph:
> > 
> > >When I take a look at the Amanda logs in /var here is what I see:
> > 
> > because the logs that contain the above lines is in /tmp/amanda/
> > unless you make special arrangements during compilation of the
> > client programs.
> > 
> 
> At some point, I know the redhat RPMS moved that to /var/log/amanda. 

Yup.  Our security team doesn't like important logs left in /tmp.
Come to think of it, I don't either.

-- JF

Re: AMANDA+Samba setup...

[Matt Hyclak]

On Fri, Jun 24, 2005 at 03:51:13PM +0200, Paul Bijnens enlightened us:
> What is also strange is your statement just before the above
> paragraph:
> 
> >When I take a look at the Amanda logs in /var here is what I see:
> 
> because the logs that contain the above lines is in /tmp/amanda/
> unless you make special arrangements during compilation of the
> client programs.
> 

At some point, I know the redhat RPMS moved that to /var/log/amanda. 

Matt

-- 
Matt Hyclak
Department of Mathematics 
Department of Social Work
Ohio University
(740) 593-1263

Re: AMANDA+Samba setup...

[Paul Bijnens]

boztr0n wrote:

selfcheck: time 0.000: checking disk //rockwell/docs
selfcheck: time 0.000: spawning /usr/bin/smbclient in pipeline
selfcheck: argument list: smbclient \\rockwell\docs -U backup -E -W
NETSETGOODS -c quit
selfcheck: time 0.311: samba access error: //rockwell/docs: Error
connecting to 192.168.1.166 (Permission denied): Connection to
rockwell failed: returned 1


This is sort of funny because when I try and execute the smbclient
command wit the same exact parameters from a shell while logged in as
amanda the command executes without error.


My first reaction would be that there is still a difference
with what you type and with what you think amanda does.

What is also strange is your statement just before the above
paragraph:


When I take a look at the Amanda logs in /var here is what I see:


because the logs that contain the above lines is in /tmp/amanda/
unless you make special arrangements during compilation of the
client programs.




Yes, I have /etc/amandapass setup correc


Obvious question:  the /etc/amandapass is createdon the right
computer, i.e. the server executing smbclient, that does not
necessarily is the amandaserver itself.



I've managed to google one fellow
(http://lists.samba.org/archive/samba/2004-February/080145.html) who
has had this problem but his suggestion (turn off "simple file
sharing" and the windoze firewall in XP) did not help.


Maybe set debugging level on samba very high somehow (e.g.
by replacing smblient with a wrapper like:

  exec /usr/bin/smbclient-real -d 9  ${1+"$@"}

(assuming you rename the real smbclient to smbclient-real before).
And hope some lights shines through in the debugging output???

--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... *
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

AMANDA+Samba setup...

[boztr0n]

hey list,

It's my first time playing with Amanda and I've been working for the
past few days to setup one central amanda server on a linux box
(generic fedora 4, using stock Amanda + Samba rpms) to backup a single
directory on 30 or so wintel boxes. Basically where I'm at now is that
amcheck fails one trying to retrieve the files from the wintel boxes
via samba with the following error:

ERROR: bosch: [samba access error: //rockwell/docs: Error connecting
to 192.168.1.166 (Permission denied): Connection to rockwell failed:
returned 1]
ERROR: bosch: [samba access error: //brown/tmp: Error conERROR: bosch:
[samba access error: //rockwell/docs: Error connecting to
192.168.1.166 (Permission denied): Connection to rockwell failed:
returned 1]
ERROR: bosch: [samba access error: //brown/tmp: Error connecting to
192.168.1.131 (Permission denied): Connection to brown failed:
returned 1]
necting to 192.168.1.131 (Permission denied): Connection to brown
failed: returned 1]

... and so on, for each host, the same error. 

When I take a look at the Amanda logs in /var here is what I see:

selfcheck: time 0.000: checking disk //rockwell/docs
selfcheck: time 0.000: spawning /usr/bin/smbclient in pipeline
selfcheck: argument list: smbclient \\rockwell\docs -U backup -E -W
NETSETGOODS -c quit
selfcheck: time 0.311: samba access error: //rockwell/docs: Error
connecting to 192.168.1.166 (Permission denied): Connection to
rockwell failed: returned 1


This is sort of funny because when I try and execute the smbclient
command wit the same exact parameters from a shell while logged in as
amanda the command executes without error.

Yes, I have /etc/amandapass setup correc
I've managed to google one fellow
(http://lists.samba.org/archive/samba/2004-February/080145.html) who
has had this problem but his suggestion (turn off "simple file
sharing" and the windoze firewall in XP) did not help.

Anyone have any ideas?

thanks,
-b

Re: New setup: problem with permissions

[Paul Bijnens]

Dwight Tovey wrote:
Jon LaBadie said:
On Tue, Dec 28, 2004 at 05:15:30PM -0700, Dwight Tovey wrote:
I'm looking at using Amanda for my backup solution, and I've run into a
bit of a problem that I'm hoping to get some help with.
...
If I try to use the device instead of the mountpoint, amcheck (and
eventually amdump) fails with "Can't open disk 'hda6'" for each
'include'.
Looking at the source I found that every time it encounters an
'include',
Amanda trys to do an 'opendir()' to verify the included name.  Since the
device is not a directory, the opendir() fails resulting in the
displayed
error.
Yes. That's a limitation of the current implementation.
For "exclude", amanda can use the --exclude flags of gnutar
doing pattern matches. Gnutar however has no support for a
similar "include" that handles patterns.  That's why amanda
had to expand the patterns herself.
Gnutar itself is run through the "runtar" executable, which is
suid-root to get all the necessary priviledges needed.  But
while expanding the patterns for the "include" directive, amanda
does that herself, and, as you noticed, does not run with root
priviledges.
Ultimately, I would like gnutar to include that include-with-pattern
functionality.   As alternative, we could write yet another suid-root
program that handles the pattern-expansion for the includes.
Also note that the man page states:
"All include expressions... must start with "./" and containing
no other "/".
I.e. amanda limits the expansion to one level deep only (to avoid 
scanning the whole directory tree).  When gnutar would do the
pattern matching of the include-directive, this limitation would
be eliminated too.  Someone brave enough to submit a patch for gnutar?


So, is there any way around this short of running Amanda as 'root'?  Is
there really any problem with running as root?
No real problem.  This only violates the rule to do everything with
the minimum necessary priviledges needed.
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: New setup: problem with permissions

[Dwight Tovey]

Jon LaBadie said:
> On Tue, Dec 28, 2004 at 05:15:30PM -0700, Dwight Tovey wrote:
>> I'm looking at using Amanda for my backup solution, and I've run into a
>> bit of a problem that I'm hoping to get some help with.
>>
> ...
>>
>> If I try to use the device instead of the mountpoint, amcheck (and
>> eventually amdump) fails with "Can't open disk 'hda6'" for each
>> 'include'.
>>  Looking at the source I found that every time it encounters an
>> 'include',
>> Amanda trys to do an 'opendir()' to verify the included name.  Since the
>> device is not a directory, the opendir() fails resulting in the
>> displayed
>> error.
>>
>> So, is there any way around this short of running Amanda as 'root'?  Is
>> there really any problem with running as root?
>
> If the amanda installation was done as root, then all the necessary
> set-user-id-root permissions should have been done at that time.
> Amanda tries to run with the minimum necessary permissions, but
> at times, the minimum is root.  Thus several amanda programs are
> setuid'ed during installation.  These include the "runtar" program
> that invokes gnutar.
>

OK.  I've verified that the "runtar" program is setuid root, so ultimatly
the 'tar' program will be run as root and it should have access to the
protected directories.  So why am I seeing the failure?

In checking the log from the last run, I found:
FAIL dumper medea home_all 20041228 0 [ [access as bkoper not allowed from
[EMAIL PROTECTED] amandahostsauth failed]

The 'medea.dtovey.net' name resolves to the external interface, but I only
had an entry in .amandahosts for the internal interface.  I've now
modified .amandahosts so that there are entries for both interfaces, so
maybe it will work tonight.  However it leaves me wondering why this one
disklist entry tried to come in on the external interface while everything
else was using the internal.

Thanks for the response.
/dwight
-- 
Dwight N. Tovey
email: [EMAIL PROTECTED]
web: http://www.dtovey.net/~dwight
---
What do you call a male ladybug?

Re: New setup: problem with permissions

[Jon LaBadie]

On Tue, Dec 28, 2004 at 05:15:30PM -0700, Dwight Tovey wrote:
> I'm looking at using Amanda for my backup solution, and I've run into a
> bit of a problem that I'm hoping to get some help with.
> 
...
> 
> If I try to use the device instead of the mountpoint, amcheck (and
> eventually amdump) fails with "Can't open disk 'hda6'" for each 'include'.
>  Looking at the source I found that every time it encounters an 'include',
> Amanda trys to do an 'opendir()' to verify the included name.  Since the
> device is not a directory, the opendir() fails resulting in the displayed
> error.
> 
> So, is there any way around this short of running Amanda as 'root'?  Is
> there really any problem with running as root?

If the amanda installation was done as root, then all the necessary
set-user-id-root permissions should have been done at that time.
Amanda tries to run with the minimum necessary permissions, but
at times, the minimum is root.  Thus several amanda programs are
setuid'ed during installation.  These include the "runtar" program
that invokes gnutar.


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

New setup: problem with permissions

[Dwight Tovey]

I'm looking at using Amanda for my backup solution, and I've run into a
bit of a problem that I'm hoping to get some help with.

My tape server (and currently only client) is running Red Hat Linux 7.3. 
It has a 4mm DAT tape drive, which means that I can get up to about
8Gb/tape.  One of my filesystems is 20Gb and currently holds about 9Gb of
data.  Since Amanda doesn't support spanning tapes, I tried to break this
filesystem into chunks using the example that was in the sample 'disklist'
file:

medea /home/dwight /home {
comp-high-tar
comment "Dwight home"
include "./dwight"
exclude "./dwight/downloads" "./dwight/News"
} 2 eth0

medea /home/public /home {
comp-high-tar
comment "Public directories"
include "./samba" "./httpd"
} 2 eth0

medea /home/all /home {
comp-high-tar
comment "Home filesystem"
exclude "./dwight" "./samba" "./httpd"
} 2 eth0


The problem here is that my backup user does not have read access to all
of the directories in this filesystem, so the backups fail.  It looks to
me like normally Amanda wants to just use the disk device (/dev/hda6 for
example) instead of the mountpoint (/home) since the backup user has read
access to the device by being in the 'disk' group.

If I try to use the device instead of the mountpoint, amcheck (and
eventually amdump) fails with "Can't open disk 'hda6'" for each 'include'.
 Looking at the source I found that every time it encounters an 'include',
Amanda trys to do an 'opendir()' to verify the included name.  Since the
device is not a directory, the opendir() fails resulting in the displayed
error.

So, is there any way around this short of running Amanda as 'root'?  Is
there really any problem with running as root?

Thanks in advance for any help/advice.

/dwight

-- 
Dwight N. Tovey
email: [EMAIL PROTECTED]
web: http://www.dtovey.net/~dwight
---
Diplomacy: The art of letting someone else have your way.

RE: HP Autloader (40x6i), Redhat (fedora), Amanda new setup

[Zadikem, Travis-taz]

mt is already installed as well as mtx, I also already have the docs for that 
drive and it should be setup to use.

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Jon LaBadie
Sent: Monday, November 15, 2004 11:35 AM
To: [EMAIL PROTECTED]
Subject: Re: HP Autloader (40x6i), Redhat (fedora), Amanda new setup


On Mon, Nov 15, 2004 at 10:29:43AM -0700, Zadikem, Travis-taz wrote:
> Can someone please point me in setting up the HP Autoloader 40x6 (C5715A) 
> with Amanda.  I am new to running Amanda.  I have never used Amanda before, 
> but have read
> good things about it.  We would like to try and use it for our CVS 
> repository. 
>  
> Thanks

First, get the excellent installation docs for that tape drive,
possibly by downloading the pdf files from their support site.

That changer has very specific OS requirements for installation
including dip switch settings, and the dip switchs are in part,
inside the case requiring case removal.

Next, make sure you have the programs mt and mtx.  Get them
working to load/unload tapes and magazines and get the status
of the tape changer status (mtx) and can rewind a tape and
get the status of the tape drive (mt).

This may require you getting your OS to recognize multiple LUN's
for the scsi driver.

Only then start looking at the amanda how-to docs.

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

Re: HP Autloader (40x6i), Redhat (fedora), Amanda new setup

[Jon LaBadie]

On Mon, Nov 15, 2004 at 10:29:43AM -0700, Zadikem, Travis-taz wrote:
> Can someone please point me in setting up the HP Autoloader 40x6 (C5715A) 
> with Amanda.  I am new to running Amanda.  I have never used Amanda before, 
> but have read
> good things about it.  We would like to try and use it for our CVS 
> repository. 
>  
> Thanks

First, get the excellent installation docs for that tape drive,
possibly by downloading the pdf files from their support site.

That changer has very specific OS requirements for installation
including dip switch settings, and the dip switchs are in part,
inside the case requiring case removal.

Next, make sure you have the programs mt and mtx.  Get them
working to load/unload tapes and magazines and get the status
of the tape changer status (mtx) and can rewind a tape and
get the status of the tape drive (mt).

This may require you getting your OS to recognize multiple LUN's
for the scsi driver.

Only then start looking at the amanda how-to docs.

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

HP Autloader (40x6i), Redhat (fedora), Amanda new setup

[Zadikem, Travis-taz]

Can someone please 
point me in setting up the HP Autoloader 40x6 (C5715A) with Amanda.  I am new to running Amanda.  I have never used Amanda before, but have read
good things about it.  We would like to try 
and use it for our CVS repository. 
 
Thanks
Travis A. Zadikem

Re: Recommended Changer script setup

[Frank Smith]

--On Monday, October 04, 2004 13:20:32 -0400 "Lauro, John" <[EMAIL PROTECTED]> wrote:

> Hello,
>  
> I am new to amanda  and after a slow-going process (not all directly
> from Amanda, some was in first getting MTX to work, etc)  I have it
> mostly working (for one machine, have to add clients to other boxes
> yet...).  (IE: It would be nice if the install script was a little more
> user friendly/complete, such as toching required files (but can be
> blank) such as tapelist...)
>  
> It's working with the chg-zd-mtx, but...  I now realize that it appears
> that chg-zd-mtx only supports one tape drive in the library, even though
> MTX supports multiple drives...
>  
> What/how should I configure it work with a multi-drive library?  There
> appears to be many options...  (It's a Spectra 1 under Linux that I
> am trying to get running, and the library has two drives installed).

You have two choices on how to use two (or more) drives in a changer.

One way is to use the RAIT (RAID using tapes) driver to stripe (or
mirror) your data across the drives.  With striping you should be
able to increase your throughput writing to tape.

Another way is to have two configs, each using one drive and each
backing up a subset of your clients.  You will also need separate
holdingdisks and changer configs for each.

It would be nice if you could just have one config and have taper
be able to write images to multiple drives simultaneously, but as far
as I know it isn't possible.

Frank

>  
> PS: I started to install juke, but it failed a quick make, so I decided
> I should get some recommendations as to what might be the best option.
>  
> Thanks.
>  



-- 
Frank Smith  [EMAIL PROTECTED]
Sr. Systems Administrator   Voice: 512-374-4673
Hoover's Online   Fax: 512-374-4501

Recommended Changer script setup

[Lauro, John]

Hello,
 
I am new to 
amanda  and after a slow-going process (not all directly from Amanda, 
some was in first getting MTX to work, etc)  I have it mostly working (for 
one machine, have to add clients to other boxes yet...).  (IE: It would be 
nice if the install script was a little more user friendly/complete, such as 
toching required files (but can be blank) such as 
tapelist...)
 
It's working with 
the chg-zd-mtx, but...  I now realize that it appears that 
chg-zd-mtx only supports one tape drive in the library, even though MTX 
supports multiple drives...
 
What/how should I 
configure it work with a multi-drive library?  There appears to be many 
options...  (It's a Spectra 1 under Linux that I am trying to get 
running, and the library has two drives installed).
 
PS: I started to 
install juke, but it failed a quick make, so I decided I should get some 
recommendations as to what might be the best option.
 
Thanks.
 

Amanda setup question

[Sabir Pasha]

Hey all,

I'm setting amanda up for the first time and I had a
queston about the optimal setup.  We have a Seagate
Scorpion autoloader over here where I work, that
contains 6 DDS4 tapes(each having a capacity of (20 GB
uncompressed/40 GB compressed) or 180GB/240GB. 

Now, I have to backup over 70 GB data.  We have four
sets of these tapes(i.e 6x4), and would like to rotate
these tapes  every month.  I.e. use the tapes for week
1 over again. 

Right now, I'm just trying to backup the data on the
server(a samba share of 70 GB).  Now from reading I
found out that amanda doesn't span dump images across
tapes.  So does this mean I need one tape for every
day of the week, and two tapes for that one day that a
full backup occurs?? What would be the best way to set
this up??  A little advice/nudge would be much
apperciated.  
I assume this isn't too atypical of a setup.  

Apperciate all the help, thanks!

Sabir Pasha

RE: please, help with samba/windows setup

[Dege, Robert C.]

A few things I would check:

*) Since you area already part of a domain, your amandapass does not require the third 
field that specifies the workgroup.

*) Your disklist specifies nocomp-user-gnutar as the backup method.  What is listed in 
there?

*) check out the errors logs in /tmp/amanda.  It might help highlight a few issues 
with the backup errors.

*) The packages that you are using are fairly old.  You might want to try upgrading 
them as the updated package might contain fixes relevant to your problem.


I am able to successfully backup windows2000 systems using amanda on a redhat box, so 
I do know that this works.  It has nothing to do with your linux box needing to be a 
PDC. So long as your linux box is a member of the domain, you should be okay.

-Rob

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]
> Sent: Thursday, July 08, 2004 9:34 AM
> To: [EMAIL PROTECTED]
> Subject: please, help with samba/windows setup
> 
> 
> Hi there ! 
>  
> I am having trouble to setup amanda to backup shares from a 
> Windows 2000 server. I am not sure this is possible, (I suspect is 
> not), so I am writing to get a confirmation that this is impossible 
> and ask for suggestions to backup the windows 2000 shares. 
>  
> The PDC is a windows 2000 machine. it's NetBIOS name is 
> kgbserver. The shares that I want to backup are named 
> "//kgbserver/Others" and "//kgbserver/Docs". 
>  
> The amanda server is a RH9.0, with samba configured as 
> security=DOMAIN . The samba NetBIOS name is KGBSERVER2.  
> 
> This is why I think this could be a problem. The samba 
> server is not the PDC for that domain. It is just a client for that 
> domain. This makes any sense ?
>  
> In the disklist file I put the following lines for the 
> shares I want to backup: 
>  
> kgbserver2  //kgbserver/Others  nocomp-user-gnutar 
> kgbserver2  //kgbserver/Docs nocomp-user-gnutar 
>  
> The file /etc/amandapass was configured with the 
> Administrator and password for the Windows machine: 
>  
> //kgbserver/Others  administrator%ajjj882 KGB 
> //kgbserver/Docsadministrator%ajjj882 KGB 
>  
>  
> On the amanda server (RH9.0) I can browse the Windows 2000's 
> shares with smbclient, for example: 
>  
> # smbclient -L //kgbserver -U administrator%ajjj882 -W KGB 
> added interface ip=192.168.155.202 bcast=192.168.155.255 
> nmask=255.255.255.0 
> Domain=[KGB] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] 
>  
> Sharename  Type  Comment 
> -    --- 
> 
> ...etc... and browsing a specific share: 
>  
> # smbclient  //kgbserver/Others -U administrator%ajjj882 -W KGB 
> added interface ip=192.168.155.202 bcast=192.168.155.255 
> nmask=255.255.255.0 
> Domain=[KGB] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] 
> smb: \> ls 
>   .   D0  Fri Jun 25 
> 14:46:26 2004 
>   ..  D0  Fri Jun 25 
> 14:46:26 2004 
>   ENVELOPES   D0  Thu Feb 28 
> 15:02:23 2002 
> 
> ...etc 
>  
>  ...and even smbtar work fine: 
>  
> # cd /tmp 
> # smbtar -s kgbserver -u administrator%ajjj882 -x Others -t 
> others.tar 
> # tar -tvf others.tar 
> ...etc... 
>  
>  ...and amcheck works fine too: 
>  
> # /bin/su - amanda -c "/usr/sbin/amcheck -c DialySet1" 
>  
> Amanda Backup Client Hosts Check 
>  
> Client check: 2 hosts checked in 10.682 seconds, 0 problems found 
>  
> (brought to you by Amanda 2.4.3) 
>  
>  
> The amdump fail with the following message: 
>  
> FAILURE AND STRANGE DUMP SUMMARY: 
>   planner: ERROR Request to kgbserver2 timed out. 
>  
>  and from DUMP summary: 
>  
> DUMP SUMMARY: 
>  DUMPER STATSTAPER 
> STATS 
> HOSTNAME DISKL ORIG-KB OUT-KB COMP% MMM:SS  KB/s MMM:SS  
> KB/s 
> -- - 
>  
> kgbserver2   -ver/Others   MISSING 
> -- 
> kgbserver2   -erver/Docs   MISSING 
> -- 
>  
> Is this setup really impossible ? 
> As last resort, I configured Amanda to backup the mount 
> point "/mnt/kgbserver/Others" and "/mnt/kgbserver/Docs" from the 
> machine kgbserver2. 
> 
>  These mount points are mounted at boot time via /etc/fstab us

please, help with samba/windows setup

[Miguel Angelo Rozsas]

Hi there ! 
 
I am having trouble to setup amanda to backup shares from a 
Windows 2000 server. I am not sure this is possible, (I suspect is 
not), so I am writing to get a confirmation that this is impossible 
and ask for suggestions to backup the windows 2000 shares. 
 
The PDC is a windows 2000 machine. it's NetBIOS name is 
kgbserver. The shares that I want to backup are named 
"//kgbserver/Others" and "//kgbserver/Docs". 
 
The amanda server is a RH9.0, with samba configured as 
security=DOMAIN . The samba NetBIOS name is KGBSERVER2.  

This is why I think this could be a problem. The samba 
server is not the PDC for that domain. It is just a client for that 
domain. This makes any sense ?
 
In the disklist file I put the following lines for the 
shares I want to backup: 
 
kgbserver2  //kgbserver/Others  nocomp-user-gnutar 
kgbserver2  //kgbserver/Docs nocomp-user-gnutar 
 
The file /etc/amandapass was configured with the 
Administrator and password for the Windows machine: 
 
//kgbserver/Others  administrator%ajjj882 KGB 
//kgbserver/Docsadministrator%ajjj882 KGB 
 
 
On the amanda server (RH9.0) I can browse the Windows 2000's 
shares with smbclient, for example: 
 
# smbclient -L //kgbserver -U administrator%ajjj882 -W KGB 
added interface ip=192.168.155.202 bcast=192.168.155.255 
nmask=255.255.255.0 
Domain=[KGB] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] 
 
Sharename  Type  Comment 
-    --- 

...etc... and browsing a specific share: 
 
# smbclient  //kgbserver/Others -U administrator%ajjj882 -W KGB 
added interface ip=192.168.155.202 bcast=192.168.155.255 
nmask=255.255.255.0 
Domain=[KGB] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] 
smb: \> ls 
  .   D0  Fri Jun 25 14:46:26 2004 
  ..  D0  Fri Jun 25 14:46:26 2004 
  ENVELOPES   D0  Thu Feb 28 15:02:23 2002 

...etc 
 
 ...and even smbtar work fine: 
 
# cd /tmp 
# smbtar -s kgbserver -u administrator%ajjj882 -x Others -t 
others.tar 
# tar -tvf others.tar 
...etc... 
 
 ...and amcheck works fine too: 
 
# /bin/su - amanda -c "/usr/sbin/amcheck -c DialySet1" 
 
Amanda Backup Client Hosts Check 
 
Client check: 2 hosts checked in 10.682 seconds, 0 problems found 
 
(brought to you by Amanda 2.4.3) 
 
 
The amdump fail with the following message: 
 
FAILURE AND STRANGE DUMP SUMMARY: 
  planner: ERROR Request to kgbserver2 timed out. 
 
 and from DUMP summary: 
 
DUMP SUMMARY: 
 DUMPER STATSTAPER 
STATS 
HOSTNAME DISKL ORIG-KB OUT-KB COMP% MMM:SS  KB/s MMM:SS  
KB/s 
-- -  
kgbserver2   -ver/Others   MISSING -- 
kgbserver2   -erver/Docs   MISSING ------ 
 
Is this setup really impossible ? 
As last resort, I configured Amanda to backup the mount 
point "/mnt/kgbserver/Others" and "/mnt/kgbserver/Docs" from the 
machine kgbserver2. 

 These mount points are mounted at boot time via /etc/fstab using 
smbfs. 
 
Is there another better solution ? 
 
I will appreciated any ideas and comments. 
 
thank you gurus in advance, 
 
Amanda server: RH9.0, running 2.4.20-30.9. 
amanda version: 2.4.3-4 (from the rpm package) 
samba version: 2.2.7a release 8.9.0 (from the rpm package) 
 
-- 
Miguel Angelo Rozsas
K3M Consultoria
19 3232 8886

---
Esta mensagem foi verificada pelo e-mail seguro Maxiweb.
McAfee VirusScan / Vacina v4374 atualizada em 07/07/2004
powered by wstation (www.wstation.com.br)

Re: Trying to figure out a rational setup

[Jon LaBadie]

On Mon, May 17, 2004 at 11:37:27PM -0600, Bret McKee wrote:
> Greetings:
> 
> Inspired by what I think that is telling me, I would like to do is
> something like this:
> 
> - Run compressed backups to the holding disk
> - while the holding disk has > 1 tape worth of data (40 GB), write a
> tape.
> 
> It isn't clear to me how to set it up to get this behavior.  It seems
> like I might be able to get it by having two configurations, but then
> it isn't clear of the tape tracking database features will work like I
> would like (i.e. if one configuration writes to the disk only, and
> another sometimes pushes the disk to a tape, will the database know
> which tape the files from the disk only configuration are on?)


If you can put up with nightly error messages, you could run your
daily backups with amanda.conf containing a fake tape device name
(not /dev/null).  Couple of ways to handle this:

 - Have your cron job execute a script instead of just amdump
   which renames /dev/nst0 (or whatever) to /dev/_nst0, then
   calls amdump, and when done restores the device name.

 - Have amanda.conf edited each night by the cron script

 - Have two otherwise identical configs (use same index/log/
   ... dirs).  One with a fake device name will be used for
   backups.  The other with the correct name will be used
   for flushing and restoring.

At a daily time of your choosing (when amdump is not running),
run a cronjob that du's the holding disk and if sufficiently
large, runs amflush.

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

Re: Trying to figure out a rational setup

[Stefan G. Weichinger]

Hi, Paul,

on Dienstag, 18. Mai 2004 at 23:43 you wrote to amanda-users:

PB> Those vtapes are probably quiet large.

Yeah, I can't hear mine at all.

;-) Stefan

Re: Trying to figure out a rational setup

[Stefan G. Weichinger]

Hi, Paul,

on Dienstag, 18. Mai 2004 at 23:43 you wrote to amanda-users:

PB> Don't forget that such a setup makes restores a PITA:

You are definitely right with this ;-)

This seem to be just some AMANDA-fantasies ...

PB> First you make "vtapes", just a set of large files with a
PB> strange name.  When you have collected enough of these files,
PB> and make a backup with amdump, and then you erase those "vtapes",
PB> and together probably their indexes (or jump through hoops to
PB> avoid this).

PB> Now you don't have a real index of the files on tape, only a list
PB> of hostnames-filesystems; even the dates are missing (because
PB> you put a lot of days on one tape).

PB> To restore, you first have to restore the image from tape,
PB> to get the "vtape" back, then you have to run restore again from
PB> that "vtape".
PB> In real life, you probably need to try a few times, because you're
PB> just guessing on which tape that file actually is.
PB> Those vtapes are probably quiet large.  First wait 2 hours for
PB> one 40 GB tape to read to disk, and then again an hour or so
PB> to restore from that vtape. And again, if you mistyped the name
PB> of the file, or if you want to see if the file was there a day
PB> earlier.

Sucks.

PB> Doing a backup to holdingdisk instead of vtapes avoids this
PB> backup-of-backup-images; amflush migrates the images from disk
PB> to tape instead, and keeps the indexes up to date.

We all like that, don't we?

PB> While creating backups should be easy, *restoring* from backups
PB> should be even more straightforward: it's then that you're working
PB> under high pressure (disk crash?  important file erased?  agry
PB> users waiting etc.)

Brrr ...

;-)

I would just suggest a conventional setup using the changer.

-- 
best regards,
Stefan

Stefan G. Weichinger
mailto:[EMAIL PROTECTED]

Re: Trying to figure out a rational setup

[Paul Bijnens]

Stefan G. Weichinger wrote:
Setup a config with device=, define a DLE
pointing at your vtapes-dir, let the dumptype contain "dumpcycle 0",
which will make it do always full dumps.
Check fillage, if it is close to tapesize, run amdump.

Don't forget that such a setup makes restores a PITA:
First you make "vtapes", just a set of large files with a
strange name.  When you have collected enough of these files,
and make a backup with amdump, and then you erase those "vtapes",
and together probably their indexes (or jump through hoops to
avoid this).
Now you don't have a real index of the files on tape, only a list
of hostnames-filesystems; even the dates are missing (because
you put a lot of days on one tape).
To restore, you first have to restore the image from tape,
to get the "vtape" back, then you have to run restore again from
that "vtape".
In real life, you probably need to try a few times, because you're
just guessing on which tape that file actually is.
Those vtapes are probably quiet large.  First wait 2 hours for
one 40 GB tape to read to disk, and then again an hour or so
to restore from that vtape. And again, if you mistyped the name
of the file, or if you want to see if the file was there a day
earlier.
Doing a backup to holdingdisk instead of vtapes avoids this
backup-of-backup-images; amflush migrates the images from disk
to tape instead, and keeps the indexes up to date.
While creating backups should be easy, *restoring* from backups
should be even more straightforward: it's then that you're working
under high pressure (disk crash?  important file erased?  agry
users waiting etc.)
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***

Re: Trying to figure out a rational setup

[Stefan G. Weichinger]

Hi, Bret ,

on Dienstag, 18. Mai 2004 at 21:59 you wrote to amanda-users:

BM> My real goal here is to set something up that minimizes the amount of
BM> attention it takes for me to make it work.  I'd just stick with what I
BM> have, except that there are new a few PCs that I need to start backing
BM> up as well.

Set up two configs, both using the file-driver.

One for the server, with loong dumpcycle, maybe with a special
"strategy" in the dumptype.

Look up the dumptype-parameter "strategy" in the example amanda.conf.
Maybe "incronly" works for you.

Maybe the second config for the Samba-shares, if you would like to
separate things.

>> If you really want to fill your tapes, you could set up 
>> vtapes (look into the HOWTO-FILE-DRIVER in the docs) that are 
>> nearly as big as your DLTs, and save the vtape-dirs with a 
>> second config to your DLTs.

BM> This sounds promising, but again I want to only write to the tape when
BM> there is enough data to fill it, and I'm not sure how to set up that
BM> config.  I'll read the vtape documents and see what I can learn.

BM> I suppose I can not put amanda into cron, and have a cron job that
BM> sees how full the vtape disk is, and when it is full enough it could
BM> kick off a backup.

Why not?

Setup a config with device=, define a DLE
pointing at your vtapes-dir, let the dumptype contain "dumpcycle 0",
which will make it do always full dumps.

Check fillage, if it is close to tapesize, run amdump.

Do an optional amverify, if OK, you could let the script remove the
vtape from the configs db and "erase" that vtape.

-- This is just quick-and-dirty stuff and should get worked out
better...

>> I think you should just take the straight way, setup your
>> changer, configure a dumptype for your nearly-static stuff 
>> that has a long dumpcycle and start going.
>> 
>> If you leave the tapes out, AMANDA will fill your 
>> holdingdisk, modify the "reserve"-parameter to even get 
>> lev0-dumps in there.

BM> I don't understand the last paragraph.

In a conventional setup it works that way:

If no/the wrong tape is inserted at the runtime of amdump, nothing
will be written to the tapedrive. The images stay in the holdingdisk
and can be flushed to the next tape via amflush or (if the parameter
autoflush is set) is getting flushed automatically at the next run of
amdump.

If the reserve-parameter in amanda.conf is anything else than 100%,
AMANDA will only do incremental backups in that case, but if you have
sufficient space, you can set "reserve" to something else, and then
get full backups to the holdingdisk, even without a tape inserted.

Better?

-- 
best regards,
Stefan

Stefan G. Weichinger
mailto:[EMAIL PROTECTED]

RE: Trying to figure out a rational setup

[Bret McKee]

> Hi, Bret,
> 
> on Dienstag, 18. Mai 2004 at 07:37 you wrote to amanda-users:
> 
> BM> I a couple servers and about a dozen PCs to backup.  
> Based upon my 
> BM> current backup, it seems that only a few GB changes every day.
I 
> BM> have a 20 tape DLT 80 autochanger, which should back 
> everything up 
> BM> for a long time.  My tape machine also has "plenty" of 
> holding disk 
> BM> space (something like 3 tapes worth/120 GB).  The servers 
> have quite 
> BM> a bit of disk space on them, but the vast majority of it never 
> BM> changes (1 server has maybe 100 GB total space that needs to be 
> BM> backed up, but 80 GB of it is basically static).
> 
> BM> My configuration is therefore one which will waste a lot 
> of tape if 
> BM> I let Amanda write a tape every night (and more 
> importantly, it will 
> BM> run out of tapes and pester me in 3 weeks instead of 6
months...)
> 
> You want to run backups for 6 months with 20 tapes?
> Tell us more about that, this is not clear to me ...

I have 20x40 GB = 800 GB, and I generate about 5GB/day of backup
traffic, which is something like 160 days if I can optimally fill the
tapes.  I know I can't but I'd like to try and get close.  Right now
I'm just backing up the server (which has most of the data) with cpio
to what will become the holding disk, doing only level 1 backups (a
restore is always only 2 steps) and I get a few months before I have
to do another archive and start over, keeping something like 30 days
worth of level 1s.  Eventually the level 1s get too big and so I start
over.

My real goal here is to set something up that minimizes the amount of
attention it takes for me to make it work.  I'd just stick with what I
have, except that there are new a few PCs that I need to start backing
up as well.

> 
> BM> Inspired by what I think that is telling me, I would like 
> to do is 
> BM> something like this:
> 
> BM> - Run compressed backups to the holding disk
> BM> - while the holding disk has > 1 tape worth of data (40 
> GB), write a 
> BM> tape.
> 
> BM> It isn't clear to me how to set it up to get this behavior.
> 
> If you really want to fill your tapes, you could set up 
> vtapes (look into the HOWTO-FILE-DRIVER in the docs) that are 
> nearly as big as your DLTs, and save the vtape-dirs with a 
> second config to your DLTs.

This sounds promising, but again I want to only write to the tape when
there is enough data to fill it, and I'm not sure how to set up that
config.  I'll read the vtape documents and see what I can learn.

I suppose I can not put amanda into cron, and have a cron job that
sees how full the vtape disk is, and when it is full enough it could
kick off a backup.

> 
> I think you should just take the straight way, setup your 
> changer, configure a dumptype for your nearly-static stuff 
> that has a long dumpcycle and start going.
> 
> If you leave the tapes out, AMANDA will fill your 
> holdingdisk, modify the "reserve"-parameter to even get 
> lev0-dumps in there.

I don't understand the last paragraph.

Thanks again for the help,

Bret

Re: Trying to figure out a rational setup

[Stefan G. Weichinger]

Hi, Bret,

on Dienstag, 18. Mai 2004 at 07:37 you wrote to amanda-users:

BM> I a couple servers and about a dozen PCs to backup.  Based upon my
BM> current backup, it seems that only a few GB changes every day.  I have
BM> a 20 tape DLT 80 autochanger, which should back everything up for a
BM> long time.  My tape machine also has "plenty" of holding disk space
BM> (something like 3 tapes worth/120 GB).  The servers have quite a bit
BM> of disk space on them, but the vast majority of it never changes (1
BM> server has maybe 100 GB total space that needs to be backed up, but 80
BM> GB of it is basically static).

BM> My configuration is therefore one which will waste a lot of tape if I
BM> let Amanda write a tape every night (and more importantly, it will run
BM> out of tapes and pester me in 3 weeks instead of 6 months...)

You want to run backups for 6 months with 20 tapes?
Tell us more about that, this is not clear to me ...

BM> Inspired by what I think that is telling me, I would like to do is
BM> something like this:

BM> - Run compressed backups to the holding disk
BM> - while the holding disk has > 1 tape worth of data (40 GB), write a
BM> tape.

BM> It isn't clear to me how to set it up to get this behavior.

If you really want to fill your tapes, you could set up vtapes (look
into the HOWTO-FILE-DRIVER in the docs) that are nearly as big as your
DLTs, and save the vtape-dirs with a second config to your DLTs.

I think you should just take the straight way, setup your changer,
configure a dumptype for your nearly-static stuff that has a long
dumpcycle and start going.

If you leave the tapes out, AMANDA will fill your holdingdisk, modify
the "reserve"-parameter to even get lev0-dumps in there.

-- 
best regards,
Stefan

Stefan G. Weichinger
mailto:[EMAIL PROTECTED]

Trying to figure out a rational setup

[Bret McKee]

Greetings:

I am setting up Amanda and am trying to get the configuration mostly
right the first time.  I like to understand what I'm doing, and not
ask really obvious questions.  To that end, I've read the chapter in
the backup book, read most of the man pages, most of the files in
doc/*, a bunch of entries in the Faq-O-Matic, and a couple hundred
posts in the archive for this mailing list.  Hopefully that gets me
started, but please accept my apologies if this still seems obvious.

I a couple servers and about a dozen PCs to backup.  Based upon my
current backup, it seems that only a few GB changes every day.  I have
a 20 tape DLT 80 autochanger, which should back everything up for a
long time.  My tape machine also has "plenty" of holding disk space
(something like 3 tapes worth/120 GB).  The servers have quite a bit
of disk space on them, but the vast majority of it never changes (1
server has maybe 100 GB total space that needs to be backed up, but 80
GB of it is basically static).

My configuration is therefore one which will waste a lot of tape if I
let Amanda write a tape every night (and more importantly, it will run
out of tapes and pester me in 3 weeks instead of 6 months...)

According to the FAQ-O-Matic:
-
This can be wasteful, specially if you have a small amount of data to
back up, but expensive large-capacity tapes. One possible approach is
to run amdump with tapes only, say once a week, to perform full
backups, and run it without tape on the other days, so that it
performs incremental backups and stores them in the holding disk. Once
or twice a week, you flush all backups in the holding disk to a single
tape. 
-

Inspired by what I think that is telling me, I would like to do is
something like this:

- Run compressed backups to the holding disk
- while the holding disk has > 1 tape worth of data (40 GB), write a
tape.

It isn't clear to me how to set it up to get this behavior.  It seems
like I might be able to get it by having two configurations, but then
it isn't clear of the tape tracking database features will work like I
would like (i.e. if one configuration writes to the disk only, and
another sometimes pushes the disk to a tape, will the database know
which tape the files from the disk only configuration are on?)

It almost seems like I should mess with tape changer scripts so that
requests for tapes are automatically failed if there is less than 1
tapes worth of information on the holding disk.  Once there is, it
could give Amanda 1 tape, and things might proceed the way I want.

Have I missed something obvious, or am I on the right track?

I appreciate any opinions, pointers to documentation, etc. that anyone
might provide...

Thanks,

Bret

Re: Advice for setup large environment?

[Joshua Baker-LePain]

On Tue, 24 Feb 2004 at 4:53pm, Michael D Schleif wrote
> 
> [A] dumpcycle
> runspercycle
> runtapes
> tapecycle
> &c.

dumpcycle depends on how much data you have.  runspercycle depends on how 
often you'll run amanda.  You said dailies could go on one tape, so 
runtapes is 1.  And tapecycle is simply the number of tapes you have.

> [B] How do I setup the autoloader?
> Yes, I know that this is a FAQ, and I will be RTFM this week.  Are
> there any specific things that I ought to know/read, based on real
> life experiences of list members in similar environments?

Read docs/TAPE.CHANGERS.  I use chg-zd-mtx with an Overland Library Pro.  
Setting it up was dead simply -- just follow the instructions in the 
chg-zd-mtx script itself.

> [C] Once we prove this in the pilot, then we will want to use a barcode
> reader, which may not be available during pilot.

This works just fine.

> [D] How do I setup regular cleaning tape usage via Amanda?

This is documented in the chg-zd-mtx script.  Personally, I just let the 
library handle it.

> [E] I am concerned about tape sequence.  We are starting with seven (7)
> tapes; but, will eventually move to twenty-seven (27).  If the
> order/sequence of tapes is not perfect, then Amanda will expect
> tapes in another magazine prior to using all seven in the current
> magazine.

When you're ready to expand tapecycle, start adding tapes only when amanda 
expects the first tape of the current tapecycle.  I.e. if you have 
Daily1-7, expand tapecycle the day after Daily7 gets used.

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University

Advice for setup large environment?

[Michael D Schleif]

I have been successfully using Amanda for nearly a year on several small
LAN's.

Now, I want to setup Amanda in a larger network in a phased approach.

First, notice that I will be using backup hardware such as I have setup
and configured for Arcserve and BackupExec; but, not yet with Amanda.

[1] Hardware

SuperDLT 1 tape drive
Overland Loader Xpress autoloader

[2] Tapes and magazines

Magazines hold ten (10) tapes
One (1) cleaning tape per magazine
Three (3) available magazines

[3] Startup configuration

Initially, one (1) magazine and seven (7) tapes

[4] Future larger configuration

Rotate all three (3) fully loaded magazines
Amanda can put daily backups on one (1) tape
Amanda must use all nine (9) tapes in a magazine *prior* to
   expecting another magazine


I am comfortable with installing Amanda in this environment, as well as
configurations such as holdingdisk, disklist, dumptypes, &c.

I am soliciting advice regarding areas such as these:

[A] dumpcycle
runspercycle
runtapes
tapecycle
&c.

[B] How do I setup the autoloader?
Yes, I know that this is a FAQ, and I will be RTFM this week.  Are
there any specific things that I ought to know/read, based on real
life experiences of list members in similar environments?

[C] Once we prove this in the pilot, then we will want to use a barcode
reader, which may not be available during pilot.

[D] How do I setup regular cleaning tape usage via Amanda?

[E] I am concerned about tape sequence.  We are starting with seven (7)
tapes; but, will eventually move to twenty-seven (27).  If the
order/sequence of tapes is not perfect, then Amanda will expect
tapes in another magazine prior to using all seven in the current
magazine.

What do you think?

-- 
Best Regards,

mds
mds resource
877.596.8237
-
Dare to fix things before they break . . .
-
Our capacity for understanding is inversely proportional to how much
we think we know.  The more I know, the more I know I don't know . . .
--


signature.asc
Description: Digital signature

Re: How to setup amanda with Hotbackup

[Paul Bijnens]

Yung Le wrote:

I already read this FAQ before I post this question. Since my database
spread to many mount points in system. If I follow the instruction from FAQ
to setup a differnet tar program to backup a few partitions. I need to
stop/start or enable/disable hotbackup mode many times when amanda call the
GNUTAR program to backup. Or when I backup a partition without database the
new GNUTAR still call to shutdown database...
You're not limited to the simple script in the FOM.
If you have many partitions, you shutdown the db when amanda issues
the first one to backup; keep the state in a file in e.g. /tmp/amanda.
When the last partition is backed up, then you can start the db again.
The script becomes a little more complicated, because you cannot be
sure which one will be first or last; but it still isn't rocket science.
Some quick, untested skeleton:

   if   # already stopped the db?   
test ! -f /tmp/amanda/db-parts  &&
# one of the db-partitions?
grep "^$CURDIR\$" /usr/local/etc/db-parts >/dev/null
   then
/usr/local/bin/dbshut   # shut it down
# keep list of partitions todo
cp /usr/local/etc/db-parts /tmp/amanda/db-parts
fi
[...]
/usr/local/bin/gtar "$@"  # now do the backup
[...]
if  # remove this part from the todo list
grep -v "^$CURDIR\$" /tmp/amanda/db-parts \
> /tmp/amanda/db-parts.$$  &&
   mv /tmp/amanda/db-parts.$$ /tmp/amanda/db-parts
`wc -l /tmp/amanda/db-parts` -eq 0 # last one?
 then
rm /tmp/amanda/db-parts
/usr/local/bin/dbstart
 fi
(Other tests need to be added, and output needs to be redirected
to the correct fd.)
Instead of shutting down, you could also switch the db to hot backup 
mode for the complete duration of the backup.
Or if you have snapshots, then you could shutdown the db (or switch to
hotbackup), snapshot all the partitions, and startup again, and then
let amanda do her thing. In this case, you only need to know when
to start the snapshot.

Because taking a snapshot takes only minutes at most, you could do
this outside the normal amanda backup with a separate cron entry that
runs a safe time before amanda starts.
(That's how I do it currently.)
A complete different approach is to set up rsh/ssh capabilities from the
amanda server the the database server and run the shutdown/startup
scripts as remote commands from within the little shell script that
start amdump.  Now the downtime window is longer, because the database
will be down including the estimate phase, and has to wait for the last
amanda clien too.  (Maybe not too bad if it is only hotbackup mode.)
--
Paul Bijnens, XplanationTel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax  +32 16 397.512
http://www.xplanation.com/  email:  [EMAIL PROTECTED]
***
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...*
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out  *
***