Re: ***UNCHECKED*** exact-match

[Jon LaBadie]

On Fri, Jun 07, 2019 at 11:04:48AM +0100, Nuno Dias wrote:
>  Hi,
> 
>  I'm trying to use amanda to backup only one dir from a list of dirs
> that are in disklist file
> 
> I run amdump like this
> 
> $ /usr/sbin/amdump  -o reserve=0 --no-taper MACHINE
> ^/dir/subdir/name/2019$
> 
> and with ps I can see the amdump running
> 
> /usr/bin/perl /usr/sbin/amdump  -o reserve=0 --no-taper MACHINE
> ^/dir/subdir/name/2019$
> 
> The problem is instead of only one dir I have two dirs in the backup
> 
> MACHINE:/dir/subdir/name/2019 20190606153859 0   486g dumping
> (18g done (3.74%)) (15:39:26)
> 
> MACHINE:/dir/subdir/name/2019/another 20190606153859 1   244g
> wait for dumping
> 
> 
>  From the amanda-match man page
> 
>   /usr$
>   Will match the disks /usr but not /usr/local.
> 
>  So, this is a bug or I'm doing something wrong?
> 
Not wrong, just a misconception.
The directory you give is a "starting" directory.  amdump will
backup everything under that directory unless it could cross
a mount point (file system) or if something is excluded.

So, assuming you have a DLE for ".../2019" and that your "another"
directory is the only problem, add a directive to the DLE to
exclude another.  If you don't have a DLE for 2019, PERHAPS
the -o option to amdump can be used to add an exclude directive.
I don't know if it can or cannot.

Jon
>  I'm using amanda-3.5.1-16.fc29.x86_64
> 
> thanks,
> Nuno
> -- 
> Nuno Dias 
> LIP


>>> End of included message <<<

-- 
Jon H. LaBadie j...@jgcomp.com
 11226 South Shore Rd.  (703) 787-0688 (H)
 Reston, VA  20190  (703) 935-6720 (C)

bug bites?

[Gene Heskett]

Greetings all;

Last nights backup should have been a typical run, which since I 
restarted it a little less than 2 weeks back, should have been in the 
17GB range, (and has been after the initial 4 vtapes, 180GB blast it 
took an extra flush run to clear the dumps assignment) except for the 
planner and something else apparently disagreeing on what a level is. 
The planner promoted /home/gene to a level 5 run, but the backup when it 
was listed actually was a level 0. So what should have been another 16G 
or so was actually still over 60GB, 4 times what it should have been.

Currently its got 10 days to do a cycle, and should stay at a given level 
for 2 days, so a level 5 should be a legal promotion.  Do a level 0. 
then 2 level 1's =3 day, then 2 level 2's=5 days, then 2 level 3's=7 
days, then 2 level 4's = 9 days, but a level 5=10 days and a reset to 0.

This has happened many times before, and while its not a showstopper 
since its using however much of a 2T drive it needs writing the size of 
the vtape, which it can use 2 of at 45GB each. So theres nothing left in 
the dumps assignment this morning.

But can we stop the planner from lying to us? Preferably by making it do 
what the planner is telling it to do? Since I've doubled the size of the 
drive, I am going to gradually reduce the cycle until the drive is 
around 75% used. But 2 weeks is not long enough to settle, and its sure 
not long enough to settle when a bug bites it like that. According to my 
math, it should have done a level 5 once before running out of days.  
Some of its math is base0 and some is base 1?

So what I've done this morning is reduce the cycle by one day. And fixed 
the exclude file for that dumptype, there were 2 subdirs getting backed 
up by 2 different DLE's. A df says that drive is at 22%. We'll look 
again in a few days.

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 

exact-match

[Nuno Dias]

 Hi,

 I'm trying to use amanda to backup only one dir from a list of dirs
that are in disklist file

I run amdump like this

$ /usr/sbin/amdump  -o reserve=0 --no-taper MACHINE
^/dir/subdir/name/2019$

and with ps I can see the amdump running

/usr/bin/perl /usr/sbin/amdump  -o reserve=0 --no-taper MACHINE
^/dir/subdir/name/2019$

The problem is instead of only one dir I have two dirs in the backup

MACHINE:/dir/subdir/name/2019 20190606153859 0   486g dumping
(18g done (3.74%)) (15:39:26)

MACHINE:/dir/subdir/name/2019/another   20190606153859 1   244g
wait for dumping


 From the amanda-match man page

  /usr$
  Will match the disks /usr but not /usr/local.

 So, this is a bug or I'm doing something wrong?

 I'm using amanda-3.5.1-16.fc29.x86_64

thanks,
Nuno
-- 
Nuno Dias 
LIP


smime.p7s
Description: S/MIME cryptographic signature

amsamba with SMB2

[Stefan G. Weichinger]

I have to backup a share that is only reachable via SMB2:

server ~ # smbclient -L 192.168.160.1 -U user%pw
protocol negotiation failed: NT_STATUS_CONNECTION_RESET

server ~ # smbclient -L 192.168.160.1 -U user%pw -m SMB2
Domain=[ANLAGEN-SERVER] OS=[] Server=[]

[..]

ADMIN$  Disk  Remoteverwaltung
C$  Disk  Standardfreigabe
IPC$IPC   Remote-IPC

[..]

Can amsamba use that "-m SMB2" somehow?

"man amsamba" doesn't tell ...