Menu
▾
▴
backuppc-users
|
From: Hervé R. <he...@pe...> - 2008-03-26 15:42:17
Attachments:
herve.vcf
|
Hello all, My backup server with backuppc don't backup anymore, backup hdd is full :-/ So I checked a little bit the backuped data to see what is taking so much space. Let me explain my configuration: I have 12Gb of data to backup on my server. I configured 1 full backup every Sunday at 3AM and Incr every other day of the week at 3AM too I configured backuppc to keep 2 full backup and 12 Incr. Then after a while my backup hdd was filled .... I check the size of the Cpool dir and it takes about 45GB on the drive and pc dir is about 26Gb... There are no big file transfer between two Incr backup which could have filled the disk and only 12GB of data to backup. Did I missed something? I don't understand why a 80GB is not sufficient. Many thanks for your help. Hervé |
|
From: Les M. <les...@gm...> - 2008-03-26 16:08:05
|
Hervé Richard wrote:
>
> My backup server with backuppc don't backup anymore, backup hdd is full :-/
> So I checked a little bit the backuped data to see what is taking so
> much space.
>
> Let me explain my configuration:
>
> I have 12Gb of data to backup on my server.
> I configured 1 full backup every Sunday at 3AM and Incr every other day
> of the week at 3AM too
> I configured backuppc to keep 2 full backup and 12 Incr.
>
> Then after a while my backup hdd was filled ....
>
> I check the size of the Cpool dir and it takes about 45GB on the drive
> and pc dir is about 26Gb...
This doesn't sound right if you only have one target. Does the server
and the backuppc service run continuously? Maybe it isn't running the
backuppc_nightly job that removes the unneeded pool/cpool copies.
> There are no big file transfer between two Incr backup which could have
> filled the disk and only 12GB of data to backup.
If you are using rsync, any large files that grow or have small changes
(databases, logs, etc.) will only transfer the difference but the server
side will reconstruct the full size file and store it as a separate copy.
> Did I missed something?
If your target is a 64-bit linux system, be sure you aren't backing up
/var/log/lastlog - or at least check its size. On some versions this is
a huge sparse file that copies generally don't handle very well.
--
Les Mikesell
le...@fu...
|
|
From: Les M. <le...@fu...> - 2008-03-26 19:33:31
|
Hervé Richard wrote:
>
>>
>> This doesn't sound right if you only have one target. Does the server
>> and the backuppc service run continuously? Maybe it isn't running the
>> backuppc_nightly job that removes the unneeded pool/cpool copies.
> Yes in fact the server and backuppc run continuously and the server
> backup files stored on itself.
> I took care of not backing up backuppc files as mentioned in the
> backuppc faq.
>
> For backuppc_nightly job:
> Maybe :-/
> I don't know how to configure/force backuppc_nightly job to removes
> unneeded pool/cpool :-/
> Maybe it's automatic... check every age of every backuped files and
> delete everyone older than configured in:
It is supposed to be automatic and it doesn't pay attention to ages. It
should delete anything in the pools with only one link - meaning that
all references in the pc directories had already been deleted.
If you cd to the cpool directory and 'find . -links 1' you shouldn't see
very many files.
> I have this message in every log:
>
> Cpool nightly clean removed 0 files of size 0.00GB
It doesn't make sense for the cpool dir to be bigger than the pc
directory if you only have one server, since all files that exist in
cpool should have links in the pc dir(s) too. However, du has some
quirks in the way it adds up the space, trying not to count linked files
multiple times. Also, the space you reported for cpool and pc should
really be the same space since all the files should be hardlinks. And
in any case it shouldn't fill the 80 gigs you said you had. We must be
missing something. Is everything on the same filesytem?
--
Les Mikesell
les...@gm...
|
|
From: Hervé R. <he...@pe...> - 2008-03-27 10:26:51
Attachments:
herve.vcf
|
Hello Les,
Les Mikesell wrote:
> Hervé Richard wrote:
>>
>>>
>>> This doesn't sound right if you only have one target. Does the
>>> server and the backuppc service run continuously? Maybe it isn't
>>> running the backuppc_nightly job that removes the unneeded
>>> pool/cpool copies.
>> Yes in fact the server and backuppc run continuously and the server
>> backup files stored on itself.
>> I took care of not backing up backuppc files as mentioned in the
>> backuppc faq.
By the way I have a rotation between two hard drive every 2 weeks.
This done manually.
Localhost config.pl is stored on each hard drive including the backuppc
dir as below:
backuppc/
cpool/
log/
pc/
pool/
trash/
I think there is bad influence on the way backuppc behave.
>>
>> For backuppc_nightly job:
>> Maybe :-/
>> I don't know how to configure/force backuppc_nightly job to removes
>> unneeded pool/cpool :-/
>> Maybe it's automatic... check every age of every backuped files and
>> delete everyone older than configured in:
>
> It is supposed to be automatic and it doesn't pay attention to ages.
> It should delete anything in the pools with only one link - meaning
> that all references in the pc directories had already been deleted.
> If you cd to the cpool directory and 'find . -links 1' you shouldn't
> see very many files.
>
When I execute ' find . -links 1 ' in the cpool directory I see a lot of
files.
In my configuration, files are stored on a RAID volume mounted on a
mount point in home directory.
In the Cpool directory I got directories from : 0 to f and same subdirs
from 0 to f...
For example:
ls cpool -> 0 1 2 3 4 5 6 7 8 9 a b c d e f
and ls cpool/0/ -> same thing as above
ls cpool/0/0/ -> same thing
ls cpool/0/0/0/ -> gives me files with 34 hexa digit.
When I check (with ls -la) files date some are more than 1 year older
than the other (eg: 2006 and 2007)
>
>> I have this message in every log:
>>
>> Cpool nightly clean removed 0 files of size 0.00GB
>
> It doesn't make sense for the cpool dir to be bigger than the pc
> directory if you only have one server, since all files that exist in
> cpool should have links in the pc dir(s) too. However, du has some
> quirks in the way it adds up the space, trying not to count linked
> files multiple times. Also, the space you reported for cpool and pc
> should really be the same space since all the files should be
> hardlinks. And in any case it shouldn't fill the 80 gigs you said you
> had. We must be missing something. Is everything on the same filesytem?
>
Of cource I think it too...
BackupPC main directory and core system is on the OS file system, the
data to be backuped up are on an another file system mounted on home/ as
I said above and the backuped data are on another filesystem on a hard
drive. Every file system are Ext3.
I didn't used symlink to write path to the backuped files in my config.pl.
Here is an overview of the system architecture:
_____
| |
|RAID|
Main FS |_____|
___ /
| | / /home/e-smith/files
|___| /
- OS /
__
- Mounting point <---/---------------------/mnt/hdd1---|__| Backup media
(HDD 80GB) ->backuppc/
- BackupPC core /opt/backuppc/
cpool/
log/
pc/
pool/
trash/
Cheers
Hervé
|
|
From: Les M. <le...@fu...> - 2008-03-27 18:39:21
|
Hervé Richard wrote:
>
> By the way I have a rotation between two hard drive every 2 weeks.
> This done manually.
> Localhost config.pl is stored on each hard drive including the backuppc
> dir as below:
> backuppc/
> cpool/
> log/
> pc/
> pool/
> trash/
>
> I think there is bad influence on the way backuppc behave.
That shouldn't cause any particular problem as long as the
backuppc_nightly jobs run.
>>
> When I execute ' find . -links 1 ' in the cpool directory I see a lot of
> files.
> In my configuration, files are stored on a RAID volume mounted on a
> mount point in home directory.
> In the Cpool directory I got directories from : 0 to f and same subdirs
> from 0 to f...
> For example:
> ls cpool -> 0 1 2 3 4 5 6 7 8 9 a b c d e f
> and ls cpool/0/ -> same thing as above
> ls cpool/0/0/ -> same thing
> ls cpool/0/0/0/ -> gives me files with 34 hexa digit.
>
> When I check (with ls -la) files date some are more than 1 year older
> than the other (eg: 2006 and 2007)
The age doesn't matter - files will remain as long as some copy of that
same content is still in a backup run under the pc directory. However,
when the last copy under the pc directory expires and is removed, the
link count on the cpool instance will go down to 1 and the next run of
backuppc_nightly should remove it. I think I would try stopping the
backuppc service and running (as the backuppc user)
BackupPC_Nightly 0 255
to see if it releases some space or has some error.
--
Les Mikesell
le...@fu...
|
|
From: Hervé R. <he...@pe...> - 2008-03-31 07:25:57
Attachments:
herve.vcf
|
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=UTF-8" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Les Mikesell wrote:
<blockquote cite="mid:47E...@fu..." type="cite">Hervé
Richard wrote: <br>
<blockquote type="cite"><br>
By the way I have a rotation between two hard drive every 2 weeks. <br>
This done manually. <br>
Localhost config.pl is stored on each hard drive including the backuppc
dir as below: <br>
backuppc/ <br>
cpool/ <br>
log/ <br>
pc/ <br>
pool/ <br>
trash/ <br>
<br>
I think there is bad influence on the way backuppc behave. <br>
</blockquote>
<br>
That shouldn't cause any particular problem as long as the
backuppc_nightly jobs run. <br>
<br>
<blockquote type="cite">
<blockquote type="cite"><br>
</blockquote>
When I execute ' find . -links 1 ' in the cpool directory I see a lot
of files. <br>
In my configuration, files are stored on a RAID volume mounted on a
mount point in home directory. <br>
In the Cpool directory I got directories from : 0 to f and same subdirs
from 0 to f... <br>
For example: <br>
ls cpool -> 0 1 2 3 4 5 6 7 8 9 a b c d e f <br>
and ls cpool/0/ -> same thing as above <br>
ls cpool/0/0/ -> same thing <br>
ls cpool/0/0/0/ -> gives me files with 34 hexa digit. <br>
<br>
When I check (with ls -la) files date some are more than 1 year older
than the other (eg: 2006 and 2007) <br>
</blockquote>
<br>
The age doesn't matter - files will remain as long as some copy of that
same content is still in a backup run under the pc directory. However,
when the last copy under the pc directory expires and is removed, the
link count on the cpool instance will go down to 1 and the next run of
backuppc_nightly should remove it. I think I would try stopping the
backuppc service and running (as the backuppc user) <br>
BackupPC_Nightly 0 255 <br>
to see if it releases some space or has some error. <br>
<br>
</blockquote>
Les,<br>
<br>
Just one thing I forgot :-/<br>
There is a symlink between /opt/backuppc/files/pc/localhost (os file
system) -> /mnt/hdd1/backuppc/pc/localhost (backup hdd file system)<br>
<br>
Maybe the problem comes from this symlink, what do you think about?<br>
Maybe I should have make a hardlink?<br>
Or maybe I shouldn't have copied /localhost subdirs to backup hdds <br>
<br>
Here is the backup log for this WE, Incr + Full:<br>
<br>
<p>Contents of file <tt>/opt/backuppc/files/pc/localhost/LOG</tt>,
modified 2008-03-31 03:30:10 </p>
<pre>2008-03-01 03:30:02 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-01 03:30:04 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-01 03:30:08 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-01 03:30:13 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/proj/files/
2008-03-01 03:30:14 incr backup started back to 2008-02-26 02:30:02 for directory /etc/
2008-03-01 03:30:20 incr backup 318 complete, 3379 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-01 03:30:20 removing incr backup 304
2008-03-02 03:30:02 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-02 03:30:04 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-02 03:30:06 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-02 03:30:11 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/proj/files/
2008-03-02 03:30:12 incr backup started back to 2008-02-26 02:30:02 for directory /etc/
2008-03-02 03:30:15 incr backup 319 complete, 3379 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-02 03:30:15 removing incr backup 305
2008-03-03 03:30:02 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-03 03:30:04 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-03 03:30:06 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-03 03:30:12 incr backup started back to 2008-02-26 02:30:02 for directory /home/e-smith/files/ibays/proj/files/
2008-03-03 03:30:13 incr backup started back to 2008-02-26 02:30:02 for directory /etc/
2008-03-03 03:30:16 incr backup 320 complete, 3379 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-03 03:30:16 removing incr backup 306
2008-03-03 03:30:24 removing full backup 300
2008-03-18 03:30:14 full backup started for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-18 03:35:32 full backup started for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-18 03:36:55 full backup started for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-18 03:51:53 full backup started for directory /home/e-smith/files/ibays/proj/files/
2008-03-18 03:57:41 full backup started for directory /etc/
2008-03-18 03:58:20 full backup 321 complete, 21267 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-19 03:30:19 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-19 03:30:24 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-19 03:30:28 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-19 03:30:31 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/proj/files/
2008-03-19 03:30:36 incr backup started back to 2008-03-18 02:30:13 for directory /etc/
2008-03-19 03:30:45 incr backup 322 complete, 3352 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-19 03:30:45 removing incr backup 308
2008-03-20 03:30:16 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-20 03:30:19 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-20 03:30:23 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-20 03:30:25 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/proj/files/
2008-03-20 03:30:29 incr backup started back to 2008-03-18 02:30:13 for directory /etc/
2008-03-20 03:30:36 incr backup 323 complete, 3368 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-20 03:30:36 removing incr backup 309
2008-03-21 03:30:19 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-21 03:30:25 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-21 03:30:31 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-21 03:31:23 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/proj/files/
2008-03-21 03:31:29 incr backup started back to 2008-03-18 02:30:13 for directory /etc/
2008-03-21 03:31:38 incr backup 324 complete, 3453 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-21 03:31:38 removing incr backup 310
2008-03-22 03:30:03 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-22 03:30:05 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-22 03:30:08 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-22 03:30:49 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/proj/files/
2008-03-22 03:30:51 incr backup started back to 2008-03-18 02:30:13 for directory /etc/
2008-03-22 03:30:56 incr backup 325 complete, 3474 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-22 03:30:56 removing incr backup 311
2008-03-23 03:30:02 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-23 03:30:04 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-23 03:30:07 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-23 03:30:48 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/proj/files/
2008-03-23 03:30:50 incr backup started back to 2008-03-18 02:30:13 for directory /etc/
2008-03-23 03:30:55 incr backup 326 complete, 3475 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-23 03:30:55 removing incr backup 312
2008-03-24 03:30:02 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-24 03:30:04 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-24 03:30:08 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-24 03:30:51 incr backup started back to 2008-03-18 02:30:13 for directory /home/e-smith/files/ibays/proj/files/
2008-03-24 03:30:54 incr backup started back to 2008-03-18 02:30:13 for directory /etc/
2008-03-24 03:30:59 incr backup 327 complete, 3475 files, bytes, xferErrs ( bad files, bad shares, 0 other)
2008-03-24 03:30:59 removing incr backup 313
2008-03-24 03:31:09 removing full backup 307
2008-03-25 03:30:02 full backup started for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-25 03:35:13 full backup started for directory /home/e-smith/files/ibays/admin-sec/files/
2008-03-25 03:36:38 full backup started for directory /home/e-smith/files/ibays/img/files/Current/
2008-03-25 03:51:52 Got fatal error during xfer (Tar exited with error 512 () status)
2008-03-25 03:51:57 Backup aborted (Tar exited with error 512 () status)
2008-03-25 03:51:57 Saved partial dump 328
2008-03-26 03:30:19 full backup started for directory /home/e-smith/files/ibays/admin-bna/files/
2008-03-26 03:30:21 Got fatal error during xfer (Tar exited with error 512 () status)
2008-03-26 03:30:26 Backup aborted (Tar exited with error 512 () status)
2008-03-26 03:30:26 Rename /opt/backuppc/files/pc/<a
href="https://bnabbot.dyndns.org/backup/cgi-bin/BackupPC_Admin?host=localhost">localhost</a>/new -> /opt/backuppc/files/pc/<a
href="https://bnabbot.dyndns.org/backup/cgi-bin/BackupPC_Admin?host=localhost">localhost</a>/0 failed
2008-03-26 03:30
</pre>
<br>
Thanks for your help cause I'm really stucked here :-(<br>
<br>
Hervé<br>
<br>
</body>
</html>
|
|
From: Les M. <le...@fu...> - 2008-04-01 04:14:46
|
Hervé Richard wrote:
>
> Just one thing I forgot :-/
> There is a symlink between /opt/backuppc/files/pc/localhost (os file system) ->
> /mnt/hdd1/backuppc/pc/localhost (backup hdd file system)
>
> Maybe the problem comes from this symlink, what do you think about?
> Maybe I should have make a hardlink?
> Or maybe I shouldn't have copied /localhost subdirs to backup hdds
The cpool and pc directories must be on the same filesystem so if you
are going to symlink to a different disk the symlink must be to the
directory above cpool and pc.
--
Les Mikesell
les...@gm...
|
|
From: Hervé R. <he...@pe...> - 2008-04-01 09:31:29
Attachments:
herve.vcf
|
Les Mikesell wrote: > Hervé Richard wrote: >> >> Just one thing I forgot :-/ >> There is a symlink between /opt/backuppc/files/pc/localhost (os file >> system) -> /mnt/hdd1/backuppc/pc/localhost (backup hdd file system) >> >> Maybe the problem comes from this symlink, what do you think about? >> Maybe I should have make a hardlink? >> Or maybe I shouldn't have copied /localhost subdirs to backup hdds > > The cpool and pc directories must be on the same filesystem so if you > are going to symlink to a different disk the symlink must be to the > directory above cpool and pc. > If I understand what you say, on my backup HDD I need conf, cpool, log, pc, pool, save and trash to be there ? Is for that reason that's not working :-/ I wish I could do a rotation between 2 hdd every 2 weeks.... Maybe if I do the same thing on the other HDD: copy all needed directory and mount the hdd partition under /files/ Many thanks for your help :-) I'll try this and let you know Cheers Hervé P.S: As you guessed I supposed, I'm quite new as using backuppc :-/ |
|
From: Hervé R. <he...@pe...> - 2008-04-01 16:34:47
Attachments:
herve.vcf
|
Hervé Richard wrote: > Les Mikesell wrote: >> Hervé Richard wrote: >>> >>> Just one thing I forgot :-/ >>> There is a symlink between /opt/backuppc/files/pc/localhost (os file >>> system) -> /mnt/hdd1/backuppc/pc/localhost (backup hdd file system) >>> >>> Maybe the problem comes from this symlink, what do you think about? >>> Maybe I should have make a hardlink? >>> Or maybe I shouldn't have copied /localhost subdirs to backup hdds >> >> The cpool and pc directories must be on the same filesystem so if you >> are going to symlink to a different disk the symlink must be to the >> directory above cpool and pc. >> > If I understand what you say, on my backup HDD I need conf, cpool, > log, pc, pool, save and trash to be there ? > Is for that reason that's not working :-/ > I wish I could do a rotation between 2 hdd every 2 weeks.... > Maybe if I do the same thing on the other HDD: copy all needed > directory and mount the hdd partition under /files/ > > > Many thanks for your help :-) > I'll try this and let you know > > Cheers > > Hervé > > P.S: As you guessed I supposed, I'm quite new as using backuppc :-/ Ok I've done some cleaning it was a little bit a mess :-/ BackuPC behave a little better now but I still have old backup that BackupPC_nightly doesn't want to remove :-/ I think I have to choice: 1°) Comment the line for the host I backup files and let backuppc propose me tout delete its backup files 2°) Use the BackupPC_deletebackup I saw on the wiki: http://backuppc.wiki.sourceforge.net/How+to+delete+backups What do you think about? Thanks, Cheers Hervé > > > > __________ NOD32 2991 (20080401) Information __________ > > This message was checked by NOD32 antivirus system. > http://www.eset.com > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace > > > __________ NOD32 2991 (20080401) Information __________ > > This message was checked by NOD32 antivirus system. > http://www.eset.com > > ------------------------------------------------------------------------ > > _______________________________________________ > BackupPC-users mailing list > Bac...@li... > List: https://lists.sourceforge.net/lists/listinfo/backuppc-users > Wiki: http://backuppc.wiki.sourceforge.net > Project: http://backuppc.sourceforge.net/ > > > > __________ NOD32 2991 (20080401) Information __________ > > This message was checked by NOD32 antivirus system. > http://www.eset.com > |
|
From: Les M. <le...@fu...> - 2008-04-01 16:12:46
|
Hervé Richard wrote: > >>> >> If I understand what you say, on my backup HDD I need conf, cpool, >> log, pc, pool, save and trash to be there ? >> Is for that reason that's not working :-/ >> I wish I could do a rotation between 2 hdd every 2 weeks.... >> Maybe if I do the same thing on the other HDD: copy all needed >> directory and mount the hdd partition under /files/ >> >> P.S: As you guessed I supposed, I'm quite new as using backuppc :-/ > Ok I've done some cleaning it was a little bit a mess :-/ > BackuPC behave a little better now but I still have old backup that > BackupPC_nightly doesn't want to remove :-/ > I think I have to choice: > 1°) Comment the line for the host I backup files and let backuppc > propose me tout delete its backup files > 2°) Use the BackupPC_deletebackup I saw on the wiki: > http://backuppc.wiki.sourceforge.net/How+to+delete+backups > > What do you think about? Since you have 2 disks, you could just start from scratch on one of them and wait until it has a complete set before you swap again. Another approach is to always run on an internal drive, but periodically image-copy to a matching external disk that you rotate offsite. I do it that way myself but recently switched from firewire externals to a swappable SATA enclosure. I do the mirroring with software raid, but it would work as well to unmount the partition temporarily and dd it. -- Les Mikesell le...@fu... |
|
From: Hervé R. <he...@pe...> - 2008-04-02 07:41:54
Attachments:
herve.vcf
|
Les Mikesell wrote: > Hervé Richard wrote: >> >>> >>> If I understand what you say, on my backup HDD I need conf, cpool, >>> log, pc, pool, save and trash to be there ? >>> Is for that reason that's not working :-/ >>> I wish I could do a rotation between 2 hdd every 2 weeks.... >>> Maybe if I do the same thing on the other HDD: copy all needed >>> directory and mount the hdd partition under /files/ >>> > >>> P.S: As you guessed I supposed, I'm quite new as using backuppc :-/ >> Ok I've done some cleaning it was a little bit a mess :-/ >> BackuPC behave a little better now but I still have old backup that >> BackupPC_nightly doesn't want to remove :-/ >> I think I have to choice: >> 1°) Comment the line for the host I backup files and let backuppc >> propose me tout delete its backup files >> 2°) Use the BackupPC_deletebackup I saw on the wiki: >> http://backuppc.wiki.sourceforge.net/How+to+delete+backups >> >> What do you think about? > > Since you have 2 disks, you could just start from scratch on one of > them and wait until it has a complete set before you swap again. > Another approach is to always run on an internal drive, but > periodically image-copy to a matching external disk that you rotate > offsite. I do it that way myself but recently switched from firewire > externals to a swappable SATA enclosure. I do the mirroring with > software raid, but it would work as well to unmount the partition > temporarily and dd it. > Hello Les, By "start from scratch" you mean comment the backuped host in the "hosts" file, let BackupPC running and deleting all backup made? Or what I have to delete manually? For now backup hdd is connected as JBOD to the main RAID controller. I made a script which is executed via Webmin to stop/start backuppc, umount/mount the hdd, remove it from RAID controller/rescan the controller for new inserted disk. Nice tips you told here :-) With a swappable SATA enclosure I could dd one backup of 2 weeks on one hdd and the backup of the following 2 weeks on the other one and do the rotation... I'll think about it. Thanks, Cheers, Hervé |
|
From: Les M. <le...@fu...> - 2008-04-02 15:46:06
|
Hervé Richard wrote: >> >> Since you have 2 disks, you could just start from scratch on one of >> them and wait until it has a complete set before you swap again. >> Another approach is to always run on an internal drive, but >> periodically image-copy to a matching external disk that you rotate >> offsite. I do it that way myself but recently switched from firewire >> externals to a swappable SATA enclosure. I do the mirroring with >> software raid, but it would work as well to unmount the partition >> temporarily and dd it. > > By "start from scratch" you mean comment the backuped host in the > "hosts" file, let BackupPC running and deleting all backup made? > Or what I have to delete manually? You can go as far as making a new file system on one of the drives and copying over everything except the contents of cpool and the backups under the pc directories. Or, if backuppc_Nightly is working now you might not need to start over. > For now backup hdd is connected as JBOD to the main RAID controller. I > made a script which is executed via Webmin to stop/start backuppc, > umount/mount the hdd, remove it from RAID controller/rescan the > controller for new inserted disk. > > Nice tips you told here :-) > With a swappable SATA enclosure I could dd one backup of 2 weeks on one > hdd and the backup of the following 2 weeks on the other one and do the > rotation... I'll think about it. My current system is using one of these with 750 gig drives: http://www.startech.com/Product/ItemDetail.aspx?productid=HSB320SATBK&c=US and I use mdadm to fail/remove/add the disks that rotate, momentarily stopping backuppc and unmounting to get a clean filesystem. On a smaller system I might consider the laptop-form external USB drives that you can get in sizes to 250 gigs now and just unmount and dd the internal partition to it. -- Les Mikesell les...@gm... |
|
From: Hervé R. <he...@pe...> - 2008-04-02 16:05:10
Attachments:
herve.vcf
|
Les Mikesell wrote: > Hervé Richard wrote: >>> >>> Since you have 2 disks, you could just start from scratch on one of >>> them and wait until it has a complete set before you swap again. >>> Another approach is to always run on an internal drive, but >>> periodically image-copy to a matching external disk that you rotate >>> offsite. I do it that way myself but recently switched from >>> firewire externals to a swappable SATA enclosure. I do the >>> mirroring with software raid, but it would work as well to unmount >>> the partition temporarily and dd it. >> >> By "start from scratch" you mean comment the backuped host in the >> "hosts" file, let BackupPC running and deleting all backup made? >> Or what I have to delete manually? > > You can go as far as making a new file system on one of the drives and > copying over everything except the contents of cpool and the backups > under the pc directories. Or, if backuppc_Nightly is working now you > might not need to start over. I made some modif in the main config.pl file and reloaded the config file then I saw some cleanup among the old backup (maybe it's a coincidence). The only strange thing is that I have an old full backup from 32 days remaining among 3 full and I only keep 2 full.... :-/ Seems that BackupPC didn't delete this one. I'll wait till the nightly Incr goes on and see tomorrow if this old backup still there... If it doesn't work I'll delete files under cpool and backup under pc directories (if I understood well, copying over and omitting content of cpool and files under pc dir is same as said) as you said and start fresh backup... > >> For now backup hdd is connected as JBOD to the main RAID controller. >> I made a script which is executed via Webmin to stop/start backuppc, >> umount/mount the hdd, remove it from RAID controller/rescan the >> controller for new inserted disk. >> >> Nice tips you told here :-) >> With a swappable SATA enclosure I could dd one backup of 2 weeks on >> one hdd and the backup of the following 2 weeks on the other one and >> do the rotation... I'll think about it. > > My current system is using one of these with 750 gig drives: > http://www.startech.com/Product/ItemDetail.aspx?productid=HSB320SATBK&c=US > > and I use mdadm to fail/remove/add the disks that rotate, momentarily > stopping backuppc and unmounting to get a clean filesystem. > > On a smaller system I might consider the laptop-form external USB > drives that you can get in sizes to 250 gigs now and just unmount and > dd the internal partition to it. > Thanks for the tips and help :-) I'll let you know how it goes. Cheers, Hervé |
|
From: Hervé R. <he...@pe...> - 2008-04-04 10:16:28
Attachments:
herve.vcf
|
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=UTF-8" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Hervé Richard wrote:
<blockquote cite="mid:47F...@pe..." type="cite"><br>
<blockquote type="cite"><br>
<blockquote type="cite">For now backup hdd is connected as JBOD to
the main RAID controller. I made a script which is executed via Webmin
to stop/start backuppc, umount/mount the hdd, remove it from RAID
controller/rescan the controller for new inserted disk.
<br>
<br>
Nice tips you told here :-)
<br>
With a swappable SATA enclosure I could dd one backup of 2 weeks on one
hdd and the backup of the following 2 weeks on the other one and do the
rotation... I'll think about it.
<br>
</blockquote>
<br>
My current system is using one of these with 750 gig drives:
<br>
<a class="moz-txt-link-freetext" href="http://www.startech.com/Product/ItemDetail.aspx?productid=HSB320SATBK&c=US">http://www.startech.com/Product/ItemDetail.aspx?productid=HSB320SATBK&c=US</a>
<br>
and I use mdadm to fail/remove/add the disks that rotate, momentarily
stopping backuppc and unmounting to get a clean filesystem.
<br>
<br>
On a smaller system I might consider the laptop-form external USB
drives that you can get in sizes to 250 gigs now and just unmount and
dd the internal partition to it.
<br>
<br>
</blockquote>
Thanks for the tips and help :-)
<br>
<br>
I'll let you know how it goes.
<br>
<br>
Cheers,
<br>
<br>
Hervé
<br>
</blockquote>
Hello Les,<br>
<br>
Now it seems to work well.<br>
BackuPC_nightly has done is job and for now old backup are gone :-)<br>
New Full nad Incr backup are going well :-)<br>
<br>
Many thanks for your help :-)<br>
<br>
Cheers<br>
<br>
Hervé<br>
<blockquote cite="mid:47F...@pe..." type="cite"><br>
<br>
<br>
<br>
__________ NOD32 2995 (20080402) Information __________
<br>
<br>
This message was checked by NOD32 antivirus system.
<br>
<a class="moz-txt-link-freetext" href="http://www.eset.com">http://www.eset.com</a>
<br>
<pre wrap="">-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
<a class="moz-txt-link-freetext" href="http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace">http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace</a>
__________ NOD32 2995 (20080402) Information __________
This message was checked by NOD32 antivirus system.
<a class="moz-txt-link-freetext" href="http://www.eset.com">http://www.eset.com</a>
</pre>
<pre wrap="">
<hr size="4" width="90%">
_______________________________________________
BackupPC-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Bac...@li...">Bac...@li...</a>
List: <a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/backuppc-users">https://lists.sourceforge.net/lists/listinfo/backuppc-users</a>
Wiki: <a class="moz-txt-link-freetext" href="http://backuppc.wiki.sourceforge.net">http://backuppc.wiki.sourceforge.net</a>
Project: <a class="moz-txt-link-freetext" href="http://backuppc.sourceforge.net/">http://backuppc.sourceforge.net/</a>
__________ NOD32 2995 (20080402) Information __________
This message was checked by NOD32 antivirus system.
<a class="moz-txt-link-freetext" href="http://www.eset.com">http://www.eset.com</a>
</pre>
</blockquote>
<br>
</body>
</html>
|
|
From: Hervé R. <he...@pe...> - 2008-03-28 15:07:49
Attachments:
herve.vcf
|
Les Mikesell wrote: > Hervé Richard wrote: >> >> By the way I have a rotation between two hard drive every 2 weeks. >> This done manually. >> Localhost config.pl is stored on each hard drive including the >> backuppc dir as below: >> backuppc/ >> cpool/ >> log/ >> pc/ >> pool/ >> trash/ >> >> I think there is bad influence on the way backuppc behave. > > That shouldn't cause any particular problem as long as the > backuppc_nightly jobs run. > >>> >> When I execute ' find . -links 1 ' in the cpool directory I see a lot >> of files. >> In my configuration, files are stored on a RAID volume mounted on a >> mount point in home directory. >> In the Cpool directory I got directories from : 0 to f and same >> subdirs from 0 to f... >> For example: >> ls cpool -> 0 1 2 3 4 5 6 7 8 9 a b c d e f >> and ls cpool/0/ -> same thing as above >> ls cpool/0/0/ -> same thing >> ls cpool/0/0/0/ -> gives me files with 34 hexa digit. >> >> When I check (with ls -la) files date some are more than 1 year older >> than the other (eg: 2006 and 2007) > > The age doesn't matter - files will remain as long as some copy of > that same content is still in a backup run under the pc directory. > However, when the last copy under the pc directory expires and is > removed, the link count on the cpool instance will go down to 1 and > the next run of backuppc_nightly should remove it. I think I would > try stopping the backuppc service and running (as the backuppc user) > BackupPC_Nightly 0 255 > to see if it releases some space or has some error. After executing the command BackupPC_nightly 0 255 I couldn't see anything in output neither on display nor in log :-/ Is this command logging somewhere? But here is a copu from log after a full backup job: 2008-03-28 03:30:00 Running 2 BackupPC_nightly jobs from 0..15 (out of 0..15) 2008-03-28 03:30:00 Running BackupPC_nightly -m 0 127 (pid=15319) 2008-03-28 03:30:00 Running BackupPC_nightly 128 255 (pid=15320) 2008-03-28 03:30:00 Next wakeup is 2008-03-29 03:30:00 2008-03-28 03:30:01 Finished admin1 (BackupPC_nightly 128 255) 2008-03-28 03:30:01 BackupPC_nightly now running BackupPC_sendEmail 2008-03-28 03:30:01 Finished admin (BackupPC_nightly -m 0 127) 2008-03-28 03:30:01 Pool nightly clean removed 0 files of size 0.00GB 2008-03-28 03:30:01 Pool is 0.00GB, 0 files (0 repeated, 0 max chain, 0 max link s), 1 directories 2008-03-28 03:30:01 Cpool nightly clean removed 0 files of size 0.00GB 2008-03-28 03:30:01 Cpool is 0.00GB, 0 files (0 repeated, 0 max chain, 0 max lin ks), 4369 directories 2008-03-28 03:30:02 Started full backup on localhost (pid=15323, share=/.....path to share) Cheers, Hervé |
|
From: Les M. <les...@gm...> - 2008-03-28 17:51:32
|
Hervé Richard wrote:
>
>>>>
>>> When I execute ' find . -links 1 ' in the cpool directory I see a lot
>>> of files.
> After executing the command BackupPC_nightly 0 255 I couldn't see
> anything in output neither on display nor in log :-/
> Is this command logging somewhere?
I don't really know the details of the code, but it should have removed
the files that only have one link. Did it? And did it free up any space?
> But here is a copu from log after a full backup job:
> 2008-03-28 03:30:01 Pool nightly clean removed 0 files of size 0.00GB
> 2008-03-28 03:30:01 Pool is 0.00GB, 0 files (0 repeated, 0 max chain, 0
> max link
> s), 1 directories
This might be normal if your manual run already removed the expired
files, but if they are still there I'm about out of ideas.
--
Les Mikesell
le...@fu...
|
