Re: [Bacula-users] Bacula 7.4.4: Strange warnings non-existent filesystem crossings

2016-10-16 Thread Andreas Koch 
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Here's a brief update on this issue: After taking the automounter out of the
equation, and hard-mounting all file systems, the last differential did not
show the spurious warnings on crossing file-system boundaries.

Whether the Accurate differential backup actually worked, I do not know yet
(see other thread titled `More Bacula 7.4.4 differential Backup
strangeness'), that will take more investigation.

Best,
  Andreas

On 10/14/2016 01:50 PM, Andreas Koch wrote:
> Hello all,
> 
> we have observed strange behavior in our recent backups. We have a number
> of filesystems, and explicitly list those to be backed up in the fileset
> 
> One of these file systems is
> 
> /home/stud
> 
> which has subdirectories for each user. Backing these up has worked
> perfectly for the last decade by just including the file system in the
> FileSet
> 
> FileSet { Name = "fs_gundabad" Include { Options { signature = SHA1; 
> verify=pins1; accurate=pins1; xattrsupport = yes; } ... File =
> /home/stud ... }
> 
> However, a recent differential backup shows warnings for _all_ of user 
> subdirectories of the form:
> 
> 13-Oct 03:25 gundabad-fd JobId 2161:  /home/stud/aqsprak01 is a
> different filesystem. Will not descend from /home/stud into it. 13-Oct
> 03:25 gundabad-fd JobId 2161:  /home/stud/aqsprak02 is a different 
> filesystem. Will not descend from /home/stud into it. ...
> 
> Even stranger, the backup _does_ appear to contain files from these 
> directories (or more accurately, shows them in the catalog):
> 
> * list files jobid=2161 ... 
> /home/stud/aqsprak01/Templates/test01/projSRC_Master/3cuS_1_l.dat ...
> 
> Has anyone observed something like this elsewhere? As I wrote above, we
> use Bacula 7.4.4, on Scientific Linux 7.2 backing-up NFS v4.0 filesystems
> mounted via automounter from an EMC VNX 5200.
> 
> Best, Andreas
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEUEARECAAYFAlgDcPgACgkQk5ta2EV7DoxaxACfVTL/LvSnZNtFqaIzPIuUCTlk
2AoAl3LNp5Lf+Yi+3FgO7Ge8q2CFtiA=
=Qes0
-END PGP SIGNATURE-

--
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula 7.4.4: Strange warnings non-existent filesystem crossings

2016-10-14 Thread Andreas Koch 
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 10/14/2016 05:31 PM, Hankins, Jonathan wrote:
> I'm not really familiar with EMC's NFS, NFSv4 and automounter, but here's
> a shot in the dark - what do your export and mount options look like? In 
> particular, look into nohide or crossmnt export options. Also, have a
> look at this:
> https://utcc.utoronto.ca/~cks/space/blog/unix/MultiFilesystemNFSIssue. 
> Just guesses, may not have any bearing on your issue.

Hello Jonathan,

we had just been using something pretty standard:

stud   -rsize=32768,wsize=32768,vers=4   moria-0:/stud

The export setup is also really simple: the exported file systems themselves
have no other filesystems mounted inside, so crossmnt and nohide are not
applicable to our use-case.

At the moment, I suspect misbehavior on the part of the automounter. I have
just hardmounted the file systems (which does not hurt us on that specific
server) and will see how the next Differential on Sunday morning fares.

Best,
  Andreas


> 
> -Jonathan Hankins
> 
> On Fri, Oct 14, 2016 at 7:34 AM Andreas Koch 
>  > wrote:
> 
> Hello all,
> 
> we have observed strange behavior in our recent backups. We have a number
> of filesystems, and explicitly list those to be backed up in the fileset
> 
> One of these file systems is
> 
> /home/stud
> 
> which has subdirectories for each user. Backing these up has worked
> perfectly for the last decade by just including the file system in the
> FileSet
> 
> FileSet { Name = "fs_gundabad" Include { Options { signature = SHA1; 
> verify=pins1; accurate=pins1; xattrsupport = yes; } ... File =
> /home/stud ... }
> 
> However, a recent differential backup shows warnings for _all_ of user 
> subdirectories of the form:
> 
> 13-Oct 03:25 gundabad-fd JobId 2161:  /home/stud/aqsprak01 is a
> different filesystem. Will not descend from /home/stud into it. 13-Oct
> 03:25 gundabad-fd JobId 2161:  /home/stud/aqsprak02 is a different 
> filesystem. Will not descend from /home/stud into it. ...
> 
> Even stranger, the backup _does_ appear to contain files from these 
> directories (or more accurately, shows them in the catalog):
> 
> * list files jobid=2161 ... 
> /home/stud/aqsprak01/Templates/test01/projSRC_Master/3cuS_1_l.dat ...
> 
> Has anyone observed something like this elsewhere? As I wrote above, we
> use Bacula 7.4.4, on Scientific Linux 7.2 backing-up NFS v4.0 filesystems
> mounted via automounter from an EMC VNX 5200.
> 
> Best, Andreas
> 
> 
- --
>
> 
Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! 
> http://sdm.link/slashdot___ 
> Bacula-users mailing list Bacula-users@lists.sourceforge.net
>  
> https://lists.sourceforge.net/lists/listinfo/bacula-users
> 
> 
> This e-mail is intended only for the recipient and may contain 
> confidential or  proprietary information. If you are not the intended 
> recipient, the review,  distribution, duplication or retention of this 
> message and its attachments is  prohibited. Please notify the sender of 
> this error immediately by reply  e-mail, and permanently delete this 
> message and its attachments in any form in  which they may have been 
> preserved.
> 
> 
> 
- --
>
> 
Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> 
> 
> 
> ___ Bacula-users mailing
> list Bacula-users@lists.sourceforge.net 
> https://lists.sourceforge.net/lists/listinfo/bacula-users
> 


- -- 
*** !!! Geänderte Rufnummern / Phone numbers have changed !!! **
Prof. Dr.-Ing. Andreas Koch  k...@esa.cs.tu-darmstadt.de
Technische Universitaet Darmstadt, FB20   Phone  : x49-6151-16-22420
Embedded Systems and Applications Group (ESA) FAX: x49-6151-16-22422
Hochschulstr. 10, D-64289 Darmstadt, Germany   * PGP key available *
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAlgBGF8ACgkQk5ta2EV7DoxXkgCgomQ1+eOamkoz8Q7AmPJKiPI6
XEkAnAq9kkl+URSgOhwja4t/xGW4Mhrm
=L4Hp
-END PGP SIGNATURE-

--
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula 7.4.4: Strange warnings non-existent filesystem crossings

2016-10-14 Thread Hankins, Jonathan 
I'm not really familiar with EMC's NFS, NFSv4 and automounter, but here's a
shot in the dark - what do your export and mount options look like? In
particular, look into nohide or crossmnt export options. Also, have a look
at this:
https://utcc.utoronto.ca/~cks/space/blog/unix/MultiFilesystemNFSIssue. Just
guesses, may not have any bearing on your issue.

-Jonathan Hankins

On Fri, Oct 14, 2016 at 7:34 AM Andreas Koch <
k...@esa.informatik.tu-darmstadt.de> wrote:

> Hello all,
>
> we have observed strange behavior in our recent backups. We have a number
> of
> filesystems, and explicitly list those to be backed up in the fileset
>
> One of these file systems is
>
> /home/stud
>
> which has subdirectories for each user. Backing these up has worked
> perfectly
> for the last decade by just including the file system in the FileSet
>
> FileSet {
>   Name = "fs_gundabad"
>   Include {
> Options {
>   signature = SHA1;
>   verify=pins1;
>   accurate=pins1;
>   xattrsupport = yes;
> }
> ...
> File = /home/stud
> ...
> }
>
> However, a recent differential backup shows warnings for _all_ of user
> subdirectories of the form:
>
> 13-Oct 03:25 gundabad-fd JobId 2161:  /home/stud/aqsprak01 is a
> different
> filesystem. Will not descend from /home/stud into it.
> 13-Oct 03:25 gundabad-fd JobId 2161:  /home/stud/aqsprak02 is a
> different
> filesystem. Will not descend from /home/stud into it.
> ...
>
> Even stranger, the backup _does_ appear to contain files from these
> directories (or more accurately, shows them in the catalog):
>
> * list files jobid=2161
> ...
> /home/stud/aqsprak01/Templates/test01/projSRC_Master/3cuS_1_l.dat
> ...
>
> Has anyone observed something like this elsewhere? As I wrote above, we use
> Bacula 7.4.4, on Scientific Linux 7.2 backing-up NFS v4.0 filesystems
> mounted
> via automounter from an EMC VNX 5200.
>
> Best,
>   Andreas
>
>
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> ___
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>

-- 
This e-mail is intended only for the recipient and may contain confidential 
or proprietary information. If you are not the intended recipient, the 
review, distribution, duplication or retention of this message and its 
attachments is prohibited. Please notify the sender of this error 
immediately by reply e-mail, and permanently delete this message and its 
attachments in any form in which they may have been preserved.
--
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users