On Thursday 25 August 2005 15:34, Tom Boyda wrote:
> Kern,
>
> Thank you for a very nice program.
>
> I am using MySQL 4.0.14 with 2 LTO-2 tape drives in a 36 slot library,
> and my backups will fit on one tape.
>
> It turns out that using the Berkeley devices /dev/rmt/0cbn instead of
> /dev/rmt/0cn was the magic to get the restores to work correctly.

Oh yes, that would create big problems.

> I verified that I could read the full backup of my Solaris server. Now to
> go test restores on some of my Windows servers.

Smart move.

>
> BTW, my Qualstar RLS-8236 36 slot library with 2 Ultrium LTO-2 tape
> drives, Solaris 9 OS (no barcode reader) works just fine under bacula
> 1.37.30. I will want to upgrade to the latest bacula again to get the
> improved autochanger capabilities soon.

If you upgrade to 1.37.36, you will need to apply two patches that I sent to 
the bacula-devel list.

Thanks for mentioning the drive specs, I'll add it to the manual.

Good luck.

>
> Thank you.
>
> Tom.
> ----
> Thomas Boyda
> [EMAIL PROTECTED]
> 276-739-2314
>
> -----Original Message-----
> From: Kern Sibbald [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, August 24, 2005 3:04 PM
> To: bacula-users@lists.sourceforge.net
> Cc: Tom Boyda
> Subject: Re: [Bacula-users] Problems restoring large(?) files
>
> If you are using PostgreSQL, and it involves multiple tapes, and the
> tapes are
> not read back in the order that they were written, it is possible that
> you
> have run into a rather obscure bug.
>
> That bug is fixed in 1.37.36, but the fix corrects a problem with
> writing the
> JobMedia records, so it won't do any good for data already written.
>
> You can manually correct it with a bit of pain by editing the bootstrap
> file
> and reordering the volumes and the entries that follow them up to the
> next
> volume in the order the volumes were written.
>
> On Wednesday 24 August 2005 20:39, Tom Boyda wrote:
> > Hello,
> >
> > I am having a problem with bacula 1.37.30 and restoring large(?)
>
> 150MB+
>
> > files. But it seems to be only some of the files (see extract from the
> > log below) and the files were *not* being used, they are
>
> self-extracting
>
> > archives being stored for download.
> >
> > The system is Solaris 9 and bacula is running on the solaris system
>
> and
>
> > was built using GCC 3.4.2. The tape drive(s) are LTO-2 with new tapes.
> >
> > Here is the snip from the failure e-mail:
> >
> > solweb-fd: -rw-r--r--   1 apache   apache   154879230 2004-11-16
> > 12:55:30
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ensemble_b150
>
> > _full.exe
> > solweb-fd: -r--r--r--   1 apache   apache   155404135 2004-11-16
> > 12:55:42
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ensemble_b150
>
> > _full.exe,v
> > solweb-fd: -rw-r--r--   1 apache   apache    13555375 2004-11-16
> > 12:55:43
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ens_short_162
>
> > .exe
> > solweb-fd: -r--r--r--   1 apache   apache    13607772 2004-11-16
> > 12:55:44
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ens_short_162
>
> > .exe,v
> > solweb-fd: -rw-r--r--   1 apache   apache   155325243 2004-11-16
> > 12:55:57
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ens_162_5.0.5
>
> > Ensemble.4.exe
> > solweb-fd: -r--r--r--   1 apache   apache   155853753 2004-11-16
> > 12:56:10
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ens_162_5.0.5
>
> > Ensemble.4.exe,v
> > 24-Aug 18:13 solweb-fd: RestoreFiles.2005-08-24_18.09.23 Error:
> > attribs.c:342 File size of restored file
>
> /local0/BACKUP/usr/local/apache2/twiki/pub/GISlab/Ensemble/Ens_162_5.0.5
>
> > Ensemble.4.exe,v not correct. Original 155853753, restored 152010752.
> > 24-Aug 18:13 solweb-sd: Got EOF at file 9  on device "LTO-TD2-1"
> > (/dev/rmt/1un), Volume "Weekly01-Vol001"
> > 24-Aug 18:13 solweb-sd: RestoreFiles.2005-08-24_18.09.23 Error:
> > block.c:928 Read error at file:blk 9:0 on device /dev/rmt/1un. ERR=I/O
> > error.
> > 24-Aug 18:13 solweb-sd: End of Volume at file 9 on device "LTO-TD2-1"
> > (/dev/rmt/1un), Volume "Weekly01-Vol001"
> > 24-Aug 18:13 solweb-sd: End of all volumes.
> > 24-Aug 18:13 solweb-dir: RestoreFiles.2005-08-24_18.09.23 Error:
>
> Bacula
>
> > 1.37.30 (14Jul05): 24-Aug-2005 18:13:33
> >
> > Any pointers would be appreciated.
> >
> > Let me know if you need any other information.
> >
> > Thank you,
> >
> > Tom.
> > ----
> > Thomas Boyda
> > Consultant
> > Daou Systems, Inc.
> > [EMAIL PROTECTED]
> > 276-739-2314
> >
> >
> >
> > -------------------------------------------------------
> > SF.Net email is Sponsored by the Better Software Conference & EXPO
> > September 19-22, 2005 * San Francisco, CA * Development Lifecycle
>
> Practices
>
> > Agile & Plan-Driven Development * Managing Projects & Teams * Testing
>
> & QA
>
> > Security * Process Improvement & Measurement *
>
> http://www.sqe.com/bsce5sf
>
> > _______________________________________________
> > Bacula-users mailing list
> > Bacula-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/bacula-users

-- 
Best regards,

Kern

  (">
  /\
  V_V


-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to