On Thursday 11 October 2007 22:59, Mariusz Czulada wrote:
> Hi,
>
> Dnia 11-10-2007 o godz. 18:55 Kern Sibbald napisał(a):
> > Hello,
> >
> > I went back through the emails, and one thing that I can say is that you
> > are
> > very persistent. You've been working on this quite a while and have not
> > given up!
>
> Have no options. I recommended this soft/hardware to the project, so it
> must (finally) works.
>
> > I'm still convinced that you have have one or more of the following
> > problems:
> > 1. A FC that is caching data/status codes, which will not work correctly
> > with
> > Bacula, which expects to get everything back from the drive
> > synchronously.
>
> Still investigating.
>
> > 2. A Device resource that is not really properly defined for the way
> > your tape
> > drive works.
>
> I take this into accout. But on the other hand: all device options were
> suggested by 'btape test'. Obviously it is not an oracle, but...
>
> Talking about definition - here is a part of bacula-sd.conf related to
> this DD (damn drive):
>
> Device {
> Name = "Dev:TS3310-drv0"
> Device Type = tape
> Media Type = LTO-3
> Archive Device = /dev/IBMtape0n
> AutomaticMount = yes; # when device opened, read it
> AlwaysOpen = yes;
> RemovableMedia = yes;
> RandomAccess = no;
> # Block positioning = no;
> Hardware End of Medium = No
> Fast Forward Space File = No
> BSF at EOM = yes
> TWO EOF = Yes
> Autochanger = yes;
> Drive Index = 0;
> }
Well, normally on a "modern" drive attached to a Linux machine, the following
directives should not be present:
Hardware End of Medium = No
Fast Forward Space File = No
TWO EOF = Yes
They virtually ensure that the drive will function very poorly or fail to work
correctly. Everything else as you have it looks pretty normal.
Have you ensured that the drive has all the right modes set? See the Tape
Testing manual.
I'll ask a Bacula user who has everything connected through FC what config he
is using.
Regards,
Kern
>
> > One suggestion that may reduce the complexity of the situation would be
> > to connect Bacula to your tape drive via SCSI rather than FC. This would
> > at least remove the potential problems with FC.
>
> This is also not an option. Hardware was already bought, and I could
> hardly explain why another buys are needed.
>
> Regards,
>
> Mariusz
>
> > > Regards,
> > >
> > > Mariusz Czulada
> > >
> > > -------------
> > >
> > > Dnia 8-10-2007 o godz. 14:52 Mariusz Czulada napisał(a):
> > > > Hi all,
> > > >
> > > > Unfortunately I didn't get any help from my local support regarding
> > > > my problem. In fact the answer was IMHO quite arrogant, like "System
> > > > must work correctly. We don't know/care about bacula. Use tar instead
> > > > or by Legato". So it seems I have to fight it alone...
> > > >
> > > > As a remainder, my configuration:
> > > >
> > > > Hardware configuration:
> > > > * xSeries 3690, CentOS 4 (like RH ES4)
> > > > * Two DS4000 FC 4 Gb PCI-X Single Port HBA; in system identified as
> > > > QLogic QLA2460 - PCI-X 2.0 to 4Gb FC, Single Channel
> > > > * TS3310 library, with two ULT3580-TD3 drives, connected directly to
> > > > HBAs.
> > > >
> > > > Drivers:
> > > > * drivers for qlaxxxx - v8.01.06
> > > > * SANsurfer FC HBA Manager High Availability Edition v2.0.30 Build 81
> > > > * IBMTape soft v. 3.0.5
> > > >
> > > > Firmware:
> > > > * QLA2460:
> > > > BIOS Version 1.12, Firmware Version: 4.00.23
> > > > * TS3310:
> > > > Name library
> > > > IBM Serial Number 1307494
> > > > 3576-MTL Firmware Version 320G.GS006
> > > > BSP Level BSP6.6
> > > > Last Update 9/29/2006 1:18
> > > > * ULT3580-TD3 drives:
> > > > Firmware Level: 69U2
> > > >
> > > > Bacula:
> > > > * 2.2.4, configured with "--prefix=/opt/bacula
> > > > --with-postgresql=/opt/pgsql-8.1/ --with-openssl"
> > > >
> > > > Probably noone else is using such or similar configuration so will
> > > > have to do myself some debugging to search for the reason. I'd like
> > > > to ask you to help me:
> > > >
> > > > * how to place timestamp in SD log file? It is quite detailed, but
> > > > almost impossible to correlate with IBMtaped trace file.
> > > >
> > > > * how to increase debug level in specific code? Most (all?) of my
> > > > problems appear in stored/dev.c/DEVICE::fsf method. It would be nice
> > > > to set detailed debugging only in this function instead of generating
> > > > megabytes of unneeded messages.
> > > >
> > > > TIA,
> > > >
> > > > Mariusz Czulada
>
> ----------------------------------------------------
> Walka o Mistrzostwo Świata! Krzysztof DIABLO Włodarczyk
> 20.10 po raz kolejny stanie ringu! Musisz to zobaczy!
> Kliknij:
> http://klik.wp.pl/?adr=http%3A%2F%2Fcorto.www.wp.pl%2Fas%2Folympic.html&sid
>=57
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems? Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
> _______________________________________________
> Bacula-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bacula-devel
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Bacula-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-devel
