Autochanger incorrectly assigning first slot number

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

Autochanger incorrectly assigning first slot number

Ivan Adzhubey
Hi,

I have configured Bacula for our Qualstar TLS-8466 autochanger and everything
seems be working fine except for the weird errors in slot numbering. The
problem I suspect is with the extra I/O slot which this model has, in addition
to normal slots in the mounted magazines. Mtx command correctly reports all
magazine slots plus the I/O slot (separately). However, Bacula director seems
to have issues with the I/O slot: when command is issued to mount slot 1, it
does mount the tape from first slot of the first magazine but somehow records
mounted slot number as (last magazine slot + 1), e.g. slot 45 when there are 4
x 11-slot magazines (total regular 44 slots) installed in the changer. Weirdly
enough, everything seems to work fine whether the this slot is referred to as
slot=1 or slot=45 in mount/umount commands. It looks like I/O slot address
overlaps with the first slot address. Not sure if this is a feature or a bug?
Should I leave first slot of the first magazine empty to avoid the confusion or
is there some other known workaround?

Thanks,
Ivan

--
Ivan Adzhubey, Ph.D.
Instructor
Division of Genetics, Dept of Medicine
Brigham & Women's Hospital
Harvard Medical School
New Research Building, Room 0464C
77 Avenue Louis Pasteur
Boston, MA 02115
tel.: (617) 525-4728
fax:  (617) 525-4705
web: http://genetics.bwh.harvard.edu/wiki/sunyaevlab/



The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Ivan Adzhubey
On Tuesday, June 27, 2017 5:42:10 PM EDT Ivan Adzhubey wrote:
> The
> problem I suspect is with the extra I/O slot which this model has, in
> addition to normal slots in the mounted magazines. Mtx command correctly
> reports all magazine slots plus the I/O slot (separately). However, Bacula
> director seems to have issues with the I/O slot

Correction: This should read "Import/Export" (I/E) slot, not I/O slot. Sorry
about that typo.

Google finds some reports of issues with mtx-changer script failing to
properly parse I/E slot lines returned by "mtx status" for the changers with
I/E slots, due to extra "IMPORT/EXPORT" tag inserted. However, my case is the
opposite since "mtx status" for my changer does not tag the (single) I/E slot
at all, it just appends one extra slot to the list, so that the last slot
listed is always a I/E one:

# mtx -f /dev/changer status
  Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
Data Transfer Element 0:Empty
Data Transfer Element 1:Empty
      Storage Element 1:Full :VolumeTag=A00000L5                        
      Storage Element 2:Full :VolumeTag=A00001L5
<...skipped...>
      Storage Element 34:Full :VolumeTag=A00033L5                        
      Storage Element 35:Full :VolumeTag=A00034L5                        
      Storage Element 36:Full :VolumeTag=A00035L5                        
      Storage Element 37:Empty
      Storage Element 38:Empty
      Storage Element 39:Empty
      Storage Element 40:Empty
      Storage Element 41:Empty
      Storage Element 42:Empty
      Storage Element 43:Empty
      Storage Element 44:Empty
      Storage Element 45:Empty
<end of status command output>

Storage element 45 listed above is in fact the I/E slot. For some reason,
Bacula flips it with the slot 1, so when I mount slot 1 in bconsole, it mounts
it as slot 45. However, actual tape from slot 1 is loaded.

--Ivan


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Ivan Adzhubey
Hi,

After more testing done today, this looks like a mtx bug: issuing command "mtx
load 1 0" correctly loads tape from the changer slot 1 into the first tape
drive. However, "mtx status" then shows this tape drive as loaded with slot 45
instead (but also correctly lists slot 1 as empty):

# mtx status
  Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
Data Transfer Element 0:Empty
Data Transfer Element 1:Empty
      Storage Element 1:Full :VolumeTag=A00000L5                        
      Storage Element 2:Full :VolumeTag=A00001L5
<...>

# mtx load 1 0
Loading media from Storage Element 1 into drive 0...done

# mtx status  
  Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
Data Transfer Element 0:Full (Storage Element 45 Loaded):VolumeTag = A00000L5                        
Data Transfer Element 1:Empty
      Storage Element 1:Empty
      Storage Element 2:Full :VolumeTag=A00001L5                        
      Storage Element 3:Full :VolumeTag=A00002L5
<...>

Unfortunately, mtx project seems dead, so no updates...

--Ivan

On Tuesday, June 27, 2017 8:34:47 PM EDT Ivan Adzhubey wrote:

> On Tuesday, June 27, 2017 5:42:10 PM EDT Ivan Adzhubey wrote:
> > The
> > problem I suspect is with the extra I/O slot which this model has, in
> > addition to normal slots in the mounted magazines. Mtx command correctly
> > reports all magazine slots plus the I/O slot (separately). However, Bacula
> > director seems to have issues with the I/O slot
>
> Correction: This should read "Import/Export" (I/E) slot, not I/O slot. Sorry
> about that typo.
>
> Google finds some reports of issues with mtx-changer script failing to
> properly parse I/E slot lines returned by "mtx status" for the changers with
> I/E slots, due to extra "IMPORT/EXPORT" tag inserted. However, my case is
> the opposite since "mtx status" for my changer does not tag the (single)
> I/E slot at all, it just appends one extra slot to the list, so that the
> last slot listed is always a I/E one:
>
> # mtx -f /dev/changer status
>   Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
> Data Transfer Element 0:Empty
> Data Transfer Element 1:Empty
>       Storage Element 1:Full :VolumeTag=A00000L5
>       Storage Element 2:Full :VolumeTag=A00001L5
> <...skipped...>
>       Storage Element 34:Full :VolumeTag=A00033L5
>       Storage Element 35:Full :VolumeTag=A00034L5
>       Storage Element 36:Full :VolumeTag=A00035L5
>       Storage Element 37:Empty
>       Storage Element 38:Empty
>       Storage Element 39:Empty
>       Storage Element 40:Empty
>       Storage Element 41:Empty
>       Storage Element 42:Empty
>       Storage Element 43:Empty
>       Storage Element 44:Empty
>       Storage Element 45:Empty
> <end of status command output>
>
> Storage element 45 listed above is in fact the I/E slot. For some reason,
> Bacula flips it with the slot 1, so when I mount slot 1 in bconsole, it
> mounts it as slot 45. However, actual tape from slot 1 is loaded.



The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Kern Sibbald
Hello,

I think this is what we politely call a "pilot error/misunderstanding".  
From what you have described, mtx is behaving correctly.  Normally I
turn off the mail slot (aka import/export slot) on my autochanger and
the problem goes away. It is my understanding that Bacula can be used
with a mail slot, but I do not know how -- probably the simplest is to
make sure there is no volume in the mail slot when you are doing an
"update slots barcodes" command.  If you do that, Bacula should never
access slot 1.

Best regards,

Kern


On 06/28/2017 05:44 AM, Ivan Adzhubey wrote:

> Hi,
>
> After more testing done today, this looks like a mtx bug: issuing command "mtx
> load 1 0" correctly loads tape from the changer slot 1 into the first tape
> drive. However, "mtx status" then shows this tape drive as loaded with slot 45
> instead (but also correctly lists slot 1 as empty):
>
> # mtx status
>    Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
> Data Transfer Element 0:Empty
> Data Transfer Element 1:Empty
>        Storage Element 1:Full :VolumeTag=A00000L5
>        Storage Element 2:Full :VolumeTag=A00001L5
> <...>
>
> # mtx load 1 0
> Loading media from Storage Element 1 into drive 0...done
>
> # mtx status
>    Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
> Data Transfer Element 0:Full (Storage Element 45 Loaded):VolumeTag = A00000L5
> Data Transfer Element 1:Empty
>        Storage Element 1:Empty
>        Storage Element 2:Full :VolumeTag=A00001L5
>        Storage Element 3:Full :VolumeTag=A00002L5
> <...>
>
> Unfortunately, mtx project seems dead, so no updates...
>
> --Ivan
>
> On Tuesday, June 27, 2017 8:34:47 PM EDT Ivan Adzhubey wrote:
>> On Tuesday, June 27, 2017 5:42:10 PM EDT Ivan Adzhubey wrote:
>>> The
>>> problem I suspect is with the extra I/O slot which this model has, in
>>> addition to normal slots in the mounted magazines. Mtx command correctly
>>> reports all magazine slots plus the I/O slot (separately). However, Bacula
>>> director seems to have issues with the I/O slot
>> Correction: This should read "Import/Export" (I/E) slot, not I/O slot. Sorry
>> about that typo.
>>
>> Google finds some reports of issues with mtx-changer script failing to
>> properly parse I/E slot lines returned by "mtx status" for the changers with
>> I/E slots, due to extra "IMPORT/EXPORT" tag inserted. However, my case is
>> the opposite since "mtx status" for my changer does not tag the (single)
>> I/E slot at all, it just appends one extra slot to the list, so that the
>> last slot listed is always a I/E one:
>>
>> # mtx -f /dev/changer status
>>    Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
>> Data Transfer Element 0:Empty
>> Data Transfer Element 1:Empty
>>        Storage Element 1:Full :VolumeTag=A00000L5
>>        Storage Element 2:Full :VolumeTag=A00001L5
>> <...skipped...>
>>        Storage Element 34:Full :VolumeTag=A00033L5
>>        Storage Element 35:Full :VolumeTag=A00034L5
>>        Storage Element 36:Full :VolumeTag=A00035L5
>>        Storage Element 37:Empty
>>        Storage Element 38:Empty
>>        Storage Element 39:Empty
>>        Storage Element 40:Empty
>>        Storage Element 41:Empty
>>        Storage Element 42:Empty
>>        Storage Element 43:Empty
>>        Storage Element 44:Empty
>>        Storage Element 45:Empty
>> <end of status command output>
>>
>> Storage element 45 listed above is in fact the I/E slot. For some reason,
>> Bacula flips it with the slot 1, so when I mount slot 1 in bconsole, it
>> mounts it as slot 45. However, actual tape from slot 1 is loaded.
>
>
> The information in this e-mail is intended only for the person to whom it is
> addressed. If you believe this e-mail was sent to you in error and the e-mail
> contains patient information, please contact the Partners Compliance HelpLine at
> http://www.partners.org/complianceline . If the e-mail was sent to you in error
> but does not contain patient information, please contact the sender and properly
> dispose of the e-mail.
>
>
> ------------------------------------------------------------------------------
> 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
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Clark, Patti
Ivan,

You’ve not indicated what OS that you are using – it’s somewhat difficult to say whether your mtx package has a bug.  I’ve never had this issue with mtx running on RHEL6.  I don’t see your I/E slot being identified as the I/E slot, just a regular slot, which can be done if slots are tight.  It’s a configurable parameter on the library itself as are the slots.  It looks like a wrap around identification problem.

My library, while not a Qualstar, does identify the I/E slots with the mtx command:
                           
 Storage Changer /dev/changer-admin:4 Drives, 280 Slots ( 10 Import/Export )
Data Transfer Element 0:Full (Storage Element 189 Loaded):VolumeTag = 405128L5                            
Data Transfer Element 1:Full (Storage Element 227 Loaded):VolumeTag = 399696L5                            
Data Transfer Element 2:Full (Storage Element 226 Loaded):VolumeTag = 399695L5                            
Data Transfer Element 3:Full (Storage Element 1 Loaded):VolumeTag = 074260L5                            
      Storage Element 1:Empty:VolumeTag=                                    
      Storage Element 2:Full :VolumeTag=074261L5                            
.
.
.                            
      Storage Element 270:Full :VolumeTag=621469L5
      Storage Element 271 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 272 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 273 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 274 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 275 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 276 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 277 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 278 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 279 IMPORT/EXPORT:Empty:VolumeTag=                                    
      Storage Element 280 IMPORT/EXPORT:Empty:VolumeTag=              

And Kern, disabling I/E ports, really?  I have used bconsole commands to move media to the I/E ports, but I’ve scripted to just change the volume location to 0 and inchanger=no, and I use the mtx commands to manage moving the volumes to the I/E.  The process is handled in reverse for importing.

Patti Clark
Sr. Linux System Administrator
Oak Ridge National Laboratory

On 6/28/17, 5:01 AM, "Kern Sibbald" <[hidden email]> wrote:

    Hello,
   
    I think this is what we politely call a "pilot error/misunderstanding".  
    From what you have described, mtx is behaving correctly.  Normally I
    turn off the mail slot (aka import/export slot) on my autochanger and
    the problem goes away. It is my understanding that Bacula can be used
    with a mail slot, but I do not know how -- probably the simplest is to
    make sure there is no volume in the mail slot when you are doing an
    "update slots barcodes" command.  If you do that, Bacula should never
    access slot 1.
   
    Best regards,
   
    Kern
   
   
    On 06/28/2017 05:44 AM, Ivan Adzhubey wrote:
    > Hi,
    >
    > After more testing done today, this looks like a mtx bug: issuing command "mtx
    > load 1 0" correctly loads tape from the changer slot 1 into the first tape
    > drive. However, "mtx status" then shows this tape drive as loaded with slot 45
    > instead (but also correctly lists slot 1 as empty):
    >
    > # mtx status
    >    Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
    > Data Transfer Element 0:Empty
    > Data Transfer Element 1:Empty
    >        Storage Element 1:Full :VolumeTag=A00000L5
    >        Storage Element 2:Full :VolumeTag=A00001L5
    > <...>
    >
    > # mtx load 1 0
    > Loading media from Storage Element 1 into drive 0...done
    >
    > # mtx status
    >    Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
    > Data Transfer Element 0:Full (Storage Element 45 Loaded):VolumeTag = A00000L5
    > Data Transfer Element 1:Empty
    >        Storage Element 1:Empty
    >        Storage Element 2:Full :VolumeTag=A00001L5
    >        Storage Element 3:Full :VolumeTag=A00002L5
    > <...>
    >
    > Unfortunately, mtx project seems dead, so no updates...
    >
    > --Ivan
    >
    > On Tuesday, June 27, 2017 8:34:47 PM EDT Ivan Adzhubey wrote:
    >> On Tuesday, June 27, 2017 5:42:10 PM EDT Ivan Adzhubey wrote:
    >>> The
    >>> problem I suspect is with the extra I/O slot which this model has, in
    >>> addition to normal slots in the mounted magazines. Mtx command correctly
    >>> reports all magazine slots plus the I/O slot (separately). However, Bacula
    >>> director seems to have issues with the I/O slot
    >> Correction: This should read "Import/Export" (I/E) slot, not I/O slot. Sorry
    >> about that typo.
    >>
    >> Google finds some reports of issues with mtx-changer script failing to
    >> properly parse I/E slot lines returned by "mtx status" for the changers with
    >> I/E slots, due to extra "IMPORT/EXPORT" tag inserted. However, my case is
    >> the opposite since "mtx status" for my changer does not tag the (single)
    >> I/E slot at all, it just appends one extra slot to the list, so that the
    >> last slot listed is always a I/E one:
    >>
    >> # mtx -f /dev/changer status
    >>    Storage Changer /dev/changer:2 Drives, 45 Slots ( 1 Import/Export )
    >> Data Transfer Element 0:Empty
    >> Data Transfer Element 1:Empty
    >>        Storage Element 1:Full :VolumeTag=A00000L5
    >>        Storage Element 2:Full :VolumeTag=A00001L5
    >> <...skipped...>
    >>        Storage Element 34:Full :VolumeTag=A00033L5
    >>        Storage Element 35:Full :VolumeTag=A00034L5
    >>        Storage Element 36:Full :VolumeTag=A00035L5
    >>        Storage Element 37:Empty
    >>        Storage Element 38:Empty
    >>        Storage Element 39:Empty
    >>        Storage Element 40:Empty
    >>        Storage Element 41:Empty
    >>        Storage Element 42:Empty
    >>        Storage Element 43:Empty
    >>        Storage Element 44:Empty
    >>        Storage Element 45:Empty
    >> <end of status command output>
    >>
    >> Storage element 45 listed above is in fact the I/E slot. For some reason,
    >> Bacula flips it with the slot 1, so when I mount slot 1 in bconsole, it
    >> mounts it as slot 45. However, actual tape from slot 1 is loaded.
    >
    >
    > The information in this e-mail is intended only for the person to whom it is
    > addressed. If you believe this e-mail was sent to you in error and the e-mail
    > contains patient information, please contact the Partners Compliance HelpLine at
    > http://www.partners.org/complianceline . If the e-mail was sent to you in error
    > but does not contain patient information, please contact the sender and properly
    > dispose of the e-mail.
    >
    >
    > ------------------------------------------------------------------------------
    > 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
    > [hidden email]
    > https://lists.sourceforge.net/lists/listinfo/bacula-users
    >
   
   
    ------------------------------------------------------------------------------
    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
    [hidden email]
    https://lists.sourceforge.net/lists/listinfo/bacula-users
   
   

------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Ivan Adzhubey
In reply to this post by Kern Sibbald
Hi Kern,

On Wednesday, June 28, 2017 11:01:01 AM Kern Sibbald wrote:
> Hello,
>
> I think this is what we politely call a "pilot error/misunderstanding".

Which one? I do not see any error I might have made, certainly nothing that
would require strong words ;-)

> From what you have described, mtx is behaving correctly.  Normally I
> turn off the mail slot (aka import/export slot) on my autochanger and
> the problem goes away. It is my understanding that Bacula can be used
> with a mail slot, but I do not know how -- probably the simplest is to
> make sure there is no volume in the mail slot when you are doing an
> "update slots barcodes" command.  If you do that, Bacula should never
> access slot 1.

Disabling mailslot in the changer's configuration was my initial idea as well.
However, I decided to first try installing full set of magazines into the
carousel, occupying all carousel faces, and lo and behold - this has fixed all
mtx slot numbering errors immediately.

Before that, we had some faces on the carousel completely vacant, with no
magazines installed. Looks like such configuration, while supported by the
changer documentation, confused mtx and caused it to loose track of the
mailslot position. Now, with all carousel faces occupied by magazines (some of
them empty), the mailslot is correctly tagged in the mtx output by the
"IMPORT/EXPORT" string and load/unload commands no longer confuse storage
element slot #1 with the mailslot.

I am not sure how common this problem could be. We previously ran Qualstar
RLS-4445 changer with varying number of magazines installed and never had any
issues with Bacula loosing track of the slot numbers. This model however, did
not have a mailslot.

--Ivan


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Ivan Adzhubey
In reply to this post by Clark, Patti
Hi Patti,

My server is running Ubuntu 16.04.2 LTS (64bit), Bacula v7.4.7 and mtx tools
are at v1.3.12.

As I mentioned in my other post, in reply to Kern's response, I've managed to
fix the issue by installing full set of magazines into the changer's carousel.
This immediately caused mtx to properly list the mailslot as IMPORT/EXPORT in
its output, which was not the case before.

Looks like the problem was caused by some mis-interaction between the
changer's firmware and mtx. Whether this qualify as a bug I could not say. But
I would take a note to be careful not to leave any of the carousel faces
vacant even temporarily, to avoid Bacula suddenly losing track of parts of its
tape inventory.

--Ivan

On Wednesday, June 28, 2017 06:46:33 PM Clark, Patti wrote:
> Ivan,
>
> You’ve not indicated what OS that you are using – it’s somewhat difficult to
> say whether your mtx package has a bug.  I’ve never had this issue with mtx
> running on RHEL6.  I don’t see your I/E slot being identified as the I/E
> slot, just a regular slot, which can be done if slots are tight.  It’s a
> configurable parameter on the library itself as are the slots.  It looks
> like a wrap around identification problem.
 
> My library, while not a Qualstar, does identify the I/E slots with the mtx
> command:
 

>  Storage Changer /dev/changer-admin:4 Drives, 280 Slots ( 10 Import/Export
> ) Data Transfer Element 0:Full (Storage Element 189 Loaded):VolumeTag =
> 405128L5 Data Transfer Element 1:Full (Storage Element 227
> Loaded):VolumeTag = 399696L5 Data Transfer Element 2:Full (Storage Element
> 226 Loaded):VolumeTag = 399695L5 Data Transfer Element 3:Full (Storage
> Element 1 Loaded):VolumeTag = 074260L5 Storage Element 1:Empty:VolumeTag=
> Storage Element 2:Full :VolumeTag=074261L5 .
> .
> .                            
>       Storage Element 270:Full :VolumeTag=621469L5
>       Storage Element 271 IMPORT/EXPORT:Empty:VolumeTag=                    
>       Storage Element 272 IMPORT/EXPORT:Empty:VolumeTag= Storage
> Element 273 IMPORT/EXPORT:Empty:VolumeTag= Storage Element 274
> IMPORT/EXPORT:Empty:VolumeTag= Storage Element 275
> IMPORT/EXPORT:Empty:VolumeTag= Storage Element 276
> IMPORT/EXPORT:Empty:VolumeTag= Storage Element 277
> IMPORT/EXPORT:Empty:VolumeTag= Storage Element 278
> IMPORT/EXPORT:Empty:VolumeTag= Storage Element 279
> IMPORT/EXPORT:Empty:VolumeTag= Storage Element 280
> IMPORT/EXPORT:Empty:VolumeTag=
> And Kern, disabling I/E ports, really?  I have used bconsole commands to
> move media to the I/E ports, but I’ve scripted to just change the volume
> location to 0 and inchanger=no, and I use the mtx commands to manage moving
> the volumes to the I/E.  The process is handled in reverse for importing.
 
> Patti Clark
> Sr. Linux System Administrator
> Oak Ridge National Laboratory


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|

Re: Autochanger incorrectly assigning first slot number

Kern Sibbald
In reply to this post by Ivan Adzhubey
Hello Ivan,

I was trying to point out that in general, to avoid spinning ones
wheels, one should not initially point at bugs in Bacula or bugs in mtx,
since most initial setup problems are a question of misunderstanding
(perhaps unclear documentation) and configuration issues. Certainly bugs
do exist. In this case, it seems to me that Patti Clark has given you
the necessary configuration "tricks" to make Bacula and mtx work
correctly with a mail slot.

Best regards,

Kern


On 06/28/2017 10:53 PM, Ivan Adzhubey wrote:

> Hi Kern,
>
> On Wednesday, June 28, 2017 11:01:01 AM Kern Sibbald wrote:
>> Hello,
>>
>> I think this is what we politely call a "pilot error/misunderstanding".
> Which one? I do not see any error I might have made, certainly nothing that
> would require strong words ;-)
>
>>  From what you have described, mtx is behaving correctly.  Normally I
>> turn off the mail slot (aka import/export slot) on my autochanger and
>> the problem goes away. It is my understanding that Bacula can be used
>> with a mail slot, but I do not know how -- probably the simplest is to
>> make sure there is no volume in the mail slot when you are doing an
>> "update slots barcodes" command.  If you do that, Bacula should never
>> access slot 1.
> Disabling mailslot in the changer's configuration was my initial idea as well.
> However, I decided to first try installing full set of magazines into the
> carousel, occupying all carousel faces, and lo and behold - this has fixed all
> mtx slot numbering errors immediately.
>
> Before that, we had some faces on the carousel completely vacant, with no
> magazines installed. Looks like such configuration, while supported by the
> changer documentation, confused mtx and caused it to loose track of the
> mailslot position. Now, with all carousel faces occupied by magazines (some of
> them empty), the mailslot is correctly tagged in the mtx output by the
> "IMPORT/EXPORT" string and load/unload commands no longer confuse storage
> element slot #1 with the mailslot.
>
> I am not sure how common this problem could be. We previously ran Qualstar
> RLS-4445 changer with varying number of magazines installed and never had any
> issues with Bacula loosing track of the slot numbers. This model however, did
> not have a mailslot.
>
> --Ivan
>


------------------------------------------------------------------------------
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
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users