Replying to myself...see at the bottom...

* Jean-Francois Malouin <Jean-Francois.Malouin AT bic.mni.mcgill DOT ca> 
[20080305 11:30]:
> Ok, here I go again. amanda-2.5.2p1 server and client debian/etch.
> 
> I have seen this in the past (last Oct iirc) and I remember that
> Jean-Louis provided me with a patch but that was for the case of a
> succesfully retried DLE. 
> In this particular case there was no retry but nevertheless amadmin
> and amfetchdump both claims that only the last tape containing the
> spanning dump is needed.
> 
> Has that patch been back-ported to 2.5.2p1 before I recompile the
> server from scratch? (Can't seem to find the patch!)
> 
> Here is the relevent info:
> 
> # /opt/amanda/sbin/amfetchdump -p -d /dev/nrst0 left1 \
> bigbrain BIG_BRAIN_pm3903_postmortem_tif 20080303 | tar -xpvf -
> 1 tape(s) needed for restoration
> The following tapes are needed: av24-1_left1_S00025L3
> Press enter when ready
> 
> But I know for a fact that the following tapes are needed:
> av24-1_left1_S00025L3
> av24-1_left1_S00024L3
> av24-1_left1_S00014L3
> av24-1_left1_S00013L3
> 
> as I can extract the chunks manually or with amrestore.
> 
> # su amanda -c "/opt/amanda/sbin/amadmin \
> left1 info bigbrain BIG_BRAIN_pm3903_postmortem_tif"
> 
> Current info for bigbrain BIG_BRAIN_pm3903_postmortem_tif:
>   Stats: dump rates (kps), Full:  67481.3, 81007.6,  -1.0
>                     Incremental:  120.0,  -1.0,  -1.0
>           compressed size, Full: -100.0%,-100.0%,-100.0%
>                     Incremental: -100.0%,-100.0%,-100.0%
>   Dumps: lev datestmp  tape             file   origK   compK secs
>           0  20080303  av24-1_left1_S00025L3  3 837780250 837780250 12415
> 
> # su amanda -c "/opt/amanda/sbin/amadmin left1 \
> find bigbrain BIG_BRAIN_pm3903_postmortem_tif"
> 
> 2008-03-03 19:10:01 bigbrain BIG_BRAIN_pm3903_postmortem_tif  0 
> av24-1_left1_S00025L3    1  158 OK
> 2008-03-03 19:10:01 bigbrain BIG_BRAIN_pm3903_postmortem_tif  0 
> av24-1_left1_S00025L3    2  159 OK
> 2008-03-03 19:10:01 bigbrain BIG_BRAIN_pm3903_postmortem_tif  0 
> av24-1_left1_S00025L3    3  160 OK
> 
> regards,
> jf
> -- 
> <° ><

I solved this by recompiling amanda from a fresh tarball and it is ok
now as amfetchdump requests all the tapes:

/opt/amanda/sbin/amfetchdump -p -d /dev/nrst0 left1 bigbrain \
BIG_BRAIN_pm3903_postmortem_tif 20080303 | tar -xpvf -
4 tape(s) needed for restoration
The following tapes are needed: av24-1_left1_S00013L3
av24-1_left1_S00014L3 av24-1_left1_S00024L3 av24-1_left1_S00025L3
Press enter when ready

However amadmin info' still reports only one tape at level 0.
Is this the expected behaviour?

su amanda -c "/opt/amanda/sbin/amadmin \
left1 info bigbrain BIG_BRAIN_pm3903_postmortem_tif"

Current info for bigbrain BIG_BRAIN_pm3903_postmortem_tif:
  Stats: dump rates (kps), Full:  67481.3, 81007.6,  -1.0
                    Incremental:  120.0,  -1.0,  -1.0
          compressed size, Full: -100.0%,-100.0%,-100.0%
                    Incremental: -100.0%,-100.0%,-100.0%
  Dumps: lev datestmp  tape             file   origK   compK secs
          0  20080303  av24-1_left1_S00025L3  3 837780250 837780250 12415
          1  20080308  av24-1_left1_S00028L3  8 120 120 0


regards,
jf
-- 
<° ><