Hi,
I have a strange problem with AMANDA. I have a few disks that
amadmin info shows having level 0s with the tape field blank,
and they aren't on holding disk. For example:
1 [atlas:amanda](~):amadmin astro0 info shemesh /mydisks/shemesh07
Current info for shemesh.astro.umn.edu /mydisks/shemesh07:
(Forcing to level 0 dump at next run)
Stats: dump rates (kps), Full: 462.0, 1077.0, 1170.0
Incremental: 32.0, 444.0, 758.0
compressed size, Full: 91.5%, 83.0%, 83.0%
Incremental: 160.0%, 55.9%, 55.9%
Dumps: lev datestmp tape file origK compK secs
0 20010803 0 8760070 8017824 17342
1 20010806 astro0_017 2 20 32 1
2 [atlas:amanda](~):
Here, amadmin find shows:
2001-08-03 shemesh.astro.umn.edu /mydisks/shemesh07 0 astro0_015 4 [out of tape]
Any idea why it wasn't marked failed?
I also have:
7 [atlas:amanda](~):amadmin astro0 info hal /user_a
Current info for hal.astro.umn.edu /user_a:
(Forcing to level 0 dump at next run)
Stats: dump rates (kps), Full: 730.0, 493.0, 550.0
Incremental: 471.0, 225.0, 277.0
compressed size, Full: 49.8%, 50.1%, 48.7%
Incremental: 47.6%, 41.8%, 38.2%
Dumps: lev datestmp tape file origK compK secs
0 20010712 0 9767430 4864544 6663
1 20010716 astro0_040 113 431110 204736 415
2 20010718 astro0_045 117 403530 184000 409
3 20010719 0 474490 203712 985
4 20010723 0 763000 341760 1627
5 20010726 astro0_000 96 604600 231488 586
6 20010730 astro0_005 111 682610 260640 506
7 20010803 astro0_014 111 494100 206336 916
8 20010806 astro0_016 95 438710 208704 443
8 [atlas:amanda](~):
8 [atlas:amanda](~):amadmin astro0 find hal /user_a
Scanning /mydisks/hold4/amanda...
20010808: found Amanda directory.
Scanning /mydisks/hold3/amanda...
20010808: found Amanda directory.
Scanning /mydisks/hold2/amanda...
20010808: found Amanda directory.
Scanning /mydisks/hold1/amanda...
20010808: found Amanda directory.
Scanning /mydisks/hold0/amanda...
gehrz: skipping cruft directory, perhaps you should delete it.
20010808: found Amanda directory.
date host disk lv tape or file file status
2001-07-08 hal.astro.umn.edu /user_a 1 astro0_035 47 OK
2001-07-09 hal.astro.umn.edu /user_a 1 astro0_037 49 OK
2001-07-10 hal.astro.umn.edu /user_a 1 astro0_030 86 OK
2001-07-16 hal.astro.umn.edu /user_a 1 astro0_040 113 OK
2001-07-17 hal.astro.umn.edu /user_a 2 astro0_043 110 OK
2001-07-18 hal.astro.umn.edu /user_a 2 astro0_045 117 OK
2001-07-26 hal.astro.umn.edu /user_a 5 astro0_000 96 OK
2001-07-27 hal.astro.umn.edu /user_a 6 astro0_002 104 OK
2001-07-30 hal.astro.umn.edu /user_a 6 astro0_005 111 OK
2001-07-31 hal.astro.umn.edu /user_a 7 astro0_008 107 OK
2001-08-01 hal.astro.umn.edu /user_a 7 astro0_010 112 OK
2001-08-02 hal.astro.umn.edu /user_a 7 astro0_012 113 OK
2001-08-03 hal.astro.umn.edu /user_a 7 astro0_014 111 OK
2001-08-06 hal.astro.umn.edu /user_a 8 astro0_016 95 OK
9 [atlas:amanda](~):
Notice here there's no record of the missing level 0,3,4 at all.
It's possible that these are the result of the AMANDA system
losing power and crashing during the run. Has anyone verified
that AMANDA keeps its state properly when the amdump run is
abruptly terminated?
Thanks for your help. Please CC any replies to [EMAIL PROTECTED]
Caio,
Ben
--
Ben Jarvis[bjarvis[at]astro[dot]umn[dot]edu]
System Administrator
Dept. of Astronomy/Space Physics Group
University of Minnesota - Twin Cities
[http://www.astro.umn.edu/~bjarvis]
PGP signature
