Am 23.02.2006 um 11:17 schrieb Paul Bijnens:

On 2006-02-23 10:45, Stefan Herrmann wrote:

this is one of my amstatus reports:
amanda@amanda:~/hourly$ amstatus hourly

Using /usr/local/etc/amanda/hourly/amdump.1 from Thu Feb 23 10:20:59 CET 2006 pille.hq.imos.net:/ 0 driver: (aborted:[request failed: timeout waiting for ACK])(too many dumper retry) pille.hq.imos.net:/opt 1 driver: (aborted:[request failed: timeout waiting for ACK])(too many dumper retry) pille.hq.imos.net:/usr 1 driver: (aborted:[request failed: timeout waiting for ACK])(too many dumper retry) pille.hq.imos.net:/var 1 driver: (aborted:[request failed: timeout waiting for ACK])(too many dumper retry)

SUMMARY          part      real  estimated
                           size       size
partition       :   4
estimated       :   4                 5154m
flush           :   0         0m
failed          :   4                 5154m           (100.00%)
wait for dumping:   0                    0m           (  0.00%)
dumping to tape :   0                    0m           (  0.00%)
dumping         :   0         0m         0m (  0.00%) (  0.00%)
dumped          :   0         0m         0m (  0.00%) (  0.00%)
wait for writing:   0         0m         0m (  0.00%) (  0.00%)
wait to flush   :   0         0m         0m (100.00%) (  0.00%)
writing to tape :   0         0m         0m (  0.00%) (  0.00%)
failed to tape  :   0         0m         0m (  0.00%) (  0.00%)
taped           :   0         0m         0m (  0.00%) (  0.00%)
  tape 1        :   0         0m         0m (  0.00%) hourly022
6 dumpers idle  : no-dumpers
taper idle
network free kps:     14000
holding space   :    263392m (100.00%)
 dumper0 busy   :  0:03:59  ( 40.97%)

0 dumpers busy : 0:09:45 ( 99.97%) not-idle: 0:06:15 ( 64.13%) no-bandwidth: 0:03:00 ( 30.74%) no-dumpers: 0:00:30 ( 5.12%)

 1 dumper busy  :  0:00:00  (  0.00%)

as you can see, amanda is doing estimates, but then doesn't do dumps. what does this error message "(aborted:[request failed: timeout waiting for ACK])(too many dumper retry)" mean really ?

Are you really really sure the estimates were done on the client,and
not on the server?

yes i am, i saw the processes running on the client.

The server side dumper sends a request to each client, and does not
receive an ACKnowkledge packet from that client within a reasonable
time.

ok thanks for the explanation.

I would take a look in the debug files on the client, usually in
the dir /tmp/amanda/.  There you can see files "amandad.datetime.debug"
which contain the packet received, and the replies.  If there
are no debug files, than amandad isn't even started.

amandad: version 2.4.5
amandad: build: VERSION="Amanda-2.4.5"
amandad:        BUILT_DATE="Thu Feb 23 10:10:42 CET 2006"

[...]
amandad: time 30.002: dgram_recv: timeout after 30 seconds
amandad: error receiving message: timeout
amandad: time 30.002: error receiving message: timeout
amandad: time 30.002: pid 20533 finish time Thu Feb 23 14:19:44 2006

Does amcheck pass all tests?

yes:

amanda@amanda:~$ amcheck hourly
Amanda Tape Server Host Check
-----------------------------
Holding disk /holding: 264416 MB disk space available, using 263392 MB
read label `hourly023', date `20060223'
read label `hourly024', date `X'

NOTE: skipping tape-writable test
Tape hourly024 label ok

NOTE: it will be created on the next run.
Server check took 0.167 seconds

Amanda Backup Client Hosts Check
--------------------------------
Client check: 1 host checked in 0.868 seconds, 0 problems found

(brought to you by Amanda 2.5.0b2)

bye
Stefan Herrmann