Amanda-Users

Re: amdump fails with badly formatted response from

2004-05-26 07:59:19
Subject: Re: amdump fails with badly formatted response from
From: Paul Bijnens <paul.bijnens AT xplanation DOT com>
To: Stefan Noll <s.noll AT metrix DOT de>
Date: Wed, 26 May 2004 13:54:58 +0200
Stefan Noll wrote:

DISKENTRY KLINGENBERG
sendsize[16506]: time 752.530: calculating for amname '/var/samba/projekte/Laufende/Klingelnberg', dirname '/var/samba/ sendsize[16506]: time 752.530: getting size via gnutar for /var/samba/projekte/Laufende/Klingelnberg level 0
sendsize[16506]: time 752.532: spawning /usr/lib/amanda/runtar in pipeline
sendsize[16506]: argument list: /bin/tar --create --file /dev/null --directory /var/samba/projekte/Laufende/Klingelnber
sendsize[16121]: time 752.536: waiting for any estimate child: 1 running

Are you very sure this was correctly copy/pasted?  The lines
seem to be broken off.

Another thing to explore is the fact you have over 100 entries for
that host.
Amanda communicates the estimates in one single udp dgram wich has
a payload a little less then 65535 bytes:
#define MAX_DGRAM      (((1<<16)-1)-24-8)
My calculations do not come yet near that size, unless you have
ridiculous long directory names.  Or you hit some other bug in sending
large UDP packets.

Actually you can find the dgram packet in one of the amandad.XXX.debug
files on the client.  I guess the name is correct in that place?
It looks like:

=======
...
SERVICE sendsize
...
GNUTAR /the/am/name /the/dir/name 4 2004:5:19:22:6:8 1 OPTIONS |;auth=bsd;compress-fast;index;exclude-list=/usr/lib/amanda/exclude.gtar;include-file=./blabla;
...
... (requesting each disklistentry)
...
amandad: time 222.547: sending REP packet:
----
Amanda 2.4 REP HANDLE 000-40B20608 SEQ 1085436125
OPTIONS features=fffffeff9ffe0f;
/dle/number/one 0 SIZE 5390
/dle/number/one 1 SIZE 10
... and so one...
======

Verify the request and reply line for that disk (just before/after too).

When server receives the reply packet with the estimates,
you can see in amdump.1 the result like:

planner: time 80.851: got result for host neptun disk /: 0 -> 457710K, 1 -> 2980K, -1 -> -1K

Is the name now mangled to "projojeke"?

You get the "badly formatted response from" error when one of these
things happen:
- The datagram line could not be parsed by
      (sscanf(t - 1, "%d SIZE %ld", &level, &size) != 2)
- The estimate was for a level that was not requested

Which one would it be?

--
Paul Bijnens, Xplanation                            Tel  +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUM    Fax  +32 16 397.512
http://www.xplanation.com/          email:  Paul.Bijnens AT xplanation DOT com
***********************************************************************
* I think I've got the hang of it now:  exit, ^D, ^C, ^\, ^Z, ^Q, F6, *
* quit,  ZZ, :q, :q!,  M-Z, ^X^C,  logoff, logout, close, bye,  /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt,  abort,  hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e,  kill -1 $$,  shutdown, *
* kill -9 1,  Alt-F4,  Ctrl-Alt-Del,  AltGr-NumLock,  Stop-A,  ...    *
* ...  "Are you sure?"  ...   YES   ...   Phew ...   I'm out          *
***********************************************************************