Amanda-Users

Re: Xinetd not starting amanda

2003-03-12 23:15:30
Subject: Re: Xinetd not starting amanda
From: Adam Smith <adam.smith AT sageautomation DOT com>
To: amanda-users AT amanda DOT org
Date: Thu, 13 Mar 2003 13:31:55 +1030
On Mon, Mar 10, 2003 at 12:10:47PM -0600, Rebecca Pakish Crum said:

I am having some similar problems relating to this thread, in regards to
loading the three services below, and running 'amcheck <config>' on FreeBSD
5.0.  I have installed both amanda-client-2.4.3, and amanda-server-2.4.3.

> Can you do a chkconfig --list and see the amanda services listed under xinetd 
> based services and are they turned on? Do you have amanda, amandaidx and 
> amidxtape files in your xinetd.d directory, and do they all look something 
> like this (respectively):

In regards to xinetd, I have not had any experience.  I looked at the
xinetd.conf man page, and it said the configs outlined below need to be in
the /usr/local/etc/xinetd.conf file.  If they are specified in that file,
xinetd can see them, because when any of the services are broken (ie an
invalid option specified) xinetd spits out errors.

My services are specified exactly as below, except the group is different
and the option 'disable = no' exists for all three.

xinetd appears to load without errors, and I can not see any amanda
processes loaded (but this appears to be the nature of Xinetd.)  I tried to
telnet to my amanda server on ports 10080, 10082 and 10083, but only ports
10082 and 10083 responded (amandaidx and amidxtape.)  Is this the correct
behaviour?  Amandad does not seem to work correctly (unless it's not
supposed to load.)

> service amanda
> {
>     protocol            = udp
>     socket_type         = dgram
>     wait                = yes
>     user                = amanda
>     group               = disk
>     groups              = yes
>     server              = /usr/local/libexec/amandad
> }
> service amandaidx
> {
>     protocol            = tcp
>     socket_type         = stream
>     wait                = no 
>     user                = amanda
>     group               = disk
>     groups              = yes
>     server              = /usr/local/libexec/amindexd
> }
> service amidxtape
> {
>     protocol            = tcp
>     socket_type         = stream
>     wait                = no
>     user                = amanda
>     group               = disk
>     groups              = yes
>     server              = /usr/local/libexec/amidxtaped
> }

These are the results I now get from running amcheck:


[amanda@testies /data/amanda]$ amcheck normal
Amanda Tape Server Host Check
-----------------------------
Holding disk /data/amanda/holdingdisk: 44031578 KB disk space available, that's 
plenty
amcheck-server: slot 1: date X        label tape00 (first labelstr match)
NOTE: skipping tape-writable test
Tape tape00 label ok
NOTE: info dir /data/amanda/normal/curinfo/testies.sageautomation.com: does not 
exist
NOTE: info dir /data/amanda/normal/curinfo/megatron.sageautomation.com: does 
not exist
Server check took 0.147 seconds

Amanda Backup Client Hosts Check
--------------------------------
ERROR: testies.sageautomation.com: [addr 192.168.8.50: hostname lookup failed]
WARNING: megatron.sageautomation.com: selfcheck request timed out.  Host down?
Client check: 2 hosts checked in 30.053 seconds, 2 problems found

(brought to you by Amanda 2.4.3)



The host 'testies.sageautomation.com' exists, and is resolvable (internally.)
The host 'megatron.sageautomation.com' exists and resolves internally, 
but it has no amanda services loaded on it, so its error messages are
correct.

Why am I getting a hostname lookup error for a host that exists when the
host is resolvable?


Thanks for your help.


Regards,

-- 
Adam Smith
Information Technology Officer
SAGE Automation Ltd.

adam.smith AT sageautomation DOT com
http://www.sageautomation.com

<Prev in Thread] Current Thread [Next in Thread>