On Mon, 19 May 2003 at 7:59am, Rebecca Pakish Crum wrote

> Okay, this is getting ridiculous. A couple of weeks ago this client was
> FINE. I now have etimeout set to 1800 and dtimeout set to 1800. I'm
> still getting:
> sending ack:
> ----
> Amanda 2.4 ACK HANDLE 002-A0AB0708 SEQ 1053160202
> ----
> 
> bsd security: remote host web.isymmetrics.com user amanda local user
> amanda
> amandahosts security check passed
> amandad: running service "/usr/local/libexec/sendsize"
> amandad: sending REP packet:
> ----
> Amanda 2.4 REP HANDLE 002-A0AB0708 SEQ 1053160202
> OPTIONS maxdumps=1;
> / 0 SIZE 5469830
> / 1 SIZE 622920
> / 2 SIZE 412490
> ----
> 
> amandad: dgram_recv: timeout after 10 seconds
> amandad: waiting for ack: timeout, retrying
> amandad: dgram_recv: timeout after 10 seconds
> amandad: waiting for ack: timeout, retrying
> amandad: dgram_recv: timeout after 10 seconds
> amandad: waiting for ack: timeout, retrying
> amandad: dgram_recv: timeout after 10 seconds
> amandad: waiting for ack: timeout, retrying
> amandad: dgram_recv: timeout after 10 seconds
> amandad: waiting for ack: timeout, giving up!
> amandad: pid 5432 finish time Sat May 17 02:28:50 2003

I assume this is a /tmp/amanda/amandad*debug from the failnig client?  It 
looks like it's sending the estimates, but not hearing back from the 
server.  If that's the case, then I think the possiblities are:

1) The server is no longer listening for a response, i.e. the client took
   too long to do the estimates.  What time do you kick off your amdump?  
   Is it more than 1800 seconds before 2:28 AM?

2) The server can't hear the response from the client becuase of firewall 
   issues.  You said the server is RH8 -- make double/triple sure that you
   don't have ipchains or iptables running.  IIRC, RH8 *still* uses 
   ipchains.  Look at 'chkconfig --list', 'ipchains -nL', and the 
   contents of /etc/sysconfig/ipchains (adjusting as appropriate for
   iptables if I'm wrong about that).

3) Any sort of firewalling on the client side -- I know diddly about 
   Solaris.

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University