[OpenAFS] openAFS over WAN

Jayme Cox jayme@reality.net
Fri, 15 Nov 2002 11:29:34 -0800


I am using openAFS between two Linux machines across the internet. 
Initial setup works just fine. Filesystem can be viewed, everyone is happy.

However, sometime later, the connection just goes away. Here is the info 
in FileLog:

Fri Nov 15 01:31:21 2002 ProbeUuid failed for host 66.214.152.42:20114
Fri Nov 15 02:37:24 2002 ProbeUuid failed for host 66.214.152.42:20279
Fri Nov 15 02:58:27 2002 ProbeUuid failed for host 66.214.152.42:20337
Fri Nov 15 04:04:30 2002 ProbeUuid failed for host 66.214.152.42:10001
Fri Nov 15 04:10:34 2002 ProbeUuid failed for host 66.214.152.42:10005
Fri Nov 15 04:26:37 2002 ProbeUuid failed for host 66.214.152.42:10008
Fri Nov 15 06:37:40 2002 ProbeUuid failed for host 66.214.152.42:10008
Fri Nov 15 10:28:37 2002 CB: RCallBackConnectBack (host.c) failed for 
host 66.214.152.42:10008

It stopped working at around 4:00 am

Here is the output from rxdebug -a

Trying 65.209.194.44 (port 7000):
Free packets: 581, packet reclaims: 490, calls: 33811, used FDs: 64
not waiting for packets.
0 calls waiting for a thread
11 threads are idle
rx stats: free packets 581, allocs 120809, alloc-failures(rcv 0/0,send 
0/0,ack 0)
    greedy 0, bogusReads 0 (last from host 0), noPackets 0, noBuffers 0, 
selects 0, sendSelects 0
    packets read: data 48174 ack 12165 busy 0 abort 34 ackall 0 
challenge 57 response 30 debug 670 params 0 unused 0 unused 0 unused 0 
version 0
    other read counters: data 48174, ack 12164, dup 2029 spurious 1 dally 0
    packets sent: data 44768 ack 29102 busy 0 abort 51 ackall 20 
challenge 30 response 57 debug 0 params 0 unused 0 unused 0 unused 0 
version 0
    other send counters: ack 29102, data 98716 (not resends), resends 
1917, pushed 0, acked&ignored 15282
         (these should be small) sendFailed 0, fatalErrors 0
    Average rtt is 0.039, with 8001 samples
    Minimum rtt is 0.000, maximum is 0.838
    4 server connections, 7 client connections, 15 peer structs, 9 call 
structs, 8 free call structs
    0 clock updates
Showing only connections from host 66.214.152.42
Connection from host 66.214.152.42, port 10170, Cuid a86dc761/f77bece0
   serial 404,  natMTU 1444, flags, security index 0, server conn
     call 0: # 17, state precall, mode: eof, flags: receive_done, 
has_input_packets
     call 1: # 0, state not initialized
     call 2: # 0, state not initialized
     call 3: # 0, state not initialized
Connection from host 66.214.152.42, port 10170, Cuid 953e24c6/f74a3510
   serial 3,  natMTU 1444, security index 0, client conn
     call 0: # 3, state not initialized
     call 1: # 0, state not initialized
     call 2: # 0, state not initialized
     call 3: # 0, state not initialized
Done.

Also, while doing a tcpdump, I noticed this:

00:25:57.020265 65.209.194.44.7000 > 66.214.152.42.20114:  [udp sum ok] 
rx abort seq 0 ser 3 fs reply get-time error #-1 (32) (DF) (ttl 64, id 
0, len 60)

Anyone have any ideas what is going on or a way to track down the problem?

Thanks!

--Jayme