Coda File System

Client server same machine OK, distant client not OK

From: Craig Hughes <craig_at_panop.com>
Date: Mon, 30 Aug 1999 11:26:45 -0700
I've set up a coda server/SCM on baal.panop.com (204.247.162.10) and can
connect to it using venus on baal.

Now, I have a laptop pookah.panop.com (204.247.162.8 when it's on LAN,
varying IP with dialup), and this machine can successfully connect to the
testserver.  But when I venus-setup baal.panop.com 20000 on pookah, then try
to start venus, I get:

Coda Venus, version 5.2.7
11:14:24 /usr/coda/LOG setup for size 0x8761e
11:14:24 /usr/coda/DATA initialized at size 0x21d878
11:14:24 brain-wiping recoverable store
11:14:24 loading recoverable store
11:14:24 starting VSGDB scan
11:14:24 0 vsg entries in table
11:14:24 0 vsg entries on free-list
11:14:24 starting VDB scan
11:14:24 1 vol entries in table (0 MLEs)
11:14:24 0 vol entries on free-list (0 MLEs)
11:14:25 starting FSDB scan (833, 20000) (25, 75, 4)
11:14:25 0 cache files in table (0 blocks)
11:14:25 833 cache files on free-list
11:14:25 starting HDB scan
11:14:25 0 hdb entries in table
11:14:25 0 hdb entries on free-list
11:14:25 Initial LRDB allocation
11:14:25 Getting Root Volume information...
11:14:26 Venus starting...
<venus process dies>

In /usr/coda/venus.cache/venus.log I see:


[ X(00) : 0000 : 11:14:23 ] Coda Venus, version 5.2.7
[ X(00) : 0000 : 11:14:23 ] LOGFILE initialized with LogLevel = 0 at Mon Aug
30 11:14:23 1999

[ X(00) : 0000 : 11:14:23 ] ProfInit(): profiling is broken; fix it!
[ X(00) : 0000 : 11:14:23 ] E StatsInit()
[ X(00) : 0000 : 11:14:23 ] L StatsInit()
[ X(00) : 0000 : 11:14:24 ] RecovDataSizes: Log = 0x8761e, Data = 0x21d878
[ X(00) : 0000 : 11:14:24 ] Recov_AllocateVM: allocated 1000 bytes at
20000000
[ X(00) : 0000 : 11:14:24 ] Recov_AllocateVM: allocated 218000 bytes at
21000000
[ X(00) : 0000 : 11:14:24 ] Recov_DeallocateVM: deallocated 1000 bytes at
20000000
[ X(00) : 0000 : 11:14:24 ] Recov_DeallocateVM: deallocated 218000 bytes at
21000000
[ X(00) : 0000 : 11:14:25 ] E adviceserver::adviceserver: AdviceServer    

[ A(12) : 0000 : 11:14:25 ] adviceserver::main()

[ H(06) : 0000 : 11:14:25 ] HDBDaemon about to sleep on hdbdaemon_sync

[ D(15) : 0000 : 11:14:31 ] WAITING(SRVRQ):

[ W(14) : 0000 : 11:14:43 ] NotifyUserOfServerDownEvent
[ W(14) : 0000 : 11:14:43 ] DiscoRef= 1

[ D(15) : 0000 : 11:14:43 ] WAIT OVER, elapsed = 12096.0

[ W(14) : 0000 : 11:14:43 ] ADMON STATS:  DMQ Advice NOT valid. (uid = 0)
<venus dies>


Server side, in /vice/SrvLog all I see is:

11:15:12 Building callback conn.

then that's it until the next client connects.


The two machines baal and pookah are running Redhat 6.0 upgraded to 2.2.12
kernel with a variety of other updated redhat RPMs, with
coda-debug-*-5.2.7-1.i386.rpm

I have no other TCP/IP connectivity problems between them.  NFS mounts work
great, etc.

Any ideas?  Seems off that I should be able to connect to my server by
client from the same machine, that I should be able to connect to the test
server from pookah, but that I can't connect to baal from pookah.  The only
thing I can't try, for lack of machines, is to connect to baal from another
client.  Can anyone figure out where I'm going wrong?

C


--
PGP Software <http://www.pgpi.com>
Craig's PGP Key <http://www.panop.org/craig/pgpkey.html>
Get PGP, use it.
Received on 1999-08-30 14:34:03