Coda File System

would two ethernet interfaces wig out a coda server?

From: sean dreilinger <sean_at_durak.org>
Date: Thu, 18 Feb 1999 01:41:37 -0800
hi again: 

thanks very much to everyone who has been writing with suggestions on
making my new coda server work with remote clients. no success with any
of the suggestions so far, but i have another question (sorry!).

the coda server is running on a machine with two ethernet interfaces.
venus client connections coming from the internal, 10.0.0.x network on
eth0 work great, i've connected to it via a couple of local-net client
machines now.

remote venus client connections coming in from the outside world through
the second ethernet interface eth1 can see the coda server OK (cmon
servername:100 looks good) but never get hooked up. have watched venus
fail on several remote client machines that otherwise talk just fine to
the testserver.coda.cs.cmu.edu machine.

does this ring a bell for any coda gurus out there? is some part of the
coda server or related services only interested in eth0?

i had a look on the coda FTP site for the rpc2-literate tcpdump, but the
/pub/tools/ location where this is reputed to live (mentioned in
codalist archives) doesn't seem to be available. 

thanks very kindly if anyone out there has further suggestions -- i'd
really like to get this coda server talking with the outside world!

--sean :-)

[root_at_SERVER /root]# tail -f /vice/srv/SrvLog
Date: Thu 02/18/99

00:02:33 Building callback conn.
00:13:06 Callback failed RPC2_DEAD (F) for ws d8292d82, port 32265
00:13:06 Unbinding RPC2 connection 344535209
00:13:06 Unbinding RPC2 connection 291002591


DMESG on CODA-SERVER after FAILED venus -init

Coda Kernel/Venus communications (module), v5.0-pre1, braam_at_cs.cmu.edu.
coda_psdev_write: downcall, no SB!
coda_read_super: rootfid is (0x7f000000,0x1,0x1)
coda_downcall: opcode 30, no sb!
Failure of coda_cnode_make for root: error -110



sean dreilinger wrote:
> i have coda server and client 5.01 working nicely on the same linux
> (kernel 2.2.1) box.
> 
> when i try to start venus on a remote client machine (same linux kernel
> and coda sources) and direct it to use my coda server, venus just dies.
> the same venus client can connect to the testserver at CMU at will.
> 
> does anyone have any thoughts? some interesting messages from
> /usr/coda/venus.cache/vanus.log are below, along with the 'Killed.'
> conclusion that venus spits out when i try to run it on the remote
> client.
> 
> thanks for any suggestions!
> 
> --sean
> --
>                           sean dreilinger, mlis
>                           mailto:sean_at_durak.org
>                           http://durak.org/sean
> 
> [sean_at_durak:/usr/coda/venus.cache]$ cat venus.log
> 
> [ X(00) : 0000 : 16:01:28 ] Coda Venus, version 5.0.1
> [ X(00) : 0000 : 16:01:28 ] LOGFILE initialized with LogLevel = 0 at Wed
> Feb 17 16:01:28 1999
> [ X(00) : 0000 : 16:01:28 ] ProfInit(): profiling is broken; fix it!
> [ X(00) : 0000 : 16:01:28 ] E StatsInit()
> [ X(00) : 0000 : 16:01:28 ] L StatsInit()
> [ X(00) : 0000 : 16:01:28 ] RecovDataSizes: Log = 0x86f9c, Data =
> 0x21be70
> [ X(00) : 0000 : 16:01:28 ] Recov_AllocateVM: allocated 1000 bytes at
> 20000000
> [ X(00) : 0000 : 16:01:28 ] Recov_AllocateVM: allocated 216000 bytes at
> 21000000
> [ X(00) : 0000 : 16:01:28 ] Recov_DeallocateVM: deallocated 1000 bytes
> at 20000000
> [ X(00) : 0000 : 16:01:28 ] Recov_DeallocateVM: deallocated 216000 bytes
> at 21000000
> [ X(00) : 0000 : 16:01:29 ] E adviceserver::adviceserver:
> AdviceServer
> [ A(13) : 0000 : 16:01:29 ] adviceserver::main()
> [ H(07) : 0000 : 16:01:29 ] HDBDaemon about to sleep on hdbdaemon_sync
> [ D(16) : 0000 : 16:01:35 ] WAITING(SRVRQ):
> [ W(15) : 0000 : 16:01:50 ] NotifyUserOfServerDownEvent
> [ W(15) : 0000 : 16:01:50 ] DiscoRef= 1
> [ D(16) : 0000 : 16:01:50 ] WAIT OVER, elapsed = 14997.0
> [ W(15) : 0000 : 16:01:51 ] ADMON STATS:  DMQ Advice NOT valid. (uid =
> 0)
> [sean_at_durak:/usr/coda/venus.cache]$
> 
> [root_at_durak:/usr/src/redhat/RPMS/i386]# rpm -U
> coda-debug-client-5.0.1-1.i386.rpm
> /etc/services already has new services registered! Good.
> [root_at_durak:/usr/src/redhat/RPMS/i386]# venus-setup SERVER.HOST.NAME
> 20000
> /etc/services already has new services registered! Good.
> [root_at_durak:/usr/src/redhat/RPMS/i386]# venus -init
> 
> Date: Wed 02/17/99
> 
> 16:01:28 /usr/coda/LOG initialized at size 0x86f9c
> 16:01:28 /usr/coda/DATA initialized at size 0x21be70
> 16:01:28 brain-wiping recoverable store
> 16:01:28 loading recoverable store
> 16:01:28 starting VSGDB scan
> 16:01:28        0 vsg entries in table
> 16:01:28        0 vsg entries on free-list
> 16:01:28 starting VDB scan
> 16:01:28        1 vol entries in table (0 MLEs)
> 16:01:28        0 vol entries on free-list (0 MLEs)
> 16:01:28 starting FSDB scan (833, 20000) (25, 75, 4)
> 16:01:28        0 cache files in table (0 blocks)
> 16:01:29        833 cache files on free-list
> 16:01:29 starting HDB scan
> 16:01:29        0 hdb entries in table
> 16:01:29        0 hdb entries on free-list
> 16:01:29 Initial LRDB allocation
> 16:01:30 Venus starting...
> Killed
> [root_at_durak:/usr/src/redhat/RPMS/i386]#

--
                          sean dreilinger, mlis
                          mailto:sean_at_durak.org
                          http://durak.org/sean
Received on 1999-02-18 04:42:47