[Coco] Telnet and DW
Aaron Wolfe
aawolfe at gmail.com
Wed Dec 30 22:48:21 EST 2009
OK, that error is happening in a routine that pulls the disk name and
some other info out of .dsk files. It's purely cosmetic at this
point. Maybe just comment out line 171 of DWDiskDrives.java It
won't have any effect on the server. If you want to email me your
.dsk file I'll try to figure out why that routine can't grok it.
-Aaron
On Wed, Dec 30, 2009 at 10:31 PM, Christopher Hawks <chawks at dls.net> wrote:
> Aaron Wolfe said the following on 12/30/2009 07:49 PM:
>>
>> I think you might have things a bit out of sync between server version
>> and driver version.. we've done an awful lot of commits in the last
>> couple days, its easy to do.
>>
>> The SERSETSTAT line there would not look like that in a current
>> version. You would instead see a SERINIT.
>>
>> I suspect the driver is sending slightly different opcodes/arguments
>> than what the server understand. If it's not too much trouble, I
>> think udpating everything from CVS on drivewire and nos9 sides will
>> help.
>
> Yup, that made a BIG difference!
>
> After I updated NitrOS9 ahd Drivewireserver, I had to re-write the
> dwserver.dsk makefile (now in 3rdparty/packages/drivewire) 'cause it only
> built 6309 code and I don't have one.
>
> I compiled and ran Drivewireserver and I get this:
> HAWKSoft(chrish)$ make run
> java -cp
> ../../src/lib/jython.jar:../../src/lib/chatengine.jar:../../src/lib/RXTXcomm.jar:../../src/lib/commons-collections-3.2.1.jar:../../src/lib/commons-configuration-1.6.jar:../../src/lib/commons-lang-2.4.jar:../../src/lib/commons-logging-1.1.1.jar:../../src/lib/commons-net-2.0-javadoc.jar:../../src/lib/commons-net-2.0.jar:../../src/lib/commons-net-ftp-2.0.jar:../../src/lib/log4j-1.2.15.jar:../../src/lib/miglayout-3.7.1-swt.jar:../../src/lib/ostermillerutils_1_07_00.jar:../../src
> com.groupunix.drivewireserver.DriveWireServer
> 1 [main] DEBUG org.apache.commons.configuration.ConfigurationUtils -
> ConfigurationUtils.locate(): base is null, name is
> DriveWireServer.properties
> 70 [main] DEBUG org.apache.commons.configuration.ConfigurationUtils -
> Loading configuration from the path DriveWireServer.properties
> 30 Dec 2009 21:26:11 INFO [main ] rver.DriveWireServer: DriveWire
> Server 3.1.2 (12/21/2009) starting up
> 30 Dec 2009 21:26:11 INFO [main ] handler.DWDiskDrives: loading
> diskset 'default.set'
> 30 Dec 2009 21:26:11 DEBUG [main ] otocolhandler.DWDisk: created memmap
> (using 4718592 bytes)
> 30 Dec 2009 21:26:11 DEBUG [main ] otocolhandler.DWDisk: set filepath
> to '/home/chrish/dwserver.dsk'
> 30 Dec 2009 21:26:11 INFO [main ] handler.DWDiskDrives: loaded disk
> '/home/chrish/dwserver.dsk' in drive 0
> Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 32
> at
> com.groupunix.drivewireserver.dwprotocolhandler.DWDisk.cocoString(DWDisk.java:177)
> at
> com.groupunix.drivewireserver.dwprotocolhandler.DWDisk.diskInfo(DWDisk.java:153)
> at
> com.groupunix.drivewireserver.dwprotocolhandler.DWDiskDrives.LoadDisk(DWDiskDrives.java:171)
> at
> com.groupunix.drivewireserver.dwprotocolhandler.DWDiskDrives.LoadDiskSet(DWDiskDrives.java:75)
> at
> com.groupunix.drivewireserver.DriveWireServer.main(DriveWireServer.java:99)
> make: *** [run] Error 1
>
> The regular drivewire build nos96809l2v030209coco3_dw3.dsk will load,
> but it won't boot. 8^(
>
>> On Wed, Dec 30, 2009 at 8:47 PM, Christopher Hawks <chawks at dls.net> wrote:
>>>
>>> Aaron Wolfe said the following on 12/30/2009 07:14 PM:
>>>>
>>>> You're very close :)
>>>>
>>>> Can you send me the output (or the gist of it) of: dw show ports
>>>
>>> dw show ports on the CoCo locked it up.
>>> Server log sez:
>>> 30 Dec 2009 19:26:28 INFO [dwproto-7 ] er.DWProtocolHandler:
>>> DoOP_SERSETSTAT for port /T0 stat: 0
>>> 30 Dec 2009 19:26:28 INFO [dwproto-7 ] erial.DWVSerialPorts: write to
>>> non
>>> init port /T0, doing init then write
>>> 30 Dec 2009 19:26:28 INFO [dwproto-7 ] erial.DWVSerialPorts: init
>>> virtual
>>> serial port /T0 mode 0
>>> 30 Dec 2009 19:26:28 DEBUG [dwproto-7 ] serial.DWVSerialPort: init port
>>> /T0
>>> 30 Dec 2009 19:27:00 INFO [dwproto-7 ] er.DWProtocolHandler: DoOP_TIME
>>>
>>>> Also you should see some log entries as the tsmon starts referring to
>>>> the port being initialized. Also, what's in your ports.set?
>>>
>>> Saw them early on:
>>> 30 Dec 2009 18:18:34 INFO [main ] erial.DWVSerialPorts: loading
>>> portset 'ports.set'
>>> 30 Dec 2009 18:18:34 INFO [main ] erial.DWVSerialPorts: init
>>> virtual
>>> serial port /T0 mode 0
>>> 30 Dec 2009 18:18:34 DEBUG [main ] serial.DWVSerialPort: init port
>>> /T0
>>> 30 Dec 2009 18:18:34 INFO [main ] erial.DWVSerialPorts: Port 0 has
>>> action file: login.action
>>> 30 Dec 2009 18:18:34 INFO [main ] erial.DWVSerialPorts: init
>>> virtual
>>> serial port /T1 mode 0
>>>
>>> Server Inited /T0 - /T7
>>>
>>> ports.set sez:
>>> 0,0,,login.action
>>> 1,0,,login.action
>>> 2,0,,login.action
>>> 3,0,,login.action
>>> 4,0,,login.action
>>> 5,0,,login.action
>>> 6,0,,login.action
>>> 7,0,,login.action
>>>
>>> login.action sez:
>>> from com.groupunix.drivewireserver.virtualserial import DWVSerialPorts
>>> DWVSerialPorts.write(port, "\r")
>>>
>>>
>>>> -Aaron
>>>>
>>>> On Wed, Dec 30, 2009 at 7:44 PM, Christopher Hawks <chawks at dls.net>
>>>> wrote:
>>>>>
>>>>> Well, I thought I could run it.
>>>>> I
>>>>> updated to the latest nitrOS9 source (compiled it)
>>>>> updated to the latest Drivewire Server (compiled the Java)
>>>>> moved librxtxSerial.so to where the server could find it
>>>>> edited DriveWireServer.properties and default.set
>>>>> ran "make run" to start Drivewire Server
>>>>> Booted nos96809l2v030209coco3_dw3.dsk from the server
>>>>> ran startup.dw to start tsmon on the virtual serial ports
>>>>>
>>>>> CoCo3 runs NitrOS9 fine with drivewire, but, I can't log in.
>>>>>
>>>>> "telnet HAWKSoft 6809" returns:
>>>>> Trying 192.168.0.102...
>>>>> Connected to hawksoft.
>>>>> Escape character is '^]'.
>>>>> DriveWire TCP Server 3.1.2
>>>>>
>>>>> No ports available.
>>>>> Connection closed by foreign host.
>>>>> HAWKSoft(chrish)$
>>>>>
>>>>> Server's log reports:
>>>>> 30 Dec 2009 18:41:17 INFO [tcpsrv-8 ] cpserver.DWTCPServer: new
>>>>> connection
>>>>> from /192.168.0.102
>>>>> 30 Dec 2009 18:41:17 DEBUG [tcpconn-17] er.DWTCPServerThread: thread
>>>>> run
>>>>> for
>>>>> hawksoft:53544
>>>>> 30 Dec 2009 18:41:17 DEBUG [tcpconn-17] er.DWTCPServerThread: no ports
>>>>> available for new client at hawksoft:53544
>>>>> 30 Dec 2009 18:41:17 DEBUG [tcpconn-17] er.DWTCPServerThread: closing
>>>>> socket
>>>>> 30 Dec 2009 18:41:17 DEBUG [tcpconn-17] er.DWTCPServerThread: thread
>>>>> exiting
>>>>>
>>>>> "telnet localhost 6809" acts the same.
>>>>>
>>>>> Did I miss something??
>>>>>
>>>>>
>>>>> --
>>>>> Christopher R. Hawks
>>>>> HAWKSoft
>>>>> ---------------------------------------------------------
>>>>> "The strongest test of any system is not how well its features conform
>>>>> to
>>>>> anticipated needs but how well it performs when one wants to do
>>>>> something
>>>>> the designer did not forsee."
>>>>> -- Alan Kay, Xerox PARC
>>>>>
>>>>> --
>>>>> Coco mailing list
>>>>> Coco at maltedmedia.com
>>>>> http://five.pairlist.net/mailman/listinfo/coco
>>>>>
>>>> --
>>>> Coco mailing list
>>>> Coco at maltedmedia.com
>>>> http://five.pairlist.net/mailman/listinfo/coco
>>>>
>>>
>>> --
>>> Christopher R. Hawks
>>> HAWKSoft
>>> ---------------------------------------------------------
>>> I develop for Linux for a living, I used to develop for DOS. Going from
>>> DOS to Linux is like trading a glider for an F117.
>>> -- Lawrence Foard
>>>
>>> --
>>> Coco mailing list
>>> Coco at maltedmedia.com
>>> http://five.pairlist.net/mailman/listinfo/coco
>>>
>>
>> --
>> Coco mailing list
>> Coco at maltedmedia.com
>> http://five.pairlist.net/mailman/listinfo/coco
>>
>
>
> --
> Christopher R. Hawks
> HAWKSoft
> ---------------------------------------------------------
> "If Java had true garbage collection, most programs would delete
> themselves upon execution." -- Robert Sewell
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco
>
More information about the Coco
mailing list