Access limitations(?), and moving to a simple LAN

In article <c643b5b64d.druck@xxxxxxxxxxxxxxxx>, druck
<news@xxxxxxxxxxxxxxxx> wrote:
> On 8 Oct 2005 Jeremy C B Nicoll <Jeremy@xxxxxxxxxxxxxxxx> wrote:

> > When we tried enabling TCP/IP in both machines (via Configure) with
> > suitable definitions in place, both machines were able to ping the
> > other. When we were doing this I removed my normal ANT Suite
> > Internet stuff from the RPC's boot-up, so only the Configure
> > definitions/ processing were in place.

> That sounds good so far. The ANT suite only complicates issues.

I can cope ok with the idea that it won't be on the Iyonix, that is,
that I'll be setting that machine up with software that I'm not
familiar with, from scratch.

> > But, although Access was turned on in both, and - on the surface
> > appeared to work - eg a "Discs" icon was visible on both machines'
> > icon bars, actually trying to use Access didn't work. The Iyonix
> > couldn't see shares defined on the RPC. On the RPC we saw aborts,
> > disappearance of the "Discs" icon, and on subsequent attempts I've
> > also seen sections of white-on-black writing appear at the top of
> > the GUI display. The writing contains snippets of words like
> > "ShareFS" and "exec", and F12-enter doesn't make it go away (or
> > perhaps it goes, but it comes back).

> Check what versions of the internet module, the ethernet driver and
> sharefs & freeway modules. If using the ANT suite you may not have
> updated these within !System.310.Modules from versions held within
> the ANT Suite.

OK, I just rebooted the RPC, at this stage without it having TCP/IP
enabled.

Internet (in podule 8 - the NIC?) is unplugged
(in system rom) is active, V5.02

ShareFS (in system rom) is active, V3.40

Freeway (in system rom) is active, V0.26

EtherH (in podule 8) is active, V4.18

After starting the ANT Suite, to post this, none of that changes.

The NIC is:

DCI Version 4

Card Info:- i-cubed ltd, EtherLan 600 Ethernet interface
Unit 0, slot 8, 16-bit driver.
Ethernet address=00:c0:32:00:9f:d3
10BaseT (twisted pair) interface in use.

I/O Stats:- Rxframes=0, Rxerrs=0, Txframes=41, Txerrs=0,
DMAframes=0
Collisions=0, Rejects=0

Interface:- Twisted pair interface:- link is bad !!
Frame Info:- Type=8035 FrmLvl=1 AddrLvl=2 ErrLvl=0 Handler=3a4d3ac
Type= 806 FrmLvl=1 AddrLvl=2 ErrLvl=0 Handler=3a4d3ac
Type= 800 FrmLvl=1 AddrLvl=2 ErrLvl=0 Handler=3a4d3ac

Controller Mode:- Multicast, reject frames with errors

Driver Info:- Unit 0: CD/Heartbeat failure (is the AUI cable
connected?)

With TCP/IP configured on in boot the module versions are the same.

> > Incidentally, can one turn Access on/off without rebooting both
> > machines? Or RMKill & RMReInit something or other at each end and
> > expect that to work?

> Yes, but best to concentrate on getting things to work with the
> normal procedure before getting in to module reinitialisation
> ordering.

OK; I was just wondering if there was a quicker option than rebooting.


[snip]

> A machine can have any anumber of interfaces each with a different
> address, as long as the routing tables are updated so the internet
> module knows which one to use for any particular address.

Ah, that must be the *route command... Is there a parameter that
displays what the routing tables contain at any particular instant?



> > I know that it's possible to define both dial-up and LAN-based
> > connections in the ANT Suite's configuration dialogues, though I
> > don't know what that actually does, nor how one tells the Suite
> > which connection to actually use, assuming that it routes traffic
> > over one or the other but not both.

> The routing tables will be updates by the dialup software, which on
> connection will set the default route to the modem so everything goes
> to there except traffic for the local network.
>
> > Apart from the transition time, I think I'd like to keep the
> > ability to dial into Demon (and maybe another ISP I have an account
> > with) as a backup/alternate way of sending news/mail for times that
> > the TW cable connection isn't working. I don't require that to be
> > on the RPC, necessarily - is there a way to define both a broadband
> > and dial-up connections on the internet software on the Iyonix?

> Yes, you just set up the networking so that the broadband router is
> the default gateway. All non local traffic will then be sent to it.
> If you then run some dialup software (such as !Dialup) that will set
> the default route to the modem so that gets used in preference. On
> hanging up it should restore the default so broadband gets used again.

Would it make more sense to shift the dial-up modem from the RPC to the
Iyonix then, and then only have !DialUp on the Iyo?

Then, presumably, the RPC and any other machines on the LAN would only
need to have hosts files, their own IP address, gateway address and
subnet mask defined?

[ I must admit I'm confused about which of the RComp packages I need -
is there some software needed to manage the broadband connection as
well? Eg, is !Netfetch necessary? Or is it "just" a bundling of
POPstar, Newshound etc - I'd rather have separate apps and learn how
they fit together, I think. I think I'll stick to Pluto as a news/mail
client.]

> My Iyonix and RPC are connected on a local network to a broadband
> router, and sharing discs via access. Occationally I run !DialUp on
> the Risc PC to log in to my backup ISPs network so that the account
> stays alive. So all you want to do is possible.

Good!

Since writing that lot last night, I've experimented a bit more. The
ShareFS problem on the RPC is consistent. I set up a share of one of
the RPC's disks, then select-click on "Disks" to get the filer view
window, which shows that disk available. Click that and the disk's
shared icon arrives on the icon bar. Click that and I get an error box
and the shared disk icon I just clicked on vanishes from the icon bar
(and isn't replaced by the Disks one). Reporter shows:

22:54:26.41 [IDEFSFiler\IDEFSFiler] Share IDEFS::Dug -protected
22:54:34.13 [ShareFS\ShareFS] Filer_OpenDir Resources:$.Discs 200 300
-LargeIcons
22:54:39.01 ** WimpError ** from ShareFS
Error : &00000298
Message: Illegal task handle
22:54:39.01 [ShareFS\ShareFS] Exec


I've got each machine connected to a switch, and I've plugged the
router into the switch as well.

My router comes with address 192.168.1.1 so I've changed the IP
addresses of both machines. The hosts files on each now contain:

192.168.1.1 router
192.168.1.101 iyonix
192.168.1.201 riscpc
192.168.1.255 jnall

and from each machine I can ping all three devices. When I use the
broadcast address the response from ping is a bit puzzling:

(from the RPC):

*ping jnall
PING jnall (192.168.1.255): 56 data bytes
64 bytes from 192.168.1.201: icmp_seq=0 ttl=255 time=0 ms
64 bytes from 192.168.1.255: icmp_seq=0 ttl=255 time=0 ms (DUP!)
64 bytes from 192.168.1.101: icmp_seq=0 ttl=255 time=90 ms (DUP!)
64 bytes from 192.168.1.201: icmp_seq=1 ttl=255 time=0 ms
64 bytes from 192.168.1.101: icmp_seq=1 ttl=255 time=0 ms (DUP!)
64 bytes from 192.168.1.255: icmp_seq=1 ttl=255 time=10 ms (DUP!)
64 bytes from 192.168.1.201: icmp_seq=2 ttl=255 time=0 ms
64 bytes from 192.168.1.255: icmp_seq=2 ttl=255 time=10 ms (DUP!)
64 bytes from 192.168.1.101: icmp_seq=2 ttl=255 time=10 ms (DUP!)

--- jnall ping statistics ---
3 packets transmitted, 3 packets received, +6 duplicates, 0% packet loss
round-trip min/avg/max = 0.000/13.333/90.000 ms
*

It's the "DUP" that surprises me, unless... the data that each device
sends back in response to a ping is identical? And if it is, how does
the issuing ping know which response data is which?

--
Jeremy C B Nicoll - my opinions are my own.
.

Relevant Pages

  • Re: Workgroup PCs cant see each other
    ... > remove the internet connection so that the only things plugged in are ... I've tried using different network cards ... > loses connectivity when I try to ping it. ... The machines are completely blind to ...
    (microsoft.public.windowsxp.network_web)
  • Re: Security issue?
    ... "Jorge Silva" wrote: ... If you ping the server by FQDN how much time to get response? ... connection took place the tools subsequently do connect/open ok - although ... machines to the domain with no change. ...
    (microsoft.public.windows.server.active_directory)
  • Re: Gateway address needed
    ... connection problems. ... Ah, well, if it's your own network then you're the admin, do what you ... Workgroup name - just fine - same as other machines. ... What does a 'ping 192.168.2.1' say ...
    (alt.internet.wireless)
  • Re: Security issue?
    ... If you ping the server by FQDN how much time to get response? ... connection took place the tools subsequently do connect/open ok - although ... Running dsa.msc as you suggested also opens but somewhat ... machines to the domain with no change. ...
    (microsoft.public.windows.server.active_directory)
  • Access limitations(?), and moving to a simple LAN
    ... Access was enabled in both machines, and TCP/IP left turned off in both. ... On the RPC we saw aborts, ... I've a TeleWest cable broadband connection, hardly ever used, here and ...
    (comp.sys.acorn.networking)