Network Adapter question in VMS

[Jorge Cocomess]

Greetings,

How do I reference which NIC adapter that I am currently using? Information on the NIC as follows:

LANCP:
Device Characteristics EWA0:
Value Characteristic
----- --------------
1500 Device buffer size
Normal Controller mode
External Internal loopback mode
00-00-F8-07-41-BC Hardware LAN address
Multicast address list
CSMA/CD Communication medium
FF-FF-FF-FF-FF-FF Current LAN address
128 Minimum receive buffers
256 Maximum receive buffers
No Full duplex enable
No Full duplex operational
00000007 Line media type
100 Line speed (mbps)
Disabled/No Failset Logical LAN state
0 Failover priority

Device Characteristics EWB0:
Value Characteristic
----- --------------
1500 Device buffer size
Normal Controller mode
External Internal loopback mode
00-10-18-10-A4-B8 Hardware LAN address
Multicast address list
CSMA/CD Communication medium
FF-FF-FF-FF-FF-FF Current LAN address
128 Minimum receive buffers
256 Maximum receive buffers
Yes Full duplex enable
Yes Full duplex operational
TwistedPair Line media type
1000 Line speed (mbps)
Disabled/No Failset Logical LAN state
0 Failover priority

TCPIP
sho interface /full
Interface: LO0
IP_Addr: 127.0.0.1 NETWRK: 255.0.0.0 BRDCST: 192.168.255.255
MTU: 4096
Flags: UP LOOP NOARP MCAST SMPX
RECEIVE SEND
Packets 28259 28259
Errors 0 0
Collisions: 0

Interface: WE1
IP_Addr: 192.168.1.15 NETWRK: 255.255.0.0 BRDCST: 192.168.255.255
Cluster
C_Addr: 192.168.1.15 C_NETWRK: 255.255.0.0 C_BRDCST: 192.168.255.255
Ethernet_Addr: 00-10-18-10-A4-B8 MTU: 1500
Flags: UP BRDCST RUN MCAST SMPX
RECEIVE SEND
Packets 931348559 532107993
Errors 0 0
Collisions: 0


Thank you in advance.

[Hoff · Accepted solution]

OpenVMS IP and DECnet will attempt to use all links that are found on the system. To see which are in active use on a host, check the link counters.

Within the specific information posted, the IP device interface WE1 is presently using physical device EWB0: for connection into a private class C address space.

LO is the IP loopback device.

EWA0: and EWB0: are two NICs that are present. Based on what is shown, EWA0: isn't used for anything. It's probably not plugged into a network, or it's been otherwise disabled, or there's a problem somewhere upstream of the EWA0: NIC in this network.

(Some systems will have link status fault indications, and will flash an attention indicator light near the failure. On various local servers, it's a bright blue LED right next to the NIC socket, and it automatically flashes when the NIC is not connected or the connection is not working. Unfortunately, I've not seen this implemented on Integrity, Alpha or VAX hardware. This would be a nice enhancement for the Integrity boxes, and would help you find out which NIC socket is -- for instance -- the EWA0: device here. Right now, it's somewhat trial-and-error, assuming there's a wire plugged into each of the NICs on this system. Pull one, see if the connections drop. Or trace the wires.)

[Andy Bustamante]

You have 2 network interfaces, TCPIP is configured to use the second interface. Which adapter depends on what you're doing. For TCPIP traffic, you'll be using the second interface. For DECnet or LAT there isn't enough information to determine which adapter may be in use.

You may want to consider LAN failover or failSAFE IP with two interfaces available. See the System Manager's manual for LAN failover documentation http://h71000.www7.hp.com/doc/os83_index.html and this link from the OpenVMS Technical Journal http://h71000.www7.hp.com/openvms/journal/v2/articles/tcpip.html for discussion on configuring TCPIP for reliable access.

For details on the network adapters you can use:
$ anal/system
> clue config

Depending on the version of your operating system and platform, the error log may inventory NICs (Diagnose or anal/error). You can access this information from the console if you halt the system as well, although that can be a considered an extreme way to get this information.


Andy

[Walter Miller_1]

You might try "show device ew /full" to see which processes are current using the device.

[Ian Miller.]

ANAL/SYS
SDA> SHOW LAN
will show the protocols enabled on each device.

[Jorge Cocomess]

Per your suggestions, I ran ANAL/SYS & Show lan and here's the stat:

-- EWA Device Summary 6-FEB-2007 16:01:37 --

LSB address = 8718A000
Device state = 00001003 RUN,INITED

UCB UCB Addr Fmt Value Client State
--- -------- --- ----- ------ -----------
EWA0 8151B1C0
EWA3 8168BF00 802E 08-00-2B-80-3C DNAME 0017 STRTN,LEN,UNIQ,STRTD
EWA4 8168C700 Eth 80-3C DNAME 0017 STRTN,LEN,UNIQ,STRTD
EWA5 817DDF40 802 FE DECNETV 0017 STRTN,LEN,UNIQ,STRTD
EWA6 817DDC00 Eth 60-03 DECNET 0017 STRTN,LEN,UNIQ,STRTD
EWA7 8192B9C0 Eth 60-04 LAT 0015 STRTN,UNIQ,STRTD


Doesn't look like I'm using EWA0, am I??

[Andy Bustamante]

EWA0 is the template for protocols running on EWA1, EWA2 . . .

So in this examples, you're running DECnet phase V and LAT on the EWA NIC. Are you using either of these for connectivity?


Andy

[Martin Hughes]

As suggested by Hoff;

$ MC LANCP SHOW DEV/COUNT EWA0:

Are your sent & received byte counters zero? if not, are they increasing?.

Also, note the speed differences with your cards;

EWA0: - 100 Line speed (mbps)
EWB0: - 1000 Line speed (mbps)

[Jorge Cocomess]

Yes, device EWA0: displayed all zeroes when I execute the command '$ MC LANCP SHOW DEV/COUNT EWA0:'

When I ran the same command on EWB0: and I get the stats and counters.

That's good news. I did not want to use the 100MBps NIC Adapter any longer. I am now running on the faster NIC EWB0: at 1000MBps.

Thanks guys!!

[Colin Butcher]

A bit more information for you:

DECnet Phase V device (CSMACD station) naming by default is CSMACD-0 (EWA0) and CSMACD-1 (EWA1). If you look at the devices with NCL you'll see which prototype EWx0 physical device the CSMACD-x station maps to. I like to change the default names to something more useful, like ETH_EWA0.

TCPIP device naming for EW devices is WE0 (EWA0) and WE1 (EWB0).

EW devices cover a wide range of hardware types, all the way from 10/100 copper devices to fibre GigE devices. You'll also find EI devices as well (generally the Intel 82559 chipset based ethernet devices).

I suggest that you turn off the protocols you don't need on EWA0 - but be careful!

Don't configure the interface in IP (or remove it with SET CONF NOINT WE0).

Don't configure it in DECnet Phase V (or remove it with DISABLE CSMA STATION and DELETE CSMA STATION and don't forget to edit the NCL scripts or re-configure the devices with NET$CONFGURE in ADVANCED mode).

Stop SCS on EWA with SCACP STOP LAN or the LAVC$STOP_BUS program in SYS$EXAMPLES.

As for which devices are which - the hardware manual for the machine usually tells you, but it doesn't always tell you in what order VMS will find the LAN devices. I'm guessing that this is something like a RX26xx where you have both a GigE port and a 10/100 port. As you rightly show - LANCP gives you a reasonable picture of what's happening and the capabilities of the device.

Other thoughts:
- You may want to look at using LAN failover.
- You may want to look at using IP failsafe if youre primary network activity is IP traffic.

Cheers, Colin.