Usage of PEA0

[Sentosa]

Hi All,

I am using OpenVMS 7.3.2 with cluster (3 machines) & found error count on device PEA0.

Could anyone help to answer my question?

1. I understand PEA0 is a cluster port emulator with driver (PEDRIVER). Is it correct?

2. Which type of port (FDDI or Ethernet) for PEA0 to emulate?

3. What is the purpose or action for PEA0 in the system?

Thanks,
Sentosa

[Karl Rohwedder]

PEA0 handles the cluster messages over LAN connections, depending on the network quality some errors may be normal. You can check state and errors of the conections to your different cluster members with the SCACP utility.

regards Kalle

[Jan van den Ende]

Sentosa,

2) PEDRIVER will use any and all interconnects, (unless explicitly blocked in SCACP).

3) purpose is simple in description (although not in functionality!) : take care of cluster communication.

hth

Proost.

Have one on me.

jpe

[Albert Öttl]

Sentosa,
ad 2) AFAIK the PEDRIVER emulates a CI port.

LG
Albert

[Sentosa]

Thanks.

Could anyone help to explain the following message? (extracted from operator log)

Could you tell me where i can find more information for this message?

Message from user SYSTEM on xxxxxx
Event: Excessive Collision from: Node LOCAL:.xxxxxx CSMA-CD Station CSMACD-1,
at: xxxx-xx-xx-20:18:04.275+08:00Iinf
eventUid 7869CDA4-3C52-11DB-94AC-00062B029C57
entityUid 017997C8-3A73-11DB-8505-AA0004004F70
streamUid 05F21E9B-3A73-11DB-85CB-AA0004004F70

Regards,
Sentosa

[Karl Rohwedder]

It's DECnet reporting too many collisions on one interface. Check with
$ MCR LANCP SHO DEV /COU
which device is the culprit. Either you have a real network problem (interface, cable, switch) or perhaps just a mismatch between the interface/switch settings (10/100, Half/Full duplex).

regards kalle

[Albert Öttl]

Hi Sentosa,
this is a message from Decnet/OSI regarding
the ethernet interface, which suffers from
too many collisions.

These collisions can only occur, if the interface
is in 10 Megabit half duplex mode.

Please look at the characteristics and counters
in LANCP.

mc lancp
LANCP> sho config

LAN Configuration:
Device Parent Medium/User Version Speed Size LAN Address
------ ------ ----------- ------- ----- ---- -----------------
EWA0 CSMA/CD 02000022 100 1500 00-06-2B-00-FB-88

LANCP> sho dev ewa0/counter

Device Counters EWA0:
Value Counter
----- -------
949233 Seconds since last zeroed
71140242287 Bytes received
453496877148 Bytes sent
580852297 Packets received
834601256 Packets sent
5870321791 Multicast bytes received
727413246 Multicast bytes sent
26284287 Multicast packets received
3330062 Multicast packets sent
0 Unrecognized unicast destination packets
2016665 Unrecognized multicast destination packets
78 Unavailable station buffers ( 9-OCT-2006 23:54:33.61)
12148 Unavailable user buffers (16-OCT-2006 13:07:27.16)
0 Alignment errors
0 Frame check errors
0 Frame size errors
0 Frame status errors
0 Frame length errors
0 Frame too long errors
0 Data overruns
0 Send data length errors
0 Receive data length errors
0 Transmit underrun errors
0 Transmit failures
0 Carrier check failures
0 Station failures
0 Initially deferred packets sent
0 Single collision packets sent
0 Multiple collision packets sent
0 Excessive collisions
0 Late collisions
0 Collision detect check failures
0 Link up transitions
0 Link down transitions
None Time of last generic transmit error
None Time of last generic receive error

[Richard Brodie_1]

Ethernet automatically retries each packet transmission in hardware after a collision, up to 16 times before reporting a failed transmission back.

On a correctly functioning network, these should be rare. If you are seeing a lot of these, you likely have a problem: malfunctioning hardware, incorrect configuration, excessive load.

As Albert suggested, looking at the network error counters is often quantitive way to start troubleshooting. The counters from each interface and a diagram of the network topology is where I would want to start from.

[comarow]

Starting with the really obvious, if you have a system that is not having these problems, can you try switching to that nodes network connection (port, cable)?

[Sentosa]

thanks