Visual Productions forum

Author Topic: B-station, Kiosc Touch and LPU-2 on corporate network  (Read 1285 times)

February 18, 2023, 01:22:52 AM
Read 1285 times

Miky

  • Member
  • *
  • Posts: 31
Hello,

few days ago I posted question regarding kiosc touch and indicators on the network with several B-stations. Now I follow up with this installation. Its big corporation network with a lot of switches and subnets where are set a strong rules and high level of secutity protection.

From beggining we fight with functionality of the system which controll lighting in 6 rooms (studios). It consist of LPU-2 as main DMX controller, which is controlled from 5  B-stations and 1 Kiosc touchs placed direct in rooms for local controll, and 1 Kiosc touch in controll room as master controll for all rooms.

Highly secured network with sofisticated switches deny all what is not allowed.. It means that first days nothing was running, then we gradually worked out that the OSC commands go through and the system just works,

Unfortunately, from time to time some B-station freezes and stops responding, only a restart by disconnecting the power supply will help. This happens about 1-3 times a month. We asked the administrator to look into this issue, but unfortunately he has no experience with this specific device. He only provided us with the log he pulled from the switch regarding the communication on the given ports. He discovered that the switch regularly sends a query to the device regarding communication, the so-called handshake, based on which it then sets mutual communication (speed, etc.) At the moment when the B-station stops responding to the switch, it appears like a laptop in sleep mode, when the speed communication will switch to 10Mbits half duplex and then only a restart will help. Attached is document with log, so If you can read something from it, it would help us a lot.

Currently, the administrator has turned off the handshake function on the given ports and fixed the communication to 100Mbits full duplex. We'll see if that helps. Otherwise, we already plan to connect the entire system to a separate simple switch and build a separate network without restrictions, but it is not systemic, because then it will be more difficult to access this network and manage devices.

Best regards, Michal


March 14, 2023, 04:07:35 PM
Reply #1

Miky

  • Member
  • *
  • Posts: 31
Hello!

As I wrote before, there is still the problem with B-stations, sometimes per week, sometimes per 3 week, b-stations is freeze and not response, only reboot helps.

Is there some possibility to read out some log to find out the reason? Should be problem with PoE? We need to find some solution to make all system reliable, otherwise customer will no satisfied.

Thanks, Michal

March 17, 2023, 03:20:00 PM
Reply #2

Julien Levaufre

  • Administrator
  • Member
  • *****
  • Posts: 3234
    • www.visualproductions.nl
Hello Michal,
We do not have any feedback of other B-Station2 installations where the B-Station free. We are using multiple units in our building and we never had this problem.
The only idea I have to explain this problem would be that you have created a loop of feedback somewhere. In this case the B-station end up being overloaded by messages and freeze.
It can also be some messages not aimed to the b-station like a broadcast message coming from any device on the network.

Anyway  to be sure that it's not a Hardware issue, the best would be to switch the B-station2 with an other one and see if the problem keeps occurring.

Last thing, when the B-station is frozen, do you know if everything is blocked or if it's only the network part. Like does the LED feedback still work when pressing the buttons?

Julien Levaufre
Visual Productions BV

March 17, 2023, 10:14:17 PM
Reply #3

Miky

  • Member
  • *
  • Posts: 31
Hello,

Its very strange issue. At first look It looks like a problem on the network, First problems appeared 2/2022, b-stations did freeze every week till 4/2022, then IT operator removed some restriction on switcher port.
Then the B-station probably worked properly until 11/2022, when there was some upgrade on the network elements and there start the problems with the B-stations freezing again. After consulting with the network administrators, the administrator told us that he had already removed all port protections and could not remove anything else. The situation has improved, but the controllers still freeze occasionally. When we tried to communicate with the administrator again, he told us that there was nothing more he could do and that the fault would be on the device's side, as all the other devices on the network were fine except for the B-stations. He also told us that he doesn't want to deal with a device he doesn't know anymore because it's not his responsibility and it's up to the supplier and the manufacturer.
I would therefore need to define more precisely what the problem might be.

In the last attempt of the IT administrator to solve the problem, he found the following information from the log output on the switch:
-------------------------------------------------- ------------------------------------------------

occasionally - ?once a week it happens that the controller disconnects from the network.
At this moment, the device (driver) behaves like a "sleeping laptop", when after a requested handshake on the switch side, the port is set to a-10M/a-half-duplex. ("a" means that the port is automatically set to the best connection between the switch and the device. The port configuration does not specify specific connection parameters and is set based on a handshake - a mutual agreement between the switch and the device.
In the case of a laptop, after startup, NB will request a handshake again and the port will be set to a-1000M/a-full-duplex. But for some reason, the driver sw can't do that, or it gets into an untreated state from the point of view of the sw, when it is unable to respond to anything and it has to be restarted.
 
See the status of the port on the switch in the event of a "fault", when the controller stops responding. Ping doesn't work either. The port is in the up/connected state.
 
Gi6/0/34 ovl-204 notconnect 94 auto auto 10/100/1000BaseTX
Gi6/0/35 Studio Plus1 connected 94 a-half a-10 10/100/1000BaseTX
Gi6/0/36 AV BluePill connected 94 a-full a-1000 10/100/1000BaseTX
 
Settings on the port:
 
swri13-3p#sho run inte gi6/0/35
Building configuration...
 
Current configuration: 140 bytes
!
interface GigabitEthernet6/0/35
description Studio Plus1
switchport access vlan 94
switchport mode access
spanning-tree portfast
end
 
Status of the port where the device (driver) is connected:
 
swri13-3p#sho inte gi6/0/35
GigabitEthernet6/0/35 is up, line protocol is up (connected)
   Hardware is Gigabit Ethernet, address is 700f.6a8e.bf16 (bia 700f.6a8e.bf16)
   Description: Studio Plus1
   MTU 1500 bytes, BW 10000 Kbit/sec, DLY 1000 usec,
      reliability 255/255, txload 1/255, rxload 1/255
   Encapsulation ARPA, loopback not set
   Keepalive set (10 sec)
   Half-duplex, 10Mb/s, media type is 10/100/1000BaseTX
   input flow-control is off, output flow-control is unsupported
   ARP type: ARPA, ARP Timeout 04:00:00
   Last input 00:13:59, output 00:00:01, output hang never
   Last clearing of "show interface" counters 7w6d
   Input queue: 0/375/0/0 (size/max/drops/flushes); Total output drops: 0
   Queuing strategy: fifo
   Output queue: 0/40 (size/max)
   5 minute input rate 0 bits/sec, 0 packets/sec
   5 minute output rate 29000 bits/sec, 25 packets/sec
      16334258 packets input, 3079636478 bytes, 0 no buffer
      Received 4306239 broadcasts (142277 multicasts)
      0 runts, 0 giants, 0 throttles
      0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
      0 watchdog, 142277 multicast, 0 pause input
      0 input packets with dribble condition detected
      106310805 packets output, 15810585982 bytes, 0 underruns
      155 output errors, 478 collisions, 0 interface resets
      0 unknown protocol drops
      0 babbles, 530 late collision, 0 deferred
      0 lost carrier, 0 no carrier, 0 pause output
      0 output buffer failures, 0 output buffers swapped out
swri13-3p#
 
Ping not going through:
 
swri13-3p# ping 192.168.94.207
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.94.207, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)
swri13-3p#
 
There is no information in the log on the switch about errors that would occur there if the port is switched to the error-disable state.
 
Feb 10 15:24:44.780: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to up
Feb 10 15:24:45.779: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to up
Feb 14 11:42:38.042: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to down
Feb 14 11:42:39.043: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to down
Feb 14 11:42:54.405: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to up
Feb 14 11:42:55.404: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to up
Feb 15 08:44:05.165: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to down
Feb 15 08:44:05.505: %ILPOWER-5-IEEE_DISCONNECT: Interface Gi6/0/35: PD removed
Feb 15 08:44:06.165: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to down
Feb 15 08:44:18.493: %ILPOWER-7-DETECT: Interface Gi6/0/35: Power Device detected: IEEE PD
Feb 15 08:44:19.492: %ILPOWER-5-POWER_GRANTED: Interface Gi6/0/35: Power granted
Feb 15 08:44:22.562: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to up
Feb 15 08:44:23.563: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to up
Feb 15 14:27:40.596: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to down
Feb 15 14:27:41.597: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to down
Feb 15 14:27:43.956: %LINK-3-UPDOWN: Interface GigabitEthernet6/0/35, changed state to up
Feb 15 14:27:44.956: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet6/0/35, changed state to up
 
After finding that the device was connecting on a-100/a-full auto handshake, I set the port hard and the device rebooted and connected.
 
swri13-3p#sho inte status | in 6/0/35
Gi6/0/35 Studio Plus1 connected 94 full 100 10/100/1000BaseTX
swri13-3p#
 
swri13-3p#ping 192.168.94.207
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.94.207, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/1 ms
 
swri13-3p#sho run inte gi6/0/35
Building configuration...
 
Current configuration: 164 bytes
!
interface GigabitEthernet6/0/35
description Studio Plus1
switchport access vlan 94
switchport mode access
speed 100
duplex full
spanning-tree portfast
end
 
Please monitor the behavior on this device. In case of problems, please - call.
 
With regards and wishes for a peaceful day
jirka
----------------------------------------------------------------------------------------------------

I don't think the fault will be on the hw, it's not just one bstation doing it, and it worked fine for a period of time. Even the backlight of the buttons does not respond when it freezes.
We just need to somehow find out the cause.
You wrote: "It can also be some messages not aimed at the b-station like a broadcast message coming from any device on the network."
We use at this subnet also a lot of NDI based devices (cameras) and other audio IP protocols. Could there be some conflict caused by such traffic?
On B-stations we use your template for LPU-2, only we change the preset numbers and backlite colour. So I don't know where there could be a loop that overwhelms Bstation

Im sorry for such a long issue, but we have to solve it.

Best regards,

Michal

March 19, 2023, 12:55:28 AM
Reply #4

Julien Levaufre

  • Administrator
  • Member
  • *****
  • Posts: 3234
    • www.visualproductions.nl
The fact that you are using the same network for NDI video is an interesting information as it surely put a lots of data on the network.
Could you tell me what is the network switch reference ( manufacturer and model).
Are the cameras on the same ip network or just on the same physical network.?
I don't think it's linked but you could also change the OSC port you are using. I guess you are using port 8000, you can change it to 8005 for exemple.
 
Julien Levaufre
Visual Productions BV

March 20, 2023, 02:41:15 PM
Reply #5

Miky

  • Member
  • *
  • Posts: 31
We use the same subnet for cameras and for b-stations and LPU..  NDI is no main transmition platform there, we use SDI connection. But cameras support the NDI protocol and we use it sometimes for monitoring. The same subnet is because of remote administration.

Model of the networking switches I send you as I find out.

Michal

March 21, 2023, 01:16:56 AM
Reply #6

Julien Levaufre

  • Administrator
  • Member
  • *****
  • Posts: 3234
    • www.visualproductions.nl
ok,
To be able to find from where the problem is coming, we will need to change one sett9ing at the time. In this case even if the cameras are not outputting video stream all the time, I would try to change the Subnet used by the visual productions products,
We need to start looking somewhere and for now this seems the best explanation.

But you said it was working fine for some time and then after some changements it did start to freese. Do you know what had been changed at this time?
Julien Levaufre
Visual Productions BV

March 21, 2023, 11:47:02 AM
Reply #7

Miky

  • Member
  • *
  • Posts: 31
Here are the switch specs:

(attached)

March 21, 2023, 11:56:23 AM
Reply #8

Miky

  • Member
  • *
  • Posts: 31
Im not 100% sure, but it started after network updates was applied..

 

SMF spam blocked by CleanTalk