Data streaming IP addressing scheme docs - #231
Conversation
|
Question:
|
|
Should discuss along with @ChrisM-S |
| - `192.168.1.250` to be the streaming control board's control port | ||
| - `192.168.1.251` to be the streaming control board's status-packet port | ||
| - `192.168.1.252` to be streaming monitors | ||
| - The rest of the range to be streaming detectors |
There was a problem hiding this comment.
before or after 250-252? guessing before, otherwise that's not many IPs for detector modules..?
There was a problem hiding this comment.
Haha yeah, HRPD will have 80+ detector modules. SANDALS might be even worse. Will make this clearer...
| - `192.168.1.251` to be the streaming control board's status-packet port | ||
| - `192.168.1.252` to be streaming monitors | ||
| - The rest of the range to be streaming detectors | ||
|
|
There was a problem hiding this comment.
I would like a section in here about how our streaming software ie. kdaectrl + kdaediag interface with clients/the NDX - they probably need to be put on the site network, or also linked into the instrument private network with galils/beckhoffs on?
There was a problem hiding this comment.
Yes the intention was that the streaming server would be dual-networked:
- one side to the streaming private network
- other side to the main ISIS network
I don't currently have a use-case where the streaming server would need to connect to the instrument private network, but one might show up...
| - Detectors | ||
|
|
||
| The current standard for network addressing is: | ||
| - `192.168.1.1` is the address of the streaming server, which will currently host both `kafka_dae_control` and `event_udp_to_kafka` processes |
There was a problem hiding this comment.
192.168.1.1 and 192.168.1.2 are currently used by NDX/NDH - so the streaming computer in a cabin will be on its own 192 network separate to NDX/NDH?
There was a problem hiding this comment.
That was the idea yes - streaming private network will be totally distinct from the existing private network containing NDH/galil/beckhoff etc. I will try to make this clearer.
|
I did talk to @ChrisM-S , he thought it might be better to use a totally different address range and also to make that address range different per-instrument to reduce risk if networks get plugged into wrong places. But I need to talk to someone like Anthony about this too to ensure we don't create wider problems if we do that. Also discussed but not on this PR yet:
|
3 participants
Document IP addressing scheme for streaming system post discussion with DSG