[INSTRM-1769] summit DNSmasq needs to be sorted out Created: 26/Oct/22  Updated: 01/Feb/24  Resolved: 01/Feb/24

Status: Done
Project: Instrument control development
Component/s: ics_dnsmasq
Affects Version/s: None
Fix Version/s: None

Type: Task Priority: Normal
Reporter: yuki.moritani Assignee: Yoshida, Hiroshige
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Relates
relates to INSTRM-2089 move pdu2-dcb (and else if any) from ... Done

 Description   

Recently DNSmasq configuration is checked to prepare SM3 AIT work. At that time, there are some points to be sorted out. Here is report from Yoshida-san on Slack:

The summit DNSmasq configuration has been updated to the latest master (commit 35bd765a) + some local mods. & fixes. Please report any issues.
Known differences from the previous production configuration:

  • agcc-optiplex-755 gets 133.40.165.106, not .248
  • pfi-temps[23] has correct MAC address for old E-Box
  • filterwheel-dcb conflicts with pdu2-dcb, thus commented out

 

 Comments   
Comment by yuki.moritani [ 28/Oct/22 ]

For the PFI-related points (the first and second bullets), it is comes from we assign the same IP to the devices on the old and new E-box.

  • 133.40.165.106 is assigned to small PC in the new E-box, and agcc-optiplex-755 gets IP (DHCP), so dnsmasq should be modified with small PC information
  • There are 2 sets of temperature devices (pfi-temps 1,2,3, see below), and the PFI uses the second set. I'm not sure which is better: to have 6 devices with different IP, or update MAC address based on the box attached on PFI.

PFI temperature devices (ADAM-6015)

set 1: in the new E-box

  • pfi-temps1: 00-D0-C9-F4-2A-78
  • pfi-temps2: 00-D0-C9-F4-29-BE  (I confirmed this is correct)
  • pfi-temps3: 00-D0-C9-F6-3F-60

set 2: in the old E-box, and attached on PFI

  • pfi-temps1: 00:d0:c9:f4:2a:78
  • pfi-temps2: 74:fe:48:5b:7d:70
  • pfi-temps3: 74:fe:48:5b:7d:9c
Comment by Yoshida, Hiroshige [ 28/Oct/22 ]

So we cloned the MAC of pfi-temps1 (they are identical - that is one way to make swapping between two E-boxes easier - perhaps we should clone MACs of pfi-temps2 and pfi-temps3 as well)?

Comment by Yoshida, Hiroshige [ 28/Oct/22 ]

filterwheel-dcb is on 133.40.165.123, as in master (commit 35bd765a). pdu2-dcb is moved to 133.40.165.124.

Comment by yuki.moritani [ 03/Nov/22 ]

In regard to pfi-temp[23], it is possible to clone the MAC address. Given that the old E-box is working in the current situation, we'll modify it after Dec run, when Chi-Hung is visiting Subaru.

Comment by cloomis [ 04/Nov/22 ]

If you just want to have unchanging IP addresses for the various eboxes, I am pretty sure it is OK to have two MAC addresses for one hostname/IP address. You just have to make sure not to power/connect them both at the same time.

Comment by yuki.moritani [ 05/Nov/22 ]

In short term, until we confirm the new E-box is fine to use on the telescope, we may turn on both old E-box and new E-box, I guess.

Comment by cloomis [ 05/Nov/22 ]

As far as dnsmasq itself goes I just complicated things a bit: I created the sm[2-4] subaru host configs in dnsmasq-ics:/etc/dnsmasq.d and checked them in as myself. But I did not push, since I am not smart enough to do that from there.

Comment by yuki.moritani [ 28/Oct/23 ]

INSTRM-2089 was created to add pdu2-dcb. If there is no further item to sort out, I'll close this ticket.

Generated at Sat Feb 10 16:39:13 JST 2024 using Jira 8.3.4#803005-sha1:1f96e09b3c60279a408a2ae47be3c745f571388b.