eculocate

commit 9be9d167c38fa699c203ce4a9cb0644547d23cdf
parent c393353ecea879978e682bc403805d6379852f73
Author: Daniel Barlow <dan@telent.net>
Date:   Mon, 12 Jan 2026 17:58:08 +0000

mdns implementation notes

Diffstat:
M
ecu-esp32/src/mdns.rs
 | 
67
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

1 file changed, 67 insertions(+), 0 deletions(-)
diff --git a/ecu-esp32/src/mdns.rs b/ecu-esp32/src/mdns.rs
@@ -1,3 +1,70 @@
+// MDNS responder, mostly but not entirely standards-compliant
+
+// ** Probe and Announce at startup
+
+// at startup it does Probe and Announce for "all those resource
+// records that a Multicast DNS responder desires to be unique on the
+// local link".  "Probe" means send packets as QU (unicast response)
+// containing ANY requests for each name: 3 packets with 250ms between
+// them; "Announce" means send 2 <= n <= 8 packets one second apart,
+// containing all the records we know
+
+// ** additional data in responses
+
+//  * for SRV, we send all the A and AAAA it names
+
+//  * for PTR "Service Instance Enumeration or Selective Instance
+//    Enumeration (subtype)": i.e. when the record data names an
+//    instance of the service, we sent the SRV and TXT records for the
+//    instance _and_ all the A/AAAA named by the SRV
+
+// I haven't investigated what happens if the packet gets too
+// large. There may be bugs
+
+// ** don't answer queries if the answers in the query already include what we'd send
+
+// This is unimplemented for now.  mdns is so fricking noisy already I
+// don't know if this is worth the extra effort
+
+// ** set the CACHE_FLUSH bit on unique records
+
+// "all the records with that name, rrtype, and rrclass are conceptually
+// under the control or ownership of a single responder"
+
+// ** NSEC?
+
+// not done
+
+// ** unicast response requested (QU)
+
+// - top bit in the class field of a DNS question is the unicast-response bit
+// - send response unicast, unless the record has not been sent recently
+//   ("within one quarter of its TTL")
+// - also for queries received fom unicast
+//   - but verify the source is local lan
+//   - and the port is 5353 (otherwise see "legacy")
+
+//  [ mostly done; we don't check "has been sent recently" and as we
+//  didn't open a unicast socket then we can't receive queries from
+//  unicast ]
+
+
+// ** Legacy Unicast Responses
+
+// if source port != 5353
+
+// - respond via unicast to source
+// - query id and question
+// - don't set cache-flush bit
+// - ttl <= 10s
+
+// This is implemented but has not really been tested
+
+// ** response compression
+
+// Unimplemented
+
+
 use alloc::vec::Vec;
 use alloc::string::String;
 use alloc::collections::BTreeSet;