fix(adminui): Request-rebirth cannot be armed on a Sparkplug plant that has not birthed since the browse window opened #514

New Issue

2026-07-27T13:51:16-04:00

dohertj2 commented

2026-07-27 13:51:16 -04:00

The browse tree is birth-driven: it is built from the births observed since the browse session opened. On a
quiet plant — every edge node already birthed long ago and is publishing only NDATA/DDATA — the tree comes up
empty, and Request rebirth… is scoped from a tree selection, so the one action that would populate the tree
cannot be armed. Chicken-and-egg.

The session side is already capable. RequestRebirthAsync accepts a bare {group}/{edgeNode} scope — it
does not require a selected device or metric. What is missing is a way for the operator to express that scope
when there is nothing in the tree to select.

Fix sketch. Let the operator type/choose an edge-node scope (or offer a broadcast to the configured
groupId) when the tree is empty, rather than gating the button on a tree selection. A group-wide rebirth is
noisy on a large plant, so prefer an explicit edge-node entry over a blanket broadcast, or warn before it.

Related: #510 (the button's enablement is also captured once, at browse-open).

Acceptance. Against a simulator that birthed before the browse window opened, an operator can request a
rebirth and see the tree populate — without restarting the simulator or the driver.

The browse tree is **birth-driven**: it is built from the births observed since the browse session opened. On a quiet plant — every edge node already birthed long ago and is publishing only NDATA/DDATA — the tree comes up empty, and *Request rebirth…* is scoped from a tree selection, so the one action that would populate the tree cannot be armed. Chicken-and-egg. **The session side is already capable.** `RequestRebirthAsync` accepts a bare `{group}/{edgeNode}` scope — it does not require a selected device or metric. What is missing is a way for the operator to *express* that scope when there is nothing in the tree to select. **Fix sketch.** Let the operator type/choose an edge-node scope (or offer a broadcast to the configured `groupId`) when the tree is empty, rather than gating the button on a tree selection. A group-wide rebirth is noisy on a large plant, so prefer an explicit edge-node entry over a blanket broadcast, or warn before it. Related: #510 (the button's enablement is also captured once, at browse-open). **Acceptance.** Against a simulator that birthed before the browse window opened, an operator can request a rebirth and see the tree populate — without restarting the simulator or the driver.

dohertj2 added the driver/mqtt label 2026-07-27 13:51:16 -04:00

dohertj2 referenced this issue from a commit

2026-07-27 13:52:13 -04:00

docs(drivers): un-stale the MQTT README rows + track every known gap as an issue

dohertj2 referenced this issue from a commit

2026-07-31 06:42:15 -04:00

fix(mqtt): close #509, #510, #512, #513, #514

dohertj2 referenced this issue from a commit

2026-07-31 06:42:15 -04:00

merge: MQTT tracker batch — #508, #509, #510, #512, #513, #514

dohertj2 commented

2026-07-31 06:42:58 -04:00

Fixed in master 2a7d1167.

Confirmed the issue's read: MqttBrowseSession.ResolveRebirthTargets already handles a bare
{group}/{edgeNode} — its own comment calls that case "the prime rebirth target". So this was purely
a missing affordance, and the fix adds an edge-node entry field shown whenever no tree scope is
selected. The driver's configured groupId pre-seeds it, so only the edge node has to be typed.

Edge-node scope only, deliberately. A group scope is rejected: the session resolves one by
enumerating observed edge nodes, of which an empty tree has none — so it could not help the very case
this exists for, and on a populated plant it would fan one click across the whole group.

The rule is extracted to SparkplugRebirthScope rather than left inline in the Razor, because this repo
has no bUnit and logic stranded in markup is untestable. It is advisory: the session re-resolves and
refuses an unusable scope before publishing anything, so being the looser of the two checks is safe
while being the stricter one would block a scope the session accepts.

Fixed in master `2a7d1167`. Confirmed the issue's read: `MqttBrowseSession.ResolveRebirthTargets` already handles a bare `{group}/{edgeNode}` — its own comment calls that case *"the prime rebirth target"*. So this was purely a missing affordance, and the fix adds an edge-node entry field shown whenever no tree scope is selected. The driver's configured `groupId` pre-seeds it, so only the edge node has to be typed. **Edge-node scope only, deliberately.** A group scope is rejected: the session resolves one by enumerating *observed* edge nodes, of which an empty tree has none — so it could not help the very case this exists for, and on a populated plant it would fan one click across the whole group. The rule is extracted to `SparkplugRebirthScope` rather than left inline in the Razor, because this repo has no bUnit and logic stranded in markup is untestable. It is **advisory**: the session re-resolves and refuses an unusable scope before publishing anything, so being the looser of the two checks is safe while being the stricter one would block a scope the session accepts.

dohertj2 closed this issue

2026-07-31 06:42:58 -04:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: dohertj2/lmxopcua#514