novox/mesh-controller
18
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Compose the packages mirror only on a change, end every serve goroutine before serve returns, and raise no alarm for a cancellation (issue 1945) #464

Merged
jschoubben merged 5 commits from fix/r6-mirror-on-change-2 into main 2026-10-25 16:20:55 +00:00
Conversation 1 Commits 5 Files Changed 20
+783 -83
5 Commits
Author SHA1 Message Date
jschoubben 8e02c4726e Clear a module's fingerprint-not-composed on every successful take-in, and stop serve's goroutines before an error return undoes anything (issue 1945)
mesh/check pass 8e02c4726eb4 on c3db00ceac25 | unit 981 run, 171 kept, 2 skipped | integration 1641 run, 232 kept, 7 skipped | over bound: integration…
mesh/delivery held for a person: merged without a passing check: only a person decides what follows
Details 
A success cleared the condition only when the same follower had raised it, so
one raised before a restart or by another follower stayed open after the take-in
worked, and the operator read that a build needed a look. Clearing a condition
that is not open costs one read. On an error return, serve now ends its
goroutines before the keeper closes and before handActConn and the bus are put
back (the review of fix 12).
 2026-10-25 17:09:35 +01:00
jschoubben cc5c9ae888 Count the take-ins and every other background work in the serving group, so serve's stop waits for them too (issue 1945) 
inBackground ran the take-ins of done keys (now in #463's slots), the start's
look, the given values' replacement and the announced reads beside serve without
serve waiting for them: a take-in could outlive its controller as the loops did.
It now takes the context and is counted in its serving group as well, and a
serve that is stopping starts none.
 2026-10-25 16:36:49 +01:00
jschoubben c7e31640cf Raise no condition for a take-in, a layout or a put cut by a cancellation (issue 1945) 
Thirty urgent fingerprint-not-composed conditions were live on 2026-10-25, each a
done key's take-in a stop or a restart cut with "context canceled". A
cancellation is no fault of the build, the chain or the bucket: a take-in or a
layout whose controller stops returns and the next one does it; one whose own
step was cancelled is tried again without a condition; a send cut by a
cancellation counts as no failure to put its files.

The put test now starts and leaves the node's failure count at nothing, so a
second run of it (-count=2) finds no count left by the first.
 2026-10-25 16:30:58 +01:00
jschoubben c75d726205 End every goroutine serve starts before serve returns, so no loop outlives its controller (issue 1945) 
The try of the on-change mirror failed with every test passed: goroutines a
served test started outlived its serve and met the next test's globals and its
closed stores. Under -race the served tests show it as a race on conditionsFrom
(watchTheMesh writing it while settlingPending, already running, read it), and
a take-in cut by the stop raised fingerprint-not-composed, a kind with no plain
words, which fails the package after its PASS line.

- serve counts every goroutine it and its parts start (serving_group.go) and
  ends and waits for them before anything else is undone: the lease's watch,
  the pending settler, the status summary, the facts, the mirrors and the new
  declared-packages keeper with its two bucket watches, the declaration sweep,
  the read-call counter, the watches' closers, the have watches, the vault move
  on a report, and the watchdogs, doctor, healers and asker under their own
  group.
- The condition store opens before anything that reads conditionsFrom starts;
  askerFrom is an atomic pointer, read from the keeper's goroutines.
- A done key's take-in cut by the stop raises nothing, and fingerprint-not-composed
  has plain words.
 2026-10-25 16:30:58 +01:00
jschoubben a4cd5ba375 Compose the declared-packages mirror only when what it composes from changed (issue 1945) 
The mirror composed every node in full on the record mirrors' 10 s tick; a
pass takes about 45 s on the control node, so the controller never rested
(130-175 % CPU, issue 1945 note 22). It now passes on a watch of the
compositions' digests and of the assignments, one pass at a time with the
wakes of a pass coalesced into one after it (at least 30 s apart), skips the
composition when the stores' Input digest is unchanged, and keeps a safety
pass every 10 minutes that composes nothing when nothing changed.
 2026-10-25 16:30:58 +01:00