3x-ui/internal/web
Sanaei f072d0448d
fix(clients): flag the restart a partly-applied edit or delete still needs
63b46cd6 made a multi-inbound client op apply its inbounds concurrently and
stop aborting at the first failure, so an error can now come back together
with needRestart=true: the inbounds that succeeded committed real changes and
their Xray still needs the restart. That commit taught the two callers it
converted — create and attach — to read the flag before the error check.

d34ec97f then routed Update, Delete, Detach and DeleteByEmail's record-less
fallback through the same fanout but touched no caller, so on a master with
several nodes a partly-applied edit or delete returned (true, err) into a
handler that returned on err first. Xray was never flagged for the work that
landed and notifyClientsChanged never fired, so the running config kept
serving the pre-edit client set and every open panel showed stale rows until
something else happened to trigger a restart.

Read needRestart before the error check in update, delete and detach, and
broadcast on needRestart || err == nil — the same shape create and attach have
had since 63b46cd6. The predicate is a strict superset of the old err == nil,
and needRestart is only ever assigned after a runSerializedTx commit, so it
firing genuinely means something landed.

The three handlers are pinned by a new controller test each: one client on two
inbounds, the second one's settings JSON corrupted so the op commits on one and
fails on the other, asserting both the success:false response and the restart
flag. All three fail without the change.

The API docs for update, del and detach now describe the partial-application
contract, as add and attach already did. Detach ends at the fanout so every one
of its errors carries the inbound prefix; update and delete write the client
record afterwards, and a failure there is reported without one.
2026-09-07 01:51:01 +02:00
..
controller fix(clients): flag the restart a partly-applied edit or delete still needs 2026-09-07 01:51:01 +02:00
entity feat(sub): client-side balancers for the JSON subscription (#6243) 2026-08-23 22:34:20 +02:00
global refactor: focused service files, leaf subpackages, and an internal/ layout (#5167) 2026-06-10 15:19:22 +02:00
job perf(clients): apply a multi-inbound client create concurrently 2026-09-04 01:01:20 +02:00
locale style: adopt golangci-lint v2 and resolve all findings 2026-06-27 15:42:22 +02:00
middleware fix(web): tighten database restore body-cap exemption (#5609) 2026-06-28 15:00:55 +02:00
network style: format struct literals and whitespace 2026-08-24 00:06:01 +02:00
runtime fix(node): let a node-reported tag outrank a stale adopted alias 2026-09-04 02:48:35 +02:00
service fix(tgbot): use a token telego accepts in the edit-message tests 2026-09-06 17:32:01 +02:00
session refactor: focused service files, leaf subpackages, and an internal/ layout (#5167) 2026-06-10 15:19:22 +02:00
translation feat(sub): leastLoad member weights for subscription balancers (#6304) 2026-09-02 20:21:27 +02:00
websocket fix(nodes): keep the credential-presence flag on the node heartbeat push 2026-07-28 17:38:46 +02:00
cadence_test.go Add remote routing URL support (#6168) 2026-08-18 16:02:11 +02:00
routes_contract_test.go docs(api): align OpenAPI with runtime contracts (#6409) 2026-09-04 15:23:41 +02:00
web.go feat(amneziawg): add native AmneziaWG protocol support (#6105) 2026-08-24 02:41:15 +02:00
web_mtls.go feat(node): node hardening — mTLS, hashed+zstd reconcile transport, per-node net metrics (#5382) 2026-06-16 12:19:33 +02:00
web_mtls_test.go feat(node): node hardening — mTLS, hashed+zstd reconcile transport, per-node net metrics (#5382) 2026-06-16 12:19:33 +02:00