В одном epoll_wait batch три события для fd: EPOLLERR (освобождает tc),
EPOLLIN (read_cb), EPOLLOUT (write_cb). read_cb/write_cb guard проверяет
tc->sock==SOCKET_INVALID, но sock обнулялся ПОСЛЕ on_error.
Перенос sock=SOCKET_INVALID до on_error защищает от use-after-free.
Доказано: без фикса segfault на tcp_io.c:374 (write_cb → tc->connected).
С фиксом тест работает без крашей.
tcp_conn_handle_error гвардил uasync_remove_socket_t по tc->socket_id,
пропуская повторный вызов. В одном epoll_wait batch два события
(EPOLLERR + EPOLLOUT) на один fd: первое делает remove (socket_id=NULL),
второе проходит guard и dispatch'ит write_cb на освобождённом tc.
Фикс: всегда вызывать uasync_remove_socket_t + epoll_ctl(DEL),
независимо от socket_id.
- pkt_normalizer: save recvpart->len before queue_data_put (may free synchronously)
- ll_queue: save uasync_call_soon handle in queue_waiter_handle, cancel on free
- uasync: generation counter in epoll data.u64 to skip stale events after fd reuse
- uasync: keep fd_to_index mapping after socket_array_remove for inactive nodes
- test: test_uasync_socket_race — 4 children × 200 iterations, currently flaky
- config_parser: tun_enabled в [global] (по умолчанию yes), не дефолтить
tun_ifname если TUN не нужен
- utun_instance: уважать tun_enabled, чистые дефолты tcp_proxy TUN
- socks_proxy: фикс дэдлока — reply сразу после CONNECT, без ожидания DATA
- utun.conf.sample: документирован tun_enabled, tun_ifname
- test_socks_http_proxy: 50 запросов (SOCKS+HTTP), 2 воркера параллельно,
файлы 1k..100k, сверка через cmp, python http.server + curl
Добавлен модуль socks_proxy.c/h — SOCKS5 и HTTP CONNECT proxy на клиенте.
Использует tcp_io для локальных TCP соединений, туннелит трафик через
существующий ETCP-протокол (ETCP_ID_TCP_PROXY/ETCP_ID_TCP_PROXY_CLIENT).
Конфиг [tcp_proxy_client]:
- socks_enabled/socks_addr — включить SOCKS5 сервер на указанном IP:PORT
- http_proxy_enabled/http_proxy_addr — включить HTTP CONNECT сервер
- Можно включать TUN, SOCKS и HTTP прокси одновременно или раздельно
Весь трефик идёт через exit node (tcp_proxy_server), менять exit не пришлось.
Root cause: when bgp->local_node was reallocated (changed=1), old route
entries remained with stale v_node_info pointing to freed memory. The freed
memory got reused by a BGP peer's NODEINFO_Q, causing route_lookup to return
wrong node_id (b3b2b1b0afaeadac instead of local), which then failed at
etcp_route_send with 'no BGP route' and dropped packets to 10.23.2.1.
Changes:
- route_node.c: route_delete(old) before u_free + route_insert(new) after alloc
- utun_instance.c: remove duplicate route_insert (now atomic inside function)
- route_bgp.c: route_delete before freeing old peer NODEINFO_Q
- route_lib.c: routes_overlap now rejects only exact prefix+network duplicates
- tests/Makefile.am: add route_lib.o to test_route6_lib
When own NODEINFO is reflected back from a peer, do not add it
to bgp->nodes. Only update bgp->local_node which is sent separately.
This prevents route duplication and the 'local node in learned list'
inconsistency.
- keepalive_adaptive=0 disables adaptive keepalive period growth,
keeping it fixed at 200ms for faster timeout detection in tests
- test_bgp_triangle: 5-node triangle topology test (6 phases):
validates alternative path recording (fix#1), cascading node
removal on full isolation, and full recovery
- config generation via utun_instance_create_from_str() with
pre-generated keys — no temp files needed
- route_bgp_process_nodeinfo: record alternative paths even when NODEINFO
version hasn't changed (fixes nodes disappearing on link down despite
having other active links)
- route_bgp_process_withdraw: only remove node when all paths gone (ret==1)
- route_bgp_remove_conn: broadcast withdraw for each removed node with
correct wd_source
- route_node_format_all(): format all BGP nodes to snprintf buffer
- control_server: 'nodes' action sends ETCPMON_RSP_ACTION_RESULT (0x89) with text
- etcpmon_protocol.h: new response type + struct, increase MAX_MSG_SIZE to 32K
- etcpmon_client: on_action_result callback, send_action via send_request with seq_id
- etcpmon_gui: ETCPMonActionResult popup window class with EDIT multiline,
on_action_result creates popup (same size as main) or appends text + scrolls
Add route_node_dump_all() in route_node.c that walks bgp->nodes list
and outputs per-node info: node_id, name, version, pubkeys, IPv4/IPv6
addrs/sockets/subnets, paths with hop list, connectivity state, conn_mgr.
Wire 'nodes' action in control_server to call this function.
Root cause: test1 used lks() (INIT links up) but conn_mgr needs BGP NODEINFO.
Nodeinfo arrives after INIT, so test1 must wait for it.
Also made result volatile to prevent loop optimization.
Both req (INIT_REQUEST) and resp (INIT_RESPONSE) point to pkt->data.
resp->peer_port at offset 23-24 overwrites req->src_ipv4[0..1] at same offset.
Save src_ipv4/src_port before building the response.
- Extract ETCP_INIT_REQUEST_PKT / ETCP_INIT_RESPONSE_PKT named structs
- Replace anonymous structs and manual offset parsing with typed struct access
- Add src_ipv4/src_port fields (V2) to INIT REQUEST for DIRECT detection
- Server compares reported client addr with observed src; match + public IP → NAT_TYPE_DIRECT
- Skip STUN NAT check for DIRECT links
- route_bgp: handle NAT_TYPE_DIRECT → NAT_VERIFIED_DIRECT in handle_nat_info
- Add ed25519_public_key[32] to NODEINFO struct for pubkey distribution
- Add ed25519_public_key[32] to ROUTE_BGP (derived from X25519 privkey at init)
- Add ROUTER_FLAG_SIGNED (0x08) to SVC_ROUTE_HDR flags byte
- router_send_one_flags: when is_signed, sign [hdr][payload] and append 64-byte sig
- etcp_router_recv_cb: verify Ed25519 signature when ROUTER_FLAG_SIGNED set
- New API: etcp_router_conn_send_signed()
- Drop signed packets if sender node not in routing table or no Ed25519 key
- 5 unit tests in test_etcp_router_unit.c covering OK/tampered/unknown/no-key/short
last_recv_local_time was not set in the server INIT processing path,
causing the keepalive timer to see a stale timestamp and trigger a
false link-down → route_bgp_remove_conn() right after reconnect.
Routes could be deleted before the first normal packet from the
client updated the timestamp via process_decrypted.
Evgeny