1. Forum
  2. DOVE-Net
  3. Synchronet Programming

  • src/sbbs3/websrvr.cpp

    From Rob Swindell (on Windows 11)@VERT to Git commit to main/sbbs/master on Tue Feb 24 16:28:36 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/fd0b9d2bc21b13fa9840e190
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    Remove (long-deprecated) access.ars file support (use webctrl.ini instead)

    Fix issue #1083

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Windows 11)@VERT to Git commit to main/sbbs/master on Sun Mar 8 05:28:49 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/0c0cb7c473285ab8c71f209a
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    Allocate extra byte for NUL-terminator in read_post_data()

    This looks like an off-by-one bug in this one call to realloc(), not adding one for the NUL terminator in this case.

    Potential cause of issue #1094

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Sun Mar 8 20:47:44 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/07ed41c51fd336b9d7c5d9b4
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    Add NUL to end of post data file

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Wed May 6 19:41:53 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/3ad3f0282bcff37fa1926121
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: cast away two best-effort unchecked returns (CIDs 639932, 639941)

    CID 639932: remove(cleanup_file[i]) in close_request Ä best-effort
    cleanup of temporary request files; failure is benign.
    CID 639941: setsockopt(TCP_NODELAY) in http_session_thread Ä latency
    hint; failure is non-fatal. Also widen the bool nodelay to
    int so it has correct setsockopt() type.

    Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Wed May 6 19:41:53 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/62b41cd33f2fa4292d83710d
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: handle getuserdat failures in http_logon and check_ars (CIDs 516407, 516410, 639949)

    Both call sites set user.number then read the rest of the user record
    via getuserdat(). On read failure the user struct was left partially
    populated, then used for password comparison or downstream session
    state. Treat the failure as a system error: log it and either fall
    back to an unauthenticated session (http_logon) or reject the auth
    attempt (check_ars).

    Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Wed May 6 19:41:53 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/65d46495860f18fb2c9a5882
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: suppress send_error ORDER_REVERSAL false-positive (CID 631137)

    Coverity reports an ORDER_REVERSAL between link_list.mutex and
    jsrt_mutex when http_session_thread calls send_error() in the
    client-limit branches. The link_list helpers in this thread
    (loginAttempts, client_on, listCountMatches) acquire+release their
    list mutex internally Ä nothing holds a list mutex when send_error
    runs js_setup() which acquires jsrt_mutex. Annotate both 503/429
    send_error sites with a SUPPRESS plus rationale.

    Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Wed May 6 22:36:57 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/6ad832522da440e614b8fcdf
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: clamp tls_sent and explicit cast in sess_sendbuf return (CID 639935)

    The TLS path assigns 'result = tls_sent' where tls_sent is int and
    could theoretically be negative on cryptlib edge cases. Adding it
    to size_t 'sent' would underflow. Guard with 'if (result > 0)'.

    Also make the size_t-to-int returns explicit casts so Coverity sees
    the narrowing is intentional.

    Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Wed May 6 22:51:40 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/c7df44f17c494f7277ac112e
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: skip getuserdat for anonymous sessions in http_logon

    Regression from 9e7649fe0: when http_logon is called with usr=NULL
    on an anonymous request (session->user.number == 0), getuserdat
    legitimately fails because user 0 doesn't exist, which now spams
    the log with '!ERROR reading user #0 data' on every anon hit.

    Only call getuserdat when there's an actual user number to read.

    Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Windows 11)@VERT to Git commit to main/sbbs/master on Wed May 6 23:04:20 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/c94f75aa58112c228a8cdce9
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: include protocol, IP, request, and ARS in no-auth log

    The "!No authentication information" debug log line now reports the
    protocol, client address, request line, and the ARS string that triggered
    the auth requirement, so it's actionable when WEB_OPT_DEBUG_RX is on.

    Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Sat May 9 14:04:17 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/f7b10a614935817ba8965ec1
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: don't call destroy_session() with sentinel tls_sess value (-1)

    When TLS setup fails after add_private_key() returns an error, the code
    calls cryptDestroySession() directly and sets tls_sess = -1, then calls close_session_no_rb() which would pass -1 to destroy_session(), triggering
    a spurious "Destroying a session (-1) that's not in sess_list" error.

    Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Tue Aug 4 22:12:21 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/0ca5676148730160b440cf02
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    websrvr: always read the file header when resolving a file-vpath request

    4de1032086 (payroll-20-queue, 2026-08-01) skipped the header record for
    a file in a free directory, on the grounds that loadfile() zeroes the
    cost there and download_is_free() short-circuits on the same flag before
    any credit comparison.

    The cost is not the only field that comes from the header. The index
    record stores the filename truncated to SMB_FILEIDX_NAMELEN (64 bytes, extension preserved), and smb_getfile() points file.name at it, so at
    index detail file.name was that truncated form. Names longer than the
    limit are not hypothetical: one in this file base is 78 characters. The
    name reaches user_downloaded_file(), which embeds it in the uploader's
    "file downloaded" notification, and mqtt_file_download(), which
    publishes it.

    The transfer and the credit accounting survived it, because deriving an
    index name from an already-truncated one is idempotent and the record
    still resolved, but the name recorded and announced was wrong.

    Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell (on Debian Linux)@VERT to Git commit to main/sbbs/master on Sat Aug 8 18:05:04 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/0791f3e3bfcdb04afeda7134
    Modified Files:
    src/sbbs3/websrvr.cpp
    Log Message:
    Don't hand a CGI the web server's descriptors (#1174)

    The CGI child sets up stdin, stdout and stderr and then execs with everything else this process had open still in hand. Close the rest, as externals now
    do. Nothing above stderr is a CGI's business: the client socket it reads
    from, where it has one, has already been duplicated onto stdin.

    Verified against a listen socket left deliberately inheritable, standing in
    for the descriptors this cannot otherwise reach - the ones opened inside libraries, where there is no call site to mark. Before, the CGI inherited it; after, the CGI starts with stdio alone.

    The exec-failure message now goes to stderr directly, which is the pipe the parent already reads and logs as a CGI error. errprintf() is not usable after the close: it is not fork-safe, and its descriptors are among those closed.

    Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net