WIP: temmie/userweb: add log processor for apache

hosts/temmie/services/userweb/apache-log-processor/.gitignore

@@ -0,0 +1 @@
+target

hosts/temmie/services/userweb/apache-log-processor/Cargo.lock

@@ -0,0 +1,171 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 4
+
+[[package]]
+name = "apache-log-processor"
+version = "0.1.0"
+dependencies = [
+ "nix",
+ "time",
+]
+
+[[package]]
+name = "bitflags"
+version = "2.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c4512299f36f043ab09a583e57bceb5a5aab7a73db1805848e8fef3c9e8c78b3"
+
+[[package]]
+name = "cfg-if"
+version = "1.0.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9330f8b2ff13f34540b44e946ef35111825727b38d33286ef986142615121801"
+
+[[package]]
+name = "cfg_aliases"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "613afe47fcd5fac7ccf1db93babcb082c5994d996f20b8b159f2ad1658eb5724"
+
+[[package]]
+name = "deranged"
+version = "0.5.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7cd812cc2bc1d69d4764bd80df88b4317eaef9e773c75226407d9bc0876b211c"
+dependencies = [
+ "powerfmt",
+]
+
+[[package]]
+name = "itoa"
+version = "1.0.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8f42a60cbdf9a97f5d2305f08a87dc4e09308d1276d28c869c684d7777685682"
+
+[[package]]
+name = "libc"
+version = "0.2.186"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "68ab91017fe16c622486840e4c83c9a37afeff978bd239b5293d61ece587de66"
+
+[[package]]
+name = "nix"
+version = "0.31.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cf20d2fde8ff38632c426f1165ed7436270b44f199fc55284c38276f9db47c3d"
+dependencies = [
+ "bitflags",
+ "cfg-if",
+ "cfg_aliases",
+ "libc",
+]
+
+[[package]]
+name = "num-conv"
+version = "0.2.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "521739c6d2bac4aa25192232afe6841231376b2b26d4d9fae5ecf8ca5772e441"
+
+[[package]]
+name = "num_threads"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c7398b9c8b70908f6371f47ed36737907c87c52af34c268fed0bf0ceb92ead9"
+dependencies = [
+ "libc",
+]
+
+[[package]]
+name = "powerfmt"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "439ee305def115ba05938db6eb1644ff94165c5ab5e9420d1c1bcedbba909391"
+
+[[package]]
+name = "proc-macro2"
+version = "1.0.106"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8fd00f0bb2e90d81d1044c2b32617f68fcb9fa3bb7640c23e9c748e53fb30934"
+dependencies = [
+ "unicode-ident",
+]
+
+[[package]]
+name = "quote"
+version = "1.0.45"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "41f2619966050689382d2b44f664f4bc593e129785a36d6ee376ddf37259b924"
+dependencies = [
+ "proc-macro2",
+]
+
+[[package]]
+name = "serde_core"
+version = "1.0.228"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "41d385c7d4ca58e59fc732af25c3983b67ac852c1a25000afe1175de458b67ad"
+dependencies = [
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_derive"
+version = "1.0.228"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d540f220d3187173da220f885ab66608367b6574e925011a9353e4badda91d79"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "syn"
+version = "2.0.117"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e665b8803e7b1d2a727f4023456bbbbe74da67099c585258af0ad9c5013b9b99"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-ident",
+]
+
+[[package]]
+name = "time"
+version = "0.3.47"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "743bd48c283afc0388f9b8827b976905fb217ad9e647fae3a379a9283c4def2c"
+dependencies = [
+ "deranged",
+ "itoa",
+ "libc",
+ "num-conv",
+ "num_threads",
+ "powerfmt",
+ "serde_core",
+ "time-core",
+ "time-macros",
+]
+
+[[package]]
+name = "time-core"
+version = "0.1.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7694e1cfe791f8d31026952abf09c69ca6f6fa4e1a1229e18988f06a04a12dca"
+
+[[package]]
+name = "time-macros"
+version = "0.2.27"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2e70e4c5a0e0a8a4823ad65dfe1a6930e4f4d756dcd9dd7939022b5e8c501215"
+dependencies = [
+ "num-conv",
+ "time-core",
+]
+
+[[package]]
+name = "unicode-ident"
+version = "1.0.24"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e6e4313cd5fcd3dad5cafa179702e2b244f760991f45397d14d4ebf38247da75"

hosts/temmie/services/userweb/apache-log-processor/Cargo.toml

@@ -0,0 +1,19 @@
+[package]
+name = "apache-log-processor"
+version = "0.1.0"
+edition = "2024"
+autobins = false
+license = "MIT"
+authors = [
+  "projects@pvv.ntnu.no",
+]
+
+[dependencies]
+nix = { version = "0.31.3", features = ["event", "fs", "user"] }
+time = { version = "0.3.47", features = ["formatting", "local-offset"] }
+
+
+[[bin]]
+name = "apache-log-processor"
+bench = false
+path = "src/main.rs"

hosts/temmie/services/userweb/apache-log-processor/default.nix

@@ -0,0 +1,33 @@
+{
+  lib
+, rustPlatform
+, stdenv
+}:
+let
+  cargoToml = fromTOML (builtins.readFile ./Cargo.toml);
+  cargoLock = ./Cargo.lock;
+  mainProgram = (lib.head cargoToml.bin).name;
+  pname = cargoToml.package.name;
+in
+rustPlatform.buildRustPackage {
+  inherit pname;
+  inherit (cargoToml.package) version;
+  src = lib.fileset.toSource {
+    root = ./.;
+    fileset = lib.fileset.unions [
+      ./Cargo.toml
+      ./Cargo.lock
+      ./src
+    ];
+  };
+
+  cargoLock.lockFile = cargoLock;
+
+  doCheck = true;
+
+  meta = with lib; {
+    license = licenses.mit;
+    platforms = platforms.linux;
+    inherit mainProgram;
+  };
+}

hosts/temmie/services/userweb/apache-log-processor/src/main.rs

@@ -0,0 +1,322 @@
+use nix::{
+    errno::Errno,
+    fcntl::{FcntlArg, OFlag, fcntl, open},
+    sys::{
+        epoll::{Epoll, EpollCreateFlags, EpollEvent, EpollFlags, EpollTimeout},
+        stat::Mode,
+    },
+    // unistd::{User, getegid, geteuid, read, setegid, seteuid, write},
+    unistd::{User, read, write},
+};
+use std::{
+    collections::VecDeque,
+    os::fd::{AsFd, BorrowedFd, OwnedFd},
+    path::PathBuf,
+    process::exit,
+};
+use time::{OffsetDateTime, format_description};
+
+const READ_BUFFER_SIZE: usize = 8 * 1024;
+
+#[derive(Debug, Clone, Copy)]
+enum LogMode {
+    Access,
+    Error,
+}
+
+fn main() -> Result<(), String> {
+    let log_mode = match std::env::args().nth(1).as_deref() {
+        Some("access") => LogMode::Access,
+        Some("error") => LogMode::Error,
+        Some(other) => {
+            return Err(format!(
+                "invalid log mode `{other}`; expected `access` or `error`"
+            ));
+        }
+        None => return Err("missing log mode argument; expected `access` or `error`".to_string()),
+    };
+
+    let tee_file = match log_mode {
+        LogMode::Access => None,
+        LogMode::Error => Some(
+            open(
+                &PathBuf::from("/var/log/httpd/error.log"),
+                OFlag::O_WRONLY | OFlag::O_APPEND | OFlag::O_CREAT | OFlag::O_CLOEXEC,
+                Mode::S_IRUSR | Mode::S_IWUSR,
+            )
+            .map_err(|error| format!("failed to open error log for teeing: {error}"))?,
+        ),
+    };
+
+    let stdin = std::io::stdin();
+
+    fcntl(stdin.as_fd(), FcntlArg::F_GETFL)
+        .map(OFlag::from_bits_retain)
+        .map(|flags| FcntlArg::F_SETFL(flags | OFlag::O_NONBLOCK))
+        .and_then(|flags| fcntl(stdin.as_fd(), flags))
+        .map_err(|error| format!("failed to make stdin nonblocking: {error}"))?;
+
+    let epoll = Epoll::new(EpollCreateFlags::EPOLL_CLOEXEC)
+        .map_err(|error| format!("failed to create epoll instance: {error}"))?;
+
+    epoll
+        .add(
+            stdin.as_fd(),
+            EpollEvent::new(
+                EpollFlags::EPOLLIN | EpollFlags::EPOLLERR | EpollFlags::EPOLLHUP,
+                0,
+            ),
+        )
+        .map_err(|error| format!("failed to register stdin with epoll: {error}"))?;
+
+    if let Err(error) = event_loop(log_mode, epoll, stdin.as_fd(), tee_file) {
+        eprintln!("Error: {error}");
+        exit(1);
+    }
+
+    Ok(())
+}
+
+fn event_loop(
+    log_mode: LogMode,
+    epoll: Epoll,
+    stdin_fd: BorrowedFd<'_>,
+    mut tee_file: Option<OwnedFd>,
+) -> Result<(), String> {
+    let mut events = [EpollEvent::empty(); 1];
+    let mut pending = VecDeque::new();
+
+    loop {
+        let ready = loop {
+            match epoll.wait(&mut events, EpollTimeout::NONE) {
+                Ok(ready) => break ready,
+                Err(Errno::EINTR) => continue,
+                Err(error) => {
+                    return Err(format!("epoll wait failed: {error}"));
+                }
+            }
+        };
+
+        if ready == 0 {
+            continue;
+        }
+
+        let mut scratch = [0u8; READ_BUFFER_SIZE];
+
+        let eof = loop {
+            match read(stdin_fd, &mut scratch) {
+                Ok(0) => break true,
+                Ok(read_bytes) => pending.extend(scratch[..read_bytes].iter().copied()),
+                Err(Errno::EINTR) => continue,
+                Err(Errno::EAGAIN) => break false,
+                Err(error) => {
+                    return Err(format!("failed to read from stdin: {error}"));
+                }
+            }
+        };
+
+        while let Some(newline_index) = pending.iter().position(|byte| *byte == b'\n') {
+            let line = pending.make_contiguous();
+            process_line(log_mode, &line[..=newline_index], &mut tee_file)?;
+            pending.drain(..=newline_index);
+        }
+
+        if eof {
+            if !pending.is_empty() {
+                process_line(log_mode, pending.make_contiguous(), &mut tee_file)?;
+                pending.clear();
+            }
+            return Ok(());
+        }
+    }
+}
+
+fn process_line(
+    log_mode: LogMode,
+    line: &[u8],
+    tee_file: &mut Option<OwnedFd>,
+) -> Result<(), String> {
+    if let Some(tee_file) = tee_file.as_ref() {
+        write_all_fd(tee_file, line).map_err(|error| {
+            format!("failed to append to APACHE_LOG_PROCESSOR_TEE_FILE: {error}")
+        })?;
+    }
+
+    if let Some(user) =
+        parse_username_from_line(line).and_then(|name| User::from_name(name).ok().flatten())
+    {
+        // let identity = EffectiveIdentity::switch_to(&user).map_err(|error| {
+        //     format!(
+        //         "failed to switch effective identity to {} (uid {}, gid {}): {error}",
+        //         user.name, user.uid, user.gid
+        //     )
+        // })?;
+
+        let result: Result<(), String> = (|| {
+            let dir = user.dir.join("nobackup/weblogs");
+
+            if !dir.is_dir() {
+                return Err(format!(
+                    "logs directory {} does not exist for user {}",
+                    dir.display(),
+                    user.name
+                ));
+            }
+
+            let now = OffsetDateTime::now_local()
+                .unwrap_or_else(|_| OffsetDateTime::now_utc())
+                .format(&format_description::parse("[year]-[month]-[day]").unwrap())
+                .map_err(|error| {
+                    format!("failed to format current date for log file name: {error}")
+                })?;
+
+            let logfile = dir.join(match log_mode {
+                LogMode::Access => format!("access-{now}.log"),
+                LogMode::Error => format!("error-{now}.log"),
+            });
+
+            let fd = open(
+                &logfile,
+                OFlag::O_WRONLY | OFlag::O_APPEND | OFlag::O_CREAT | OFlag::O_CLOEXEC,
+                Mode::S_IRUSR
+                    | Mode::S_IWUSR
+                    | Mode::S_IRGRP
+                    | Mode::S_IROTH
+                    | Mode::S_IWGRP
+                    | Mode::S_IWOTH,
+            )
+            .map_err(|error| format!("failed to open log file for user {}: {error}", user.name))?;
+
+            write_all_fd(fd.as_fd(), line).map_err(|error| {
+                format!(
+                    "failed to append to log file for user {}: {error}",
+                    user.name
+                )
+            })?;
+
+            Ok(())
+        })();
+
+        if let Err(error) = result {
+            eprintln!("Error processing log line for user {}: {error}", user.name);
+        }
+
+        // identity.restore().map_err(|error| {
+        //     format!(
+        //         "failed to restore original effective identity after handling {}: {error}",
+        //         user.name
+        //     )
+        // })?;
+    }
+
+    Ok(())
+}
+
+fn parse_username_from_line(line: &[u8]) -> Option<&str> {
+    line.splitn(8, |&b| b == b' ')
+        .nth(6)
+        .and_then(|path| {
+            path.strip_prefix(b"/~")
+                .and_then(|rest| rest.split(|&b| b == b'/').next())
+        })
+        .or_else(|| {
+            line.windows(b"/home/pvv/".len())
+                .enumerate()
+                .find_map(|(start, window)| {
+                    (window == b"/home/pvv/")
+                        .then_some(start + b"/home/pvv/".len())
+                        .and_then(|start| line.get(start..))
+                        .filter(|rest| rest.get(1) == Some(&b'/'))
+                        .and_then(|rest| rest.get(2..))
+                        .and_then(|rest| rest.split(|&b| b == b'/').next())
+                })
+        })
+        .filter(|segment| !segment.is_empty())
+        .and_then(|segment| std::str::from_utf8(segment).ok())
+}
+
+fn write_all_fd<Fd: AsFd>(fd: Fd, mut buffer: &[u8]) -> nix::Result<()> {
+    while !buffer.is_empty() {
+        match write(fd.as_fd(), buffer) {
+            Ok(0) => return Err(Errno::EIO),
+            Ok(written) => buffer = &buffer[written..],
+            Err(Errno::EINTR) => continue,
+            Err(error) => return Err(error),
+        }
+    }
+
+    Ok(())
+}
+
+// struct EffectiveIdentity {
+//     saved_euid: nix::unistd::Uid,
+//     saved_egid: nix::unistd::Gid,
+//     restored: bool,
+// }
+
+// impl EffectiveIdentity {
+//     fn switch_to(user: &User) -> nix::Result<Self> {
+//         let guard = Self {
+//             saved_euid: geteuid(),
+//             saved_egid: getegid(),
+//             restored: false,
+//         };
+
+//         setegid(user.gid)?;
+//         if let Err(error) = seteuid(user.uid) {
+//             let _ = setegid(guard.saved_egid);
+//             return Err(error);
+//         }
+
+//         Ok(guard)
+//     }
+
+//     fn restore(mut self) -> nix::Result<()> {
+//         let restore_uid = seteuid(self.saved_euid);
+//         let restore_gid = setegid(self.saved_egid);
+//         self.restored = true;
+
+//         restore_uid?;
+//         restore_gid?;
+//         Ok(())
+//     }
+// }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_parse_user_from_access_log() {
+        let inputs = [(
+            "1.2.3.4 - - [25/May/2026:10:07:24 +0200] \"GET /~oysteikt/ HTTP/2.0\" 200 3708",
+            "oysteikt",
+        )];
+
+        for (line, expected_user) in inputs {
+            let parsed_user = parse_username_from_line(line.as_bytes());
+            assert_eq!(
+                parsed_user,
+                Some(expected_user),
+                "Failed to parse user from line: {line}"
+            );
+        }
+    }
+
+    #[test]
+    fn test_parse_user_from_error_log() {
+        let inputs = [(
+            "[Sat May 09 20:45:21.480016 2026] [authz_core:error] [pid 3555:tid 3617] [remote 1::2:42000] AH01630: client denied by server configuration: /home/pvv/d/oysteikt/web-docs/.git",
+            "oysteikt",
+        )];
+
+        for (line, expected_user) in inputs {
+            let parsed_user = parse_username_from_line(line.as_bytes());
+            assert_eq!(
+                parsed_user,
+                Some(expected_user),
+                "Failed to parse user from line: {line}"
+            );
+        }
+    }
+}

hosts/temmie/services/userweb/default.nix

@@ -11,6 +11,8 @@ let
     upload_max_filesize = "40M";
   });
 
+  apache-log-processor = pkgs.callPackage ./apache-log-processor { };
+
   # https://nixos.org/manual/nixpkgs/stable/#ssec-php-user-guide-installing-with-extensions
   phpEnv = pkgs.php.buildEnv {
     extensions = { all, ... }: with all; [
@@ -193,10 +195,11 @@ in
       }
     ];
 
+    logPerVirtualHost = false;
+
     extraConfig = ''
       TraceEnable on
       LogLevel warn rewrite:trace3
-      ScriptLog ${cfg.logDir}/cgi.log
     '';
 
     virtualHosts."temmie.pvv.ntnu.no" = {
@@ -208,6 +211,11 @@ in
       ];
 
       extraConfig = ''
+        CustomLog "${cfg.logDir}/access.log" combined
+        CustomLog "|${lib.getExe apache-log-processor} access" combined
+        ErrorLog "|${lib.getExe apache-log-processor} error"
+        ScriptLog "${cfg.logDir}/cgi.log"
+
         UserDir ${lib.concatMapStringsSep " " (l: "/home/pvv/${l}/*/web-docs") homeLetters}
         UserDir disabled root
         AddHandler cgi-script .cgi