Merge pull request #1459 from dermetfan/fix-1429-backport-nix-2.25

(backport to nix-2.25) hydra-eval-jobset: do not wait on n-e-j inside transaction

flake.lock

@@ -1,10 +1,30 @@
 {
   "nodes": {
+    "libgit2": {
+      "flake": false,
+      "locked": {
+        "lastModified": 1715853528,
+        "narHash": "sha256-J2rCxTecyLbbDdsyBWn9w7r3pbKRMkI9E7RvRgAqBdY=",
+        "owner": "libgit2",
+        "repo": "libgit2",
+        "rev": "36f7e21ad757a3dacc58cf7944329da6bc1d6e96",
+        "type": "github"
+      },
+      "original": {
+        "owner": "libgit2",
+        "ref": "v1.8.1",
+        "repo": "libgit2",
+        "type": "github"
+      }
+    },
     "nix": {
       "inputs": {
         "flake-compat": [],
         "flake-parts": [],
         "git-hooks-nix": [],
+        "libgit2": [
+          "libgit2"
+        ],
         "nixpkgs": [
           "nixpkgs"
         ],
@@ -12,16 +32,16 @@
         "nixpkgs-regression": []
       },
       "locked": {
-        "lastModified": 1739899400,
-        "narHash": "sha256-q/RgA4bB7zWai4oPySq9mch7qH14IEeom2P64SXdqHs=",
+        "lastModified": 1739390454,
+        "narHash": "sha256-mIpJgIwPS4o4xYhN1B+/fHESEXoxpu6nVoZTzZ0MfTg=",
         "owner": "NixOS",
         "repo": "nix",
-        "rev": "e310c19a1aeb1ce1ed4d41d5ab2d02db596e0918",
+        "rev": "d652513e4519ed4eb48c92f8670e5a71c7793fc3",
         "type": "github"
       },
       "original": {
         "owner": "NixOS",
-        "ref": "2.26-maintenance",
+        "ref": "2.25-maintenance",
         "repo": "nix",
         "type": "github"
       }
@@ -29,11 +49,11 @@
     "nix-eval-jobs": {
       "flake": false,
       "locked": {
-        "lastModified": 1739500569,
-        "narHash": "sha256-3wIReAqdTALv39gkWXLMZQvHyBOc3yPkWT2ZsItxedY=",
+        "lastModified": 1739426028,
+        "narHash": "sha256-1dZLPw+nlFQzzswfyTxW+8VF1AJ4ZvoYvLTjlHiz1SA=",
         "owner": "nix-community",
         "repo": "nix-eval-jobs",
-        "rev": "4b392b284877d203ae262e16af269f702df036bc",
+        "rev": "6d4fd5a93d7bc953ffa4dcd6d53ad7056a71eff7",
         "type": "github"
       },
       "original": {
@@ -44,22 +64,23 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1739461644,
-        "narHash": "sha256-1o1qR0KYozYGRrnqytSpAhVBYLNBHX+Lv6I39zGRzKM=",
+        "lastModified": 1726688310,
+        "narHash": "sha256-Xc9lEtentPCEtxc/F1e6jIZsd4MPDYv4Kugl9WtXlz0=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "97a719c9f0a07923c957cf51b20b329f9fb9d43f",
+        "rev": "dbebdd67a6006bb145d98c8debf9140ac7e651d0",
         "type": "github"
       },
       "original": {
         "owner": "NixOS",
-        "ref": "nixos-24.11-small",
+        "ref": "nixos-24.05-small",
         "repo": "nixpkgs",
         "type": "github"
       }
     },
     "root": {
       "inputs": {
+        "libgit2": "libgit2",
         "nix": "nix",
         "nix-eval-jobs": "nix-eval-jobs",
         "nixpkgs": "nixpkgs"

flake.nix

@@ -1,11 +1,17 @@
 {
   description = "A Nix-based continuous build system";
 
-  inputs.nixpkgs.url = "github:NixOS/nixpkgs/nixos-24.11-small";
+  inputs.nixpkgs.url = "github:NixOS/nixpkgs/nixos-24.05-small";
+
+  inputs.libgit2 = {
+    url = "github:libgit2/libgit2/v1.8.1";
+    flake = false;
+  };
 
   inputs.nix = {
-    url = "github:NixOS/nix/2.26-maintenance";
+    url = "github:NixOS/nix/2.25-maintenance";
     inputs.nixpkgs.follows = "nixpkgs";
+    inputs.libgit2.follows = "libgit2";
 
     # hide nix dev tooling from our lock file
     inputs.flake-parts.follows = "";
@@ -81,13 +87,7 @@
           inherit (nixpkgs.lib) fileset;
           inherit (self.packages.${system}) nix-eval-jobs;
           rawSrc = self;
-          inherit (nix.packages.${system})
-            nix-util
-            nix-store
-            nix-main
-            nix-cmd
-            nix-cli
-            ;
+          nix = nix.packages.${system}.nix;
           nix-perl-bindings = nix.hydraJobs.perlBindings.${system};
         };
         default = self.packages.${system}.hydra;

foreman/start-manual.sh

@@ -1,6 +1,6 @@
 #!/bin/sh
 
-exec mdbook serve \
+mdbook serve \
   --port 63332 \
   --dest-dir ./.hydra-data/manual \
   ./doc/manual/

meson.build

@@ -8,22 +8,22 @@ project('hydra', 'cpp',
   ],
 )
 
-nix_util_dep = dependency('nix-util', required: true)
 nix_store_dep = dependency('nix-store', required: true)
 nix_main_dep = dependency('nix-main', required: true)
+nix_expr_dep = dependency('nix-expr', required: true)
+nix_flake_dep = dependency('nix-flake', required: true)
+nix_cmd_dep = dependency('nix-cmd', required: true)
 
 # Nix need extra flags not provided in its pkg-config files.
 nix_dep = declare_dependency(
   dependencies: [
-    nix_util_dep,
     nix_store_dep,
     nix_main_dep,
+    nix_expr_dep,
+    nix_flake_dep,
+    nix_cmd_dep,
   ],
-  compile_args: [
-    '-include', 'nix/config-util.hh',
-    '-include', 'nix/config-store.hh',
-    '-include', 'nix/config-main.hh',
-  ],
+  compile_args: ['-include', 'nix/config.h'],
 )
 
 pqxx_dep = dependency('libpqxx', required: true)

package.nix

@@ -8,11 +8,7 @@
 
 , perlPackages
 
-, nix-util
-, nix-store
-, nix-main
-, nix-cmd
-, nix-cli
+, nix
 , nix-perl-bindings
 , git
 
@@ -166,7 +162,7 @@ stdenv.mkDerivation (finalAttrs: {
     nukeReferences
     pkg-config
     mdbook
-    nix-cli
+    nix
     perlDeps
     perl
     unzip
@@ -176,10 +172,7 @@ stdenv.mkDerivation (finalAttrs: {
     libpqxx
     openssl
     libxslt
-    nix-util
-    nix-store
-    nix-main
-    nix-cmd
+    nix
     perlDeps
     perl
     boost
@@ -206,14 +199,13 @@ stdenv.mkDerivation (finalAttrs: {
     glibcLocales
     libressl.nc
     python3
-    nix-cli
   ];
 
   hydraPath = lib.makeBinPath (
     [
       subversion
       openssh
-      nix-cli
+      nix
       coreutils
       findutils
       pixz
@@ -274,7 +266,7 @@ stdenv.mkDerivation (finalAttrs: {
             --prefix PATH ':' $out/bin:$hydraPath \
             --set HYDRA_RELEASE ${version} \
             --set HYDRA_HOME $out/libexec/hydra \
-            --set NIX_RELEASE ${nix-cli.name or "unknown"} \
+            --set NIX_RELEASE ${nix.name or "unknown"} \
             --set NIX_EVAL_JOBS_RELEASE ${nix-eval-jobs.name or "unknown"}
     done
   '';
@@ -282,5 +274,5 @@ stdenv.mkDerivation (finalAttrs: {
   dontStrip = true;
 
   meta.description = "Build of Hydra on ${stdenv.system}";
-  passthru = { inherit perlDeps; };
+  passthru = { inherit perlDeps nix; };
 })

src/hydra-build-step/hydra-build-step.cc

@@ -1,213 +0,0 @@
-/* This is a helper program that performs a build step, i.e. a single
-   derivation. In addition to a derivation path, it takes three store
-   URLs as arguments:
-
-   * --store: The store that will hold the resulting store paths
-       (typically a binary cache).
-
-   * --eval-store: The store that holds the .drv files, as produced by
-       hydra-evaluator.
-
-   * --build-store: The store that performs the build (often a
-       SSHStore for remote builds).
-
-   The build log is written to the path indicated by --log-file.
-*/
-
-#include "util.hh"
-#include "shared.hh"
-#include "common-eval-args.hh"
-#include "store-api.hh"
-#include "build-result.hh"
-#include "derivations.hh"
-#include "worker-protocol.hh"
-
-#include <chrono>
-
-using namespace nix;
-
-// FIXME: cut&paste
-static std::string_view getS(const std::vector<Logger::Field> & fields, size_t n)
-{
-    assert(n < fields.size());
-    assert(fields[n].type == Logger::Field::tString);
-    return fields[n].s;
-}
-
-void mainWrapped(std::list<std::string> args)
-{
-    verbosity = lvlError;
-
-    struct MyArgs : MixEvalArgs, MixCommonArgs, RootArgs
-    {
-        Path drvPath;
-        std::optional<std::string> buildStoreUrl;
-        std::optional<Path> logPath;
-        std::optional<uint64_t> maxOutputSize;
-
-        MyArgs() : MixCommonArgs("hydra-build-step")
-        {
-            expectArg("drv-path", &drvPath);
-
-            addFlag({
-                .longName = "build-store",
-                .description = "The Nix store to use for building the derivation.",
-                //.category = category,
-                .labels = {"store-url"},
-                .handler = {&buildStoreUrl},
-            });
-
-            addFlag({
-                .longName = "log-file",
-                .description = "The path to the build log.",
-                .labels = {"path"},
-                .handler = {&logPath},
-            });
-
-            addFlag({
-                .longName = "max-output-size",
-                .description = "Maximum size of the outputs.",
-                .labels = {"bytes"},
-                .handler = {&maxOutputSize},
-            });
-        }
-    };
-
-    /* A logger that intercepts all build log lines and writes them to
-       the log file. */
-    MyArgs myArgs;
-    myArgs.parseCmdline(args);
-
-    struct MyLogger : public Logger
-    {
-        Logger & prev;
-        AutoCloseFD logFile;
-
-        MyLogger(Logger & prev, Path logPath) : prev(prev)
-        {
-            logFile = open(logPath.c_str(), O_CREAT | O_TRUNC | O_WRONLY, 0666);
-            if (!logFile)
-                throw SysError("creating log file '%s'", logPath);
-        }
-
-        void log(Verbosity lvl, std::string_view s) override
-        { prev.log(lvl, s); }
-
-        void logEI(const ErrorInfo & ei) override
-        { prev.logEI(ei); }
-
-        void writeToStdout(std::string_view s) override
-        { prev.writeToStdout(s); }
-
-        void result(ActivityId act, ResultType type, const Fields & fields) override
-        {
-            if (type == resBuildLogLine)
-                writeLine(logFile.get(), std::string(getS(fields, 0)));
-            else
-                prev.result(act, type, fields);
-        }
-    };
-
-    auto destStore = openStore();
-    auto evalStore = myArgs.evalStoreUrl ? openStore(*myArgs.evalStoreUrl) : destStore;
-    auto buildStore = myArgs.buildStoreUrl ? openStore(*myArgs.buildStoreUrl) : destStore;
-
-    auto drvPath = evalStore->parseStorePath(myArgs.drvPath);
-
-    auto drv = evalStore->readDerivation(drvPath);
-    BasicDerivation basicDrv(drv);
-
-    uint64_t overhead = 0;
-
-    /* Gather the inputs. */
-    StorePathSet inputs;
-
-    for (auto & p : drv.inputSrcs)
-        inputs.insert(p);
-
-    for (auto & [drvPath, node] : drv.inputDrvs.map) {
-        auto drv2 = evalStore->readDerivation(drvPath);
-        for (auto & name : node.value) {
-            if (auto i = get(drv2.outputs, name)) {
-                auto outPath = i->path(*evalStore, drv2.name, name);
-                inputs.insert(*outPath);
-                basicDrv.inputSrcs.insert(*outPath);
-            }
-        }
-    }
-
-    /* Ensure that the inputs exist in the destination store (so that
-       the builder can substitute them from the destination
-       store). This is a no-op for regular stores, but for the binary
-       cache store, this will copy the inputs to the binary cache from
-       the local store. */
-    {
-        auto now1 = std::chrono::steady_clock::now();
-
-        debug("sending closure of '%s' to '%s'",
-            evalStore->printStorePath(drvPath), destStore->getUri());
-
-        if (evalStore != destStore)
-            copyClosure(*evalStore, *destStore, drv.inputSrcs, NoRepair, NoCheckSigs);
-
-        copyClosure(*destStore, *buildStore, inputs, NoRepair, NoCheckSigs, Substitute);
-
-        auto now2 = std::chrono::steady_clock::now();
-
-        overhead += std::chrono::duration_cast<std::chrono::milliseconds>(now2 - now1).count();
-    }
-
-    /* Perform the build. */
-    if (myArgs.logPath)
-        logger = new MyLogger(*logger, *myArgs.logPath);
-
-    auto buildResult = buildStore->buildDerivation(drvPath, basicDrv);
-
-    /* Copy the output paths from the build store to the destination
-       store. */
-    size_t totalNarSize = 0;
-
-    if (buildResult.success()) {
-
-        std::map<StorePath, ValidPathInfo> infos;
-        StorePathSet outputs;
-        for (auto & [output, realisation] : buildResult.builtOutputs) {
-            auto info = buildStore->queryPathInfo(realisation.outPath);
-            totalNarSize += info->narSize;
-            infos.insert_or_assign(info->path, *info);
-            outputs.insert(info->path);
-        }
-
-        if ((!myArgs.maxOutputSize || totalNarSize <= *myArgs.maxOutputSize)
-            && buildStore != destStore)
-        {
-            debug("copying outputs of '%s' from '%s' (%d bytes)",
-                buildStore->printStorePath(drvPath), buildStore->getUri(), totalNarSize);
-
-            auto now1 = std::chrono::steady_clock::now();
-
-            copyPaths(*buildStore, *destStore, outputs, NoRepair, NoCheckSigs);
-
-            auto now2 = std::chrono::steady_clock::now();
-
-            overhead += std::chrono::duration_cast<std::chrono::milliseconds>(now2 - now1).count();
-        }
-    }
-
-    FdSink to { STDOUT_FILENO };
-    WorkerProto::WriteConn wconn {
-        .to = to,
-        // Hardcode latest version because we are deploying hydra
-        // itself atomically
-        .version = PROTOCOL_VERSION,
-    };
-    WorkerProto::write(*evalStore, wconn, buildResult);
-}
-
-int main(int argc, char * * argv)
-{
-    return handleExceptions(argv[0], [&]() {
-        initNix();
-        mainWrapped(argvToStrings(argc, argv));
-    });
-}

src/hydra-build-step/meson.build

@@ -1,14 +0,0 @@
-srcs = files(
-  'hydra-build-step.cc',
-)
-
-hydra_build_step = executable('hydra-build-step',
-  'hydra-build-step.cc',
-  srcs,
-  dependencies: [
-    libhydra_dep,
-    nix_dep,
-    dependency('nix-cmd', required: true)
-  ],
-  install: true,
-)

src/hydra-queue-runner/build-remote.cc

@@ -1,27 +1,330 @@
-#include <math.h>
+#include <algorithm>
+#include <cmath>
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
 
 #include "build-result.hh"
+#include "path.hh"
+#include "legacy-ssh-store.hh"
 #include "serve-protocol.hh"
+#include "serve-protocol-impl.hh"
 #include "state.hh"
 #include "current-process.hh"
 #include "processes.hh"
 #include "util.hh"
+#include "serve-protocol.hh"
+#include "serve-protocol-impl.hh"
+#include "ssh.hh"
 #include "finally.hh"
 #include "url.hh"
-#include "worker-protocol.hh"
 
 using namespace nix;
 
+bool ::Machine::isLocalhost() const
+{
+    return storeUri.params.empty() && std::visit(overloaded {
+        [](const StoreReference::Auto &) {
+            return true;
+        },
+        [](const StoreReference::Specified & s) {
+            return
+               (s.scheme == "local" || s.scheme == "unix") ||
+               ((s.scheme == "ssh" || s.scheme == "ssh-ng") &&
+                s.authority == "localhost");
+        },
+    }, storeUri.variant);
+}
+
 namespace nix::build_remote {
 
-static Path createLogFileDir(const std::string & logDir, const StorePath & drvPath)
+static std::unique_ptr<SSHMaster::Connection> openConnection(
+    ::Machine::ptr machine, SSHMaster & master)
+{
+    Strings command = {"nix-store", "--serve", "--write"};
+    if (machine->isLocalhost()) {
+        command.push_back("--builders");
+        command.push_back("");
+    } else {
+        auto remoteStore = machine->storeUri.params.find("remote-store");
+        if (remoteStore != machine->storeUri.params.end()) {
+            command.push_back("--store");
+            command.push_back(shellEscape(remoteStore->second));
+        }
+    }
+
+    auto ret = master.startCommand(std::move(command), {
+        "-a", "-oBatchMode=yes", "-oConnectTimeout=60", "-oTCPKeepAlive=yes"
+    });
+
+    // XXX: determine the actual max value we can use from /proc.
+
+    // FIXME: Should this be upstreamed into `startCommand` in Nix?
+
+    int pipesize = 1024 * 1024;
+
+    fcntl(ret->in.get(), F_SETPIPE_SZ, &pipesize);
+    fcntl(ret->out.get(), F_SETPIPE_SZ, &pipesize);
+
+    return ret;
+}
+
+
+static void copyClosureTo(
+    ::Machine::Connection & conn,
+    Store & destStore,
+    const StorePathSet & paths,
+    SubstituteFlag useSubstitutes = NoSubstitute)
+{
+    StorePathSet closure;
+    destStore.computeFSClosure(paths, closure);
+
+    /* Send the "query valid paths" command with the "lock" option
+       enabled. This prevents a race where the remote host
+       garbage-collect paths that are already there. Optionally, ask
+       the remote host to substitute missing paths. */
+    // FIXME: substitute output pollutes our build log
+    /* Get back the set of paths that are already valid on the remote
+       host. */
+    auto present = conn.queryValidPaths(
+        destStore, true, closure, useSubstitutes);
+
+    if (present.size() == closure.size()) return;
+
+    auto sorted = destStore.topoSortPaths(closure);
+
+    StorePathSet missing;
+    for (auto i = sorted.rbegin(); i != sorted.rend(); ++i)
+        if (!present.count(*i)) missing.insert(*i);
+
+    printMsg(lvlDebug, "sending %d missing paths", missing.size());
+
+    std::unique_lock<std::timed_mutex> sendLock(conn.machine->state->sendLock,
+        std::chrono::seconds(600));
+
+    conn.to << ServeProto::Command::ImportPaths;
+    destStore.exportPaths(missing, conn.to);
+    conn.to.flush();
+
+    if (readInt(conn.from) != 1)
+        throw Error("remote machine failed to import closure");
+}
+
+
+// FIXME: use Store::topoSortPaths().
+static StorePaths reverseTopoSortPaths(const std::map<StorePath, UnkeyedValidPathInfo> & paths)
+{
+    StorePaths sorted;
+    StorePathSet visited;
+
+    std::function<void(const StorePath & path)> dfsVisit;
+
+    dfsVisit = [&](const StorePath & path) {
+        if (!visited.insert(path).second) return;
+
+        auto info = paths.find(path);
+        auto references = info == paths.end() ? StorePathSet() : info->second.references;
+
+        for (auto & i : references)
+            /* Don't traverse into paths that don't exist.  That can
+               happen due to substitutes for non-existent paths. */
+            if (i != path && paths.count(i))
+                dfsVisit(i);
+
+        sorted.push_back(path);
+    };
+
+    for (auto & i : paths)
+        dfsVisit(i.first);
+
+    return sorted;
+}
+
+static std::pair<Path, AutoCloseFD> openLogFile(const std::string & logDir, const StorePath & drvPath)
 {
     std::string base(drvPath.to_string());
     auto logFile = logDir + "/" + std::string(base, 0, 2) + "/" + std::string(base, 2);
 
     createDirs(dirOf(logFile));
 
-    return logFile;
+    AutoCloseFD logFD = open(logFile.c_str(), O_CREAT | O_TRUNC | O_WRONLY, 0666);
+    if (!logFD) throw SysError("creating log file ‘%s’", logFile);
+
+    return {std::move(logFile), std::move(logFD)};
+}
+
+static BasicDerivation sendInputs(
+    State & state,
+    Step & step,
+    Store & localStore,
+    Store & destStore,
+    ::Machine::Connection & conn,
+    unsigned int & overhead,
+    counter & nrStepsWaiting,
+    counter & nrStepsCopyingTo
+)
+{
+    /* Replace the input derivations by their output paths to send a
+       minimal closure to the builder.
+
+       `tryResolve` currently does *not* rewrite input addresses, so it
+       is safe to do this in all cases. (It should probably have a mode
+       to do that, however, but we would not use it here.)
+     */
+    BasicDerivation basicDrv = ({
+        auto maybeBasicDrv = step.drv->tryResolve(destStore, &localStore);
+        if (!maybeBasicDrv)
+            throw Error(
+                "the derivation '%s' can’t be resolved. It’s probably "
+                "missing some outputs",
+                localStore.printStorePath(step.drvPath));
+        *maybeBasicDrv;
+    });
+
+    /* Ensure that the inputs exist in the destination store. This is
+       a no-op for regular stores, but for the binary cache store,
+       this will copy the inputs to the binary cache from the local
+       store. */
+    if (&localStore != &destStore) {
+        copyClosure(localStore, destStore,
+            step.drv->inputSrcs,
+            NoRepair, NoCheckSigs, NoSubstitute);
+    }
+
+    {
+        auto mc1 = std::make_shared<MaintainCount<counter>>(nrStepsWaiting);
+        mc1.reset();
+        MaintainCount<counter> mc2(nrStepsCopyingTo);
+
+        printMsg(lvlDebug, "sending closure of ‘%s’ to ‘%s’",
+            localStore.printStorePath(step.drvPath), conn.machine->storeUri.render());
+
+        auto now1 = std::chrono::steady_clock::now();
+
+        /* Copy the input closure. */
+        if (conn.machine->isLocalhost()) {
+            StorePathSet closure;
+            destStore.computeFSClosure(basicDrv.inputSrcs, closure);
+            copyPaths(destStore, localStore, closure, NoRepair, NoCheckSigs, NoSubstitute);
+        } else {
+            copyClosureTo(conn, destStore, basicDrv.inputSrcs, Substitute);
+        }
+
+        auto now2 = std::chrono::steady_clock::now();
+
+        overhead += std::chrono::duration_cast<std::chrono::milliseconds>(now2 - now1).count();
+    }
+
+    return basicDrv;
+}
+
+static BuildResult performBuild(
+    ::Machine::Connection & conn,
+    Store & localStore,
+    StorePath drvPath,
+    const BasicDerivation & drv,
+    const ServeProto::BuildOptions & options,
+    counter & nrStepsBuilding
+)
+{
+    conn.putBuildDerivationRequest(localStore, drvPath, drv, options);
+
+    BuildResult result;
+
+    time_t startTime, stopTime;
+
+    startTime = time(0);
+    {
+        MaintainCount<counter> mc(nrStepsBuilding);
+        result = ServeProto::Serialise<BuildResult>::read(localStore, conn);
+    }
+    stopTime = time(0);
+
+    if (!result.startTime) {
+        // If the builder gave `startTime = 0`, use our measurements
+        // instead of the builder's.
+        //
+        // Note: this represents the duration of a single round, rather
+        // than all rounds.
+        result.startTime = startTime;
+        result.stopTime = stopTime;
+    }
+
+    // If the protocol was too old to give us `builtOutputs`, initialize
+    // it manually by introspecting the derivation.
+    if (GET_PROTOCOL_MINOR(conn.remoteVersion) < 6)
+    {
+        // If the remote is too old to handle CA derivations, we can’t get this
+        // far anyways
+        assert(drv.type().hasKnownOutputPaths());
+        DerivationOutputsAndOptPaths drvOutputs = drv.outputsAndOptPaths(localStore);
+        // Since this a `BasicDerivation`, `staticOutputHashes` will not
+        // do any real work.
+        auto outputHashes = staticOutputHashes(localStore, drv);
+        for (auto & [outputName, output] : drvOutputs) {
+            auto outputPath = output.second;
+            // We’ve just asserted that the output paths of the derivation
+            // were known
+            assert(outputPath);
+            auto outputHash = outputHashes.at(outputName);
+            auto drvOutput = DrvOutput { outputHash, outputName };
+            result.builtOutputs.insert_or_assign(
+                std::move(outputName),
+                Realisation { drvOutput, *outputPath });
+        }
+    }
+
+    return result;
+}
+
+static void copyPathFromRemote(
+    ::Machine::Connection & conn,
+    NarMemberDatas & narMembers,
+    Store & localStore,
+    Store & destStore,
+    const ValidPathInfo & info
+)
+{
+      /* Receive the NAR from the remote and add it to the
+          destination store. Meanwhile, extract all the info from the
+          NAR that getBuildOutput() needs. */
+      auto source2 = sinkToSource([&](Sink & sink)
+      {
+          /* Note: we should only send the command to dump the store
+              path to the remote if the NAR is actually going to get read
+              by the destination store, which won't happen if this path
+              is already valid on the destination store. Since this
+              lambda function only gets executed if someone tries to read
+              from source2, we will send the command from here rather
+              than outside the lambda. */
+          conn.to << ServeProto::Command::DumpStorePath << localStore.printStorePath(info.path);
+          conn.to.flush();
+
+          TeeSource tee(conn.from, sink);
+          extractNarData(tee, localStore.printStorePath(info.path), narMembers);
+      });
+
+      destStore.addToStore(info, *source2, NoRepair, NoCheckSigs);
+}
+
+static void copyPathsFromRemote(
+    ::Machine::Connection & conn,
+    NarMemberDatas & narMembers,
+    Store & localStore,
+    Store & destStore,
+    const std::map<StorePath, UnkeyedValidPathInfo> & infos
+)
+{
+      auto pathsSorted = reverseTopoSortPaths(infos);
+
+      for (auto & path : pathsSorted) {
+          auto & info = infos.find(path)->second;
+          copyPathFromRemote(
+              conn, narMembers, localStore, destStore,
+              ValidPathInfo { path, info });
+      }
+
 }
 
 }
@@ -30,14 +333,11 @@ static Path createLogFileDir(const std::string & logDir, const StorePath & drvPa
 
 void RemoteResult::updateWithBuildResult(const nix::BuildResult & buildResult)
 {
-    // FIXME: make RemoteResult inherit BuildResult.
+    startTime = buildResult.startTime;
+    stopTime = buildResult.stopTime;
     timesBuilt = buildResult.timesBuilt;
     errorMsg = buildResult.errorMsg;
     isNonDeterministic = buildResult.isNonDeterministic;
-    if (buildResult.startTime && buildResult.stopTime) {
-        startTime = buildResult.startTime;
-        stopTime = buildResult.stopTime;
-    }
 
     switch ((BuildResult::Status) buildResult.status) {
         case BuildResult::Built:
@@ -96,44 +396,38 @@ void State::buildRemote(ref<Store> destStore,
 {
     assert(BuildResult::TimedOut == 8);
 
-    result.logFile = build_remote::createLogFileDir(logDir, step->drvPath);
+    auto [logFile, logFD] = build_remote::openLogFile(logDir, step->drvPath);
+    AutoDelete logFileDel(logFile, false);
+    result.logFile = logFile;
 
     try {
 
-        updateStep(ssBuilding);
-        result.startTime = time(0);
+        updateStep(ssConnecting);
 
-        auto buildStoreUrl = machine->completeStoreReference().render();
+        auto storeRef = machine->completeStoreReference();
 
-        Strings args = {
-            localStore->printStorePath(step->drvPath),
-            "--store", destStore->getUri(),
-            "--eval-store", localStore->getUri(),
-            "--build-store", buildStoreUrl,
-            "--max-silent-time", std::to_string(buildOptions.maxSilentTime),
-            "--timeout", std::to_string(buildOptions.buildTimeout),
-            "--max-build-log-size", std::to_string(buildOptions.maxLogSize),
-            "--max-output-size", std::to_string(maxOutputSize),
-            "--repeat", std::to_string(buildOptions.nrRepeats),
-            "--log-file", result.logFile,
-            // FIXME: step->isDeterministic
+        auto * pSpecified = std::get_if<StoreReference::Specified>(&storeRef.variant);
+        if (!pSpecified || pSpecified->scheme != "ssh") {
+            throw Error("Currently, only (legacy-)ssh stores are supported!");
+        }
+
+        LegacySSHStoreConfig storeConfig {
+            pSpecified->scheme,
+            pSpecified->authority,
+            storeRef.params
         };
 
-        // FIXME: set pid for cancellation
+        auto master = storeConfig.createSSHMaster(
+            false, // no SSH master yet
+            logFD.get());
 
-        auto [status, childStdout] = [&]() {
-            MaintainCount<counter> mc(nrStepsBuilding);
-            return runProgram({
-                .program = "hydra-build-step",
-                .args = std::move(args),
-            });
-        }();
+        // FIXME: rewrite to use Store.
+        auto child = build_remote::openConnection(machine, master);
 
-        #if 0
         {
             auto activeStepState(activeStep->state_.lock());
             if (activeStepState->cancelled) throw Error("step cancelled");
-            activeStepState->pid = conn.store->getConnectionPid();
+            activeStepState->pid = child->sshPid;
         }
 
         Finally clearPid([&]() {
@@ -147,32 +441,76 @@ void State::buildRemote(ref<Store> destStore,
                possibility that we end up killing another
                process. Meh. */
         });
-        #endif
 
-        result.stopTime = time(0);
+        ::Machine::Connection conn {
+            {
+                .to = child->in.get(),
+                .from = child->out.get(),
+                /* Handshake. */
+                .remoteVersion = 0xdadbeef, // FIXME avoid dummy initialize
+            },
+            /*.machine =*/ machine,
+        };
 
-        if (!statusOk(status))
-            throw ExecError(status, fmt("hydra-build-step %s with output:\n%s", statusToString(status), stdout));
+        Finally updateStats([&]() {
+            bytesReceived += conn.from.read;
+            bytesSent += conn.to.written;
+        });
+
+        constexpr ServeProto::Version our_version = 0x206;
+
+        try {
+            conn.remoteVersion = decltype(conn)::handshake(
+                conn.to,
+                conn.from,
+                our_version,
+                machine->storeUri.render());
+        } catch (EndOfFile & e) {
+            child->sshPid.wait();
+            std::string s = chomp(readFile(result.logFile));
+            throw Error("cannot connect to ‘%1%’: %2%", machine->storeUri.render(), s);
+        }
 
-        /* The build was executed successfully, so clear the failure
-           count for this machine. */
         {
             auto info(machine->state->connectInfo.lock());
             info->consecutiveFailures = 0;
         }
 
-        StringSource from { childStdout };
-        /* Read the BuildResult from the child. */
-        WorkerProto::ReadConn rconn {
-            .from = from,
-            // Hardcode latest version because we are deploying hydra
-            // itself atomically
-            .version = PROTOCOL_VERSION,
-        };
-        result.overhead += readNum<uint64_t>(rconn.from);
-        auto totalNarSize = readNum<uint64_t>(rconn.from);
-        auto buildResult = WorkerProto::Serialise<BuildResult>::read(*localStore, rconn);
+        /* Gather the inputs. If the remote side is Nix <= 1.9, we have to
+           copy the entire closure of ‘drvPath’, as well as the required
+           outputs of the input derivations. On Nix > 1.9, we only need to
+           copy the immediate sources of the derivation and the required
+           outputs of the input derivations. */
+        updateStep(ssSendingInputs);
+        BasicDerivation resolvedDrv = build_remote::sendInputs(*this, *step, *localStore, *destStore, conn, result.overhead, nrStepsWaiting, nrStepsCopyingTo);
 
+        logFileDel.cancel();
+
+        /* Truncate the log to get rid of messages about substitutions
+            etc. on the remote system. */
+        if (lseek(logFD.get(), SEEK_SET, 0) != 0)
+            throw SysError("seeking to the start of log file ‘%s’", result.logFile);
+
+        if (ftruncate(logFD.get(), 0) == -1)
+            throw SysError("truncating log file ‘%s’", result.logFile);
+
+        logFD = -1;
+
+        /* Do the build. */
+        printMsg(lvlDebug, "building ‘%s’ on ‘%s’",
+            localStore->printStorePath(step->drvPath),
+            machine->storeUri.render());
+
+        updateStep(ssBuilding);
+
+        BuildResult buildResult = build_remote::performBuild(
+            conn,
+            *localStore,
+            step->drvPath,
+            resolvedDrv,
+            buildOptions,
+            nrStepsBuilding
+        );
 
         result.updateWithBuildResult(buildResult);
 
@@ -180,13 +518,6 @@ void State::buildRemote(ref<Store> destStore,
 
         result.errorMsg = "";
 
-        /* If the NAR size limit was exceeded, then hydra-build-step
-           will not have copied the output paths. */
-        if (totalNarSize > maxOutputSize) {
-            result.stepStatus = bsNarSizeLimitExceeded;
-            return;
-        }
-
         /* If the path was substituted or already valid, then we didn't
            get a build log. */
         if (result.isCached) {
@@ -196,6 +527,57 @@ void State::buildRemote(ref<Store> destStore,
             result.logFile = "";
         }
 
+        StorePathSet outputs;
+        for (auto & [_, realisation] : buildResult.builtOutputs)
+            outputs.insert(realisation.outPath);
+
+        /* Copy the output paths. */
+        if (!machine->isLocalhost() || localStore != std::shared_ptr<Store>(destStore)) {
+            updateStep(ssReceivingOutputs);
+
+            MaintainCount<counter> mc(nrStepsCopyingFrom);
+
+            auto now1 = std::chrono::steady_clock::now();
+
+            auto infos = conn.queryPathInfos(*localStore, outputs);
+
+            size_t totalNarSize = 0;
+            for (auto & [_, info] : infos) totalNarSize += info.narSize;
+
+            if (totalNarSize > maxOutputSize) {
+                result.stepStatus = bsNarSizeLimitExceeded;
+                return;
+            }
+
+            /* Copy each path. */
+            printMsg(lvlDebug, "copying outputs of ‘%s’ from ‘%s’ (%d bytes)",
+                localStore->printStorePath(step->drvPath), machine->storeUri.render(), totalNarSize);
+
+            build_remote::copyPathsFromRemote(conn, narMembers, *localStore, *destStore, infos);
+            auto now2 = std::chrono::steady_clock::now();
+
+            result.overhead += std::chrono::duration_cast<std::chrono::milliseconds>(now2 - now1).count();
+        }
+
+        /* Register the outputs of the newly built drv */
+        if (experimentalFeatureSettings.isEnabled(Xp::CaDerivations)) {
+            auto outputHashes = staticOutputHashes(*localStore, *step->drv);
+            for (auto & [outputName, realisation] : buildResult.builtOutputs) {
+                // Register the resolved drv output
+                destStore->registerDrvOutput(realisation);
+
+                // Also register the unresolved one
+                auto unresolvedRealisation = realisation;
+                unresolvedRealisation.signatures.clear();
+                unresolvedRealisation.id.drvHash = outputHashes.at(outputName);
+                destStore->registerDrvOutput(unresolvedRealisation);
+            }
+        }
+
+        /* Shut down the connection. */
+        child->in = -1;
+        child->sshPid.wait();
+
     } catch (Error & e) {
         /* Disable this machine until a certain period of time has
            passed. This period increases on every consecutive

src/hydra-queue-runner/builder.cc

@@ -3,7 +3,6 @@
 #include "state.hh"
 #include "hydra-build-result.hh"
 #include "finally.hh"
-#include "terminal.hh"
 #include "binary-cache-store.hh"
 
 using namespace nix;
@@ -220,7 +219,7 @@ State::StepResult State::doBuildStep(nix::ref<Store> destStore,
                 result.canRetry = false;
             } else {
                 result.stepStatus = bsAborted;
-                result.errorMsg = filterANSIEscapes(e.msg(), true);
+                result.errorMsg = e.msg();
                 result.canRetry = true;
             }
         }

src/hydra-queue-runner/hydra-queue-runner.cc

@@ -550,11 +550,12 @@ void State::dumpStatus(Connection & conn)
         {"nrQueuedBuilds", builds.lock()->size()},
         {"nrActiveSteps", activeSteps_.lock()->size()},
         {"nrStepsBuilding", nrStepsBuilding.load()},
-        #if 0
         {"nrStepsCopyingTo", nrStepsCopyingTo.load()},
         {"nrStepsCopyingFrom", nrStepsCopyingFrom.load()},
-        #endif
+        {"nrStepsWaiting", nrStepsWaiting.load()},
         {"nrUnsupportedSteps", nrUnsupportedSteps.load()},
+        {"bytesSent", bytesSent.load()},
+        {"bytesReceived", bytesReceived.load()},
         {"nrBuildsRead", nrBuildsRead.load()},
         {"buildReadTimeMs", buildReadTimeMs.load()},
         {"buildReadTimeAvgMs", nrBuildsRead == 0 ? 0.0 : (float) buildReadTimeMs / nrBuildsRead},

src/hydra-queue-runner/state.hh

@@ -21,6 +21,8 @@
 #include "sync.hh"
 #include "nar-extractor.hh"
 #include "serve-protocol.hh"
+#include "serve-protocol-impl.hh"
+#include "serve-protocol-connection.hh"
 #include "machines.hh"
 
 
@@ -288,6 +290,12 @@ struct Machine : nix::Machine
     }
 
     bool isLocalhost() const;
+
+    // A connection to a machine
+    struct Connection : nix::ServeProto::BasicClientConnection {
+        // Backpointer to the machine
+        ptr machine;
+    };
 };
 
 
@@ -352,10 +360,9 @@ private:
     counter nrStepsStarted{0};
     counter nrStepsDone{0};
     counter nrStepsBuilding{0};
-    #if 0
     counter nrStepsCopyingTo{0};
     counter nrStepsCopyingFrom{0};
-    #endif
+    counter nrStepsWaiting{0};
     counter nrUnsupportedSteps{0};
     counter nrRetries{0};
     counter maxNrRetries{0};
@@ -364,6 +371,8 @@ private:
     counter nrQueueWakeups{0};
     counter nrDispatcherWakeups{0};
     counter dispatchTimeMs{0};
+    counter bytesSent{0};
+    counter bytesReceived{0};
     counter nrActiveDbUpdates{0};
 
     /* Specific build to do for --build-one (testing only). */

src/meson.build

@@ -1,6 +1,5 @@
 # Native code
 subdir('libhydra')
-subdir('hydra-build-step')
 subdir('hydra-evaluator')
 subdir('hydra-queue-runner')
 

src/script/hydra-eval-jobset

@@ -773,6 +773,9 @@ sub checkJobsetWrapped {
     my $jobsetChanged = 0;
     my %buildMap;
 
+    my @jobs;
+    push @jobs, $_ while defined($_ = $jobsIter->());
+
     $db->txn_do(sub {
         my $prevEval = getPrevJobsetEval($db, $jobset, 1);
 
@@ -796,7 +799,7 @@ sub checkJobsetWrapped {
 
         my @jobsWithConstituents;
 
-        while (defined(my $job = $jobsIter->())) {
+        foreach my $job (@jobs) {
             if ($jobsetsJobset) {
                 die "The .jobsets jobset must only have a single job named 'jobsets'"
                     unless $job->{attr} eq "jobsets";

t/meson.build

@@ -27,8 +27,6 @@ testenv.prepend('PERL5LIB',
   separator: ':'
 )
 testenv.prepend('PATH',
-  fs.parent(find_program('nix').full_path()),
-  fs.parent(hydra_build_step.full_path()),
   fs.parent(hydra_evaluator.full_path()),
   fs.parent(hydra_queue_runner.full_path()),
   meson.project_source_root() / 'src/script',