Merge remote-tracking branch 'origin/main' into sled-storage

.config/hakari.toml

@@ -0,0 +1,33 @@
+# This file contains settings for `cargo hakari`.
+# See https://docs.rs/cargo-hakari/latest/cargo_hakari/config for a full list of options.
+
+hakari-package = "omicron-workspace-hack"
+
+# Format for `workspace-hack = ...` lines in other Cargo.tomls. Requires cargo-hakari 0.9.8 or above.
+dep-format-version = "4"
+
+# Output lines as `omicron-workspace-hack.workspace = true`. Requires
+# cargo-hakari 0.9.28 or above.
+workspace-hack-line-style = "workspace-dotted"
+
+# Setting workspace.resolver = "2" in the root Cargo.toml is HIGHLY recommended.
+# Hakari works much better with the new feature resolver.
+# For more about the new feature resolver, see:
+# https://blog.rust-lang.org/2021/03/25/Rust-1.51.0.html#cargos-new-feature-resolver
+resolver = "2"
+
+# Add triples corresponding to platforms commonly used by developers here.
+# https://doc.rust-lang.org/rustc/platform-support.html
+platforms = [
+    "x86_64-unknown-linux-gnu",
+    "x86_64-apple-darwin",
+    "aarch64-apple-darwin",
+    "x86_64-unknown-illumos",
+    # "x86_64-pc-windows-msvc",
+]
+
+# Write out exact versions rather than a semver range. (Defaults to false.)
+exact-versions = true
+
+[traversal-excludes]
+workspace-members = ["xtask"]

.config/nextest.toml

@@ -8,7 +8,9 @@ nextest-version = { required = "0.9.59", recommended = "0.9.59" }
 experimental = ["setup-scripts"]
 
 [[profile.default.scripts]]
-filter = 'rdeps(nexus-test-utils)'
+# Exclude omicron-dev tests from crdb-seed as we explicitly want to simulate an
+# environment where the seed file doesn't exist.
+filter = 'rdeps(nexus-test-utils) - package(omicron-dev)'
 setup = 'crdb-seed'
 
 [profile.ci]

.github/buildomat/build-and-test.sh

@@ -69,7 +69,7 @@ ptime -m timeout 1h cargo test --doc --locked --verbose --no-fail-fast
 # We expect the seed CRDB to be placed here, so we explicitly remove it so the
 # rmdir check below doesn't get triggered. Nextest doesn't have support for
 # teardown scripts so this is the best we've got.
-rm -rf "$TEST_TMPDIR/crdb-base"
+rm -rf "$TEST_TMPDIR/crdb-base"*
 
 #
 # Make sure that we have left nothing around in $TEST_TMPDIR.  The easiest way

.github/buildomat/config.toml

@@ -17,5 +17,6 @@ org_only = true
 allow_users = [
 	"dependabot[bot]",
 	"oxide-reflector-bot[bot]",
+	"oxide-renovate[bot]",
 	"renovate[bot]",
 ]

.github/buildomat/jobs/build-and-test-helios.sh

@@ -3,7 +3,7 @@
 #: name = "build-and-test (helios)"
 #: variety = "basic"
 #: target = "helios-2.0"
-#: rust_toolchain = "1.72.0"
+#: rust_toolchain = "1.72.1"
 #: output_rules = [
 #:	"/var/tmp/omicron_tmp/*",
 #:	"!/var/tmp/omicron_tmp/crdb-base*",

.github/buildomat/jobs/build-and-test-linux.sh

@@ -1,9 +1,9 @@
 #!/bin/bash
 #:
-#: name = "build-and-test (ubuntu-20.04)"
+#: name = "build-and-test (ubuntu-22.04)"
 #: variety = "basic"
-#: target = "ubuntu-20.04"
-#: rust_toolchain = "1.72.0"
+#: target = "ubuntu-22.04"
+#: rust_toolchain = "1.72.1"
 #: output_rules = [
 #:	"/var/tmp/omicron_tmp/*",
 #:	"!/var/tmp/omicron_tmp/crdb-base*",

.github/buildomat/jobs/ci-tools.sh

@@ -3,7 +3,7 @@
 #: name = "helios / CI tools"
 #: variety = "basic"
 #: target = "helios-2.0"
-#: rust_toolchain = "1.72.0"
+#: rust_toolchain = "1.72.1"
 #: output_rules = [
 #:	"=/work/end-to-end-tests/*.gz",
 #:	"=/work/caboose-util.gz",

.github/buildomat/jobs/clippy.sh

@@ -3,7 +3,7 @@
 #: name = "clippy (helios)"
 #: variety = "basic"
 #: target = "helios-2.0"
-#: rust_toolchain = "1.72.0"
+#: rust_toolchain = "1.72.1"
 #: output_rules = []
 
 # Run clippy on illumos (not just other systems) because a bunch of our code
@@ -29,3 +29,4 @@ ptime -m bash ./tools/install_builder_prerequisites.sh -y
 
 banner clippy
 ptime -m cargo xtask clippy
+ptime -m cargo doc

.github/buildomat/jobs/deploy.sh

@@ -2,7 +2,7 @@
 #:
 #: name = "helios / deploy"
 #: variety = "basic"
-#: target = "lab-2.0-opte-0.23"
+#: target = "lab-2.0-opte-0.25"
 #: output_rules = [
 #:  "%/var/svc/log/oxide-sled-agent:default.log*",
 #:  "%/pool/ext/*/crypt/zone/oxz_*/root/var/svc/log/oxide-*.log*",
@@ -143,6 +143,7 @@ cd /opt/oxide/work
 
 ptime -m tar xvzf /input/package/work/package.tar.gz
 cp /input/package/work/zones/* out/
+mv out/omicron-nexus-single-sled.tar.gz out/omicron-nexus.tar.gz
 mkdir tests
 for p in /input/ci-tools/work/end-to-end-tests/*.gz; do
 	ptime -m gunzip < "$p" > "tests/$(basename "${p%.gz}")"
@@ -231,11 +232,11 @@ infra_ip_first = \"$UPLINK_IP\"
 		/^infra_ip_last/c\\
 infra_ip_last = \"$UPLINK_IP\"
 	}
-	/^\\[\\[rack_network_config.uplinks/,/^\$/ {
-		/^gateway_ip/c\\
-gateway_ip = \"$GATEWAY_IP\"
-		/^uplink_cidr/c\\
-uplink_cidr = \"$UPLINK_IP/32\"
+	/^\\[\\[rack_network_config.ports/,/^\$/ {
+		/^routes/c\\
+routes = \\[{nexthop = \"$GATEWAY_IP\", destination = \"0.0.0.0/0\"}\\]
+		/^addresses/c\\
+addresses = \\[\"$UPLINK_IP/32\"\\]
 	}
 " pkg/config-rss.toml
 diff -u pkg/config-rss.toml{~,} || true

.github/buildomat/jobs/host-image.sh

@@ -1,11 +1,12 @@
 #!/bin/bash
 #:
-#: name = "helios / build OS image"
+#: name = "helios / build OS images"
 #: variety = "basic"
 #: target = "helios-2.0"
-#: rust_toolchain = "1.72.0"
+#: rust_toolchain = "1.72.1"
 #: output_rules = [
-#:	"=/work/helios/image/output/os.tar.gz",
+#:	"=/work/helios/upload/os-host.tar.gz",
+#:	"=/work/helios/upload/os-trampoline.tar.gz",
 #: ]
 #: access_repos = [
 #:	"oxidecomputer/amd-apcb",
@@ -44,14 +45,49 @@ TOP=$PWD
 
 source "$TOP/tools/include/force-git-over-https.sh"
 
-# Checkout helios at a pinned commit into /work/helios
-git clone https://github.com/oxidecomputer/helios.git /work/helios
-cd /work/helios
+# Check out helios into /work/helios
+HELIOSDIR=/work/helios
+git clone https://github.com/oxidecomputer/helios.git "$HELIOSDIR"
+cd "$HELIOSDIR"
+# Record the branch and commit in the output
+git status --branch --porcelain=2
+# Setting BUILD_OS to no makes setup skip repositories we don't need for
+# building the OS itself (we are just building an image from already built OS).
+BUILD_OS=no gmake setup
+
+# Commands that "helios-build" would ask us to run (either explicitly or
+# implicitly, to avoid an error).
+rc=0
+pfexec pkg install -q /system/zones/brand/omicron1/tools || rc=$?
+case $rc in
+    # `man pkg` notes that exit code 4 means no changes were made because
+    # there is nothing to do; that's fine. Any other exit code is an error.
+    0 | 4) ;;
+    *) exit $rc ;;
+esac
+
+pfexec zfs create -p "rpool/images/$USER"
+
 
 # TODO: Consider importing zones here too?
 
 cd "$TOP"
+OUTPUTDIR="$HELIOSDIR/upload"
+mkdir "$OUTPUTDIR"
+
+banner OS
 ./tools/build-host-image.sh -B \
     -S /input/package/work/zones/switch-asic.tar.gz \
-    /work/helios \
+    "$HELIOSDIR" \
     /input/package/work/global-zone-packages.tar.gz
+
+mv "$HELIOSDIR/image/output/os.tar.gz" "$OUTPUTDIR/os-host.tar.gz"
+
+banner Trampoline
+
+./tools/build-host-image.sh -R \
+    "$HELIOSDIR" \
+    /input/package/work/trampoline-global-zone-packages.tar.gz
+
+mv "$HELIOSDIR/image/output/os.tar.gz" "$OUTPUTDIR/os-trampoline.tar.gz"
+

.github/buildomat/jobs/package.sh

@@ -3,7 +3,7 @@
 #: name = "helios / package"
 #: variety = "basic"
 #: target = "helios-2.0"
-#: rust_toolchain = "1.72.0"
+#: rust_toolchain = "1.72.1"
 #: output_rules = [
 #:	"=/work/version.txt",
 #:	"=/work/package.tar.gz",
@@ -37,15 +37,15 @@ rustc --version
 # trampoline global zone images.
 #
 COMMIT=$(git rev-parse HEAD)
-VERSION="1.0.2-0.ci+git${COMMIT:0:11}"
+VERSION="1.0.3-0.ci+git${COMMIT:0:11}"
 echo "$VERSION" >/work/version.txt
 
 ptime -m ./tools/install_builder_prerequisites.sh -yp
 ptime -m ./tools/ci_download_softnpu_machinery
 
 # Build the test target
 ptime -m cargo run --locked --release --bin omicron-package -- \
-  -t test target create -i standard -m non-gimlet -s softnpu
+  -t test target create -i standard -m non-gimlet -s softnpu -r single-sled
 ptime -m cargo run --locked --release --bin omicron-package -- \
   -t test package
 
@@ -71,7 +71,7 @@ tarball_src_dir="$(pwd)/out/versioned"
 stamp_packages() {
 	for package in "$@"; do
 		# TODO: remove once https://github.com/oxidecomputer/omicron-package/pull/54 lands
-		if [[ $package == maghemite ]]; then
+		if [[ $package == mg-ddm-gz ]]; then
 			echo "0.0.0" > VERSION
 			tar rvf "out/$package.tar" VERSION
 			rm VERSION
@@ -81,12 +81,16 @@ stamp_packages() {
 	done
 }
 
+# Keep the single-sled Nexus zone around for the deploy job. (The global zone
+# build below overwrites the file.)
+mv out/omicron-nexus.tar.gz out/omicron-nexus-single-sled.tar.gz
+
 # Build necessary for the global zone
 ptime -m cargo run --locked --release --bin omicron-package -- \
-  -t host target create -i standard -m gimlet -s asic
+  -t host target create -i standard -m gimlet -s asic -r multi-sled
 ptime -m cargo run --locked --release --bin omicron-package -- \
   -t host package
-stamp_packages omicron-sled-agent maghemite propolis-server overlay
+stamp_packages omicron-sled-agent mg-ddm-gz propolis-server overlay
 
 # Create global zone package @ /work/global-zone-packages.tar.gz
 ptime -m ./tools/build-global-zone-packages.sh "$tarball_src_dir" /work
@@ -111,6 +115,7 @@ zones=(
   out/external-dns.tar.gz
   out/internal-dns.tar.gz
   out/omicron-nexus.tar.gz
+  out/omicron-nexus-single-sled.tar.gz
   out/oximeter-collector.tar.gz
   out/propolis-server.tar.gz
   out/switch-*.tar.gz
@@ -130,7 +135,7 @@ ptime -m cargo run --locked --release --bin omicron-package -- \
   -t recovery target create -i trampoline
 ptime -m cargo run --locked --release --bin omicron-package -- \
   -t recovery package
-stamp_packages installinator maghemite
+stamp_packages installinator mg-ddm-gz
 
 # Create trampoline global zone package @ /work/trampoline-global-zone-packages.tar.gz
 ptime -m ./tools/build-trampoline-global-zone-packages.sh "$tarball_src_dir" /work