chore: prepare for first release

.github/workflows/release.yml

@@ -0,0 +1,26 @@
+name: goreleaser
+
+on:
+  push:
+    tags:
+      - "*"
+
+jobs:
+  goreleaser:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+        with:
+          fetch-depth: 0
+      - name: Set up Go
+        uses: actions/setup-go@v2
+        with:
+          go-version: 1.17
+      - name: Run GoReleaser
+        uses: goreleaser/goreleaser-action@v2
+        with:
+          version: latest
+          args: release --rm-dist
+        env:
+          GITHUB_TOKEN: ${{ secrets.GH_GORELEASER_TOKEN }}

.goreleaser.yml

@@ -0,0 +1,21 @@
+env:
+  - GO111MODULE=on
+  - CGO_ENABLED=0
+
+builds:
+  - binary: nomad-vector-logger.bin
+    id: nomad-vector-logger
+    goos:
+      - linux
+    goarch:
+      - amd64
+    ldflags:
+      - -s -w -X "main.buildString={{ .Tag }} ({{ .ShortCommit }} {{ .Date }})"
+    dir: .
+
+archives:
+  - format: tar.gz
+    files:
+      - README.md
+      - LICENSE
+      - config.sample.toml

Makefile

@@ -1,4 +1,4 @@
-APP-BIN := ./bin/nomad-events-sink.bin
+APP-BIN := ./bin/nomad-vector-logger.bin
 
 LAST_COMMIT := $(shell git rev-parse --short HEAD)
 LAST_COMMIT_DATE := $(shell git show -s --format=%ci ${LAST_COMMIT})

README.md

@@ -1,2 +1,110 @@
+<a href="https://zerodha.tech"><img src="https://zerodha.tech/static/images/github-badge.svg" align="right" /></a>
+
 # nomad-vector-logger
-A daemon which continuously watches jobs running in a Nomad cluster and templates out a Vector configuration file which can be used to collect application logs enriched with Nomad metadata.
+
+A daemon which continuously watches jobs running in a [Nomad](https://www.nomadproject.io/) cluster and templates out a [Vector](https://vector.dev/) configuration file which can be used to collect application logs enriched with Nomad metadata.
+
+## Why
+
+### The Problem
+
+Currently, Nomad stores all application logs inside `$NOMAD_DATA_DIR/$NOMAD_ALLOC_DIR/logs/` directory. The limitation here is that these logs don't come with any metadata about the task/job/allocation. If you're running multiple applications on the same host, a log collecting agent cannot distinguish these logs. For `docker` driver, this is a solved problem by configuring the `log_driver` for the container.
+
+However for `raw_exec` and `exec` this still seems to be an issue until [10219](https://github.com/hashicorp/nomad/issues/10219) is addressed.
+
+### The Solution
+
+- Nomad provides an [Events stream](https://github.com/mr-karan/nomad-events-sink) which continuosly streams events from the Nomad cluster.`nomad-vector-logger` is a Go daemon which subscribes to the Events stream on topic `Allocation` and listens for allocation updates.
+- It then **templates** out a `vector` configuration to collect app logs from the allocation's log directory and enriches the configuration with metadata such as `NodeID`, `Namesapce`, `Task Name`, `JobName`, `AllocID` etc.
+- `vector` is started with a [`--watch-config`](https://vector.dev/docs/administration/management/#reloading) flag which automatically live-reloads `vector` whenever config changes. A config change can happen whenever an allocation is _created/stopped/restarted_.
+
+You can see a sample [config file](./examples/vector/nomad.toml) that is generated by this program. This can be used in addition with other vector config files to provide the config for rest of pipeline (additional transformations, sinks etc).
+
+#### Before
+
+Logs without any metdata on `/opt/nomad/data/alloc/$ALLOC_ID/alloc/logs`:
+
+```
+==> proxy.stdout.0 <==
+192.168.29.76 - - [17/Jun/2022:11:01:42 +0000] "GET / HTTP/1.1" 200 27 "-" "curl/7.81.0" "-"
+```
+
+#### After
+
+This is an example JSON log collected from `nginx` task running with `raw_exec` task driver on Nomad, collected using `vector`:
+
+```json
+{
+  "file": "/opt/nomad/data/alloc/2cbf80ff-ef2a-e634-33ec-e097a6061001/alloc/logs/proxy.stdout.0",
+  "host": "pop-os",
+  "message": "192.168.29.76 - - [17/Jun/2022:11:00:03 +0000] \"GET / HTTP/1.1\" 200 27 \"-\" \"curl/7.81.0\" \"-\"",
+  "nomad_alloc_id": "2cbf80ff-ef2a-e634-33ec-e097a6061001",
+  "nomad_group_name": "nginx",
+  "nomad_job_name": "nginx",
+  "nomad_namespace": "default",
+  "nomad_node_name": "pop-os",
+  "nomad_task_name": "proxy",
+  "source_type": "file",
+  "timestamp": "2022-06-17T11:00:03.922205021Z"
+}
+```
+
+## Deploy
+
+- This is meant to run inside a Nomad cluster and should have proper ACL to listen to `Allocation:*` events.
+- This is meant to be run as a `system` job. Each allocation of this program is responsible to configure `vector` for the allocations running on the host.
+- Vector should be deployed externally, as a system job as well and should have access to the same directory that this program uses to generate the files.
+
+You can choose one of the various deployment options:
+
+### Binary
+
+Grab the latest release from [Releases](https://github.com/mr-karan/nomad-vector-logger/releases).
+
+To run:
+
+```
+$ ./nomad-vector-logger.bin --config config.toml
+```
+
+### Nomad
+
+TODO
+
+### Docker 
+
+TODO
+
+## Configuration
+
+Refer to [config.sample.toml](./config.sample.toml) for a list of configurable values.
+
+### Environment Variables
+
+All config variables can also be populated as env vairables by prefixing `NOMAD_VECTOR_LOGGER_` and replacing `.` with `__`.
+
+For eg: `app.data_dir` becomes `NOMAD_VECTOR_LOGGER_app__data_dir`.
+
+Nomad API client reads the following environment variables:
+
+- `NOMAD_TOKEN`
+- `NOMAD_ADDR`
+- `NOMAD_REGION`
+- `NOMAD_NAMESPACE`
+- `NOMAD_HTTP_AUTH`
+- `NOMAD_CACERT`
+- `NOMAD_CAPATH`
+- `NOMAD_CLIENT_CERT`
+- `NOMAD_CLIENT_KEY`
+- `NOMAD_TLS_SERVER_NAME`
+- `NOMAD_SKIP_VERIFY`
+
+You can read about them in detail [here](https://www.nomadproject.io/docs/runtime/environment).
+
+## Contribution
+
+Please feel free to open a new issue for bugs, feedback etc.
+
+## LICENSE
+
+[LICENSE](./LICENSE)

config.sample.toml

@@ -2,10 +2,10 @@
 log = "debug" # `debug` for verbose logs. `info` otherwise.
 data_dir = "./data/events" # Directory to store `index.json` file which is used to set index offset when listening for events.
 commit_index_interval = "5s" # Interval at which `index.json` gets stored in `data_dir`.
-nomad_data_dir = "/opt/nomad/data/alloc"
-remove_alloc_delay = "5s"
+nomad_data_dir = "/opt/nomad/data/alloc" # Nomad data directory where allocs are stored.
+remove_alloc_delay = "5s" # If the alloc is completed or stopped, the allocation isn't removed immediately from vector's config. You can delay the removal of alloc by `n` duration to ensure that vector has finished collecting all logs till then.
 vector_config_dir = "examples/vector/" # Path to the generated vector config file.
-extra_templates_dir = "static/"
+extra_templates_dir = "static/" # Extra templates that can be given. They will be rendered in `$vector_config_dir`. You can use variables mentioned in vector.tmpl if required.
 
 [stream]
 max_reconnect_attempts = 5 # Maximum reconnection attempts with Nomad Events API. After this limit is breached, program exits.

examples/vector/nomad.toml

@@ -0,0 +1,19 @@
+
+[sources.source_nomad_alloc_2cbf80ff-ef2a-e634-33ec-e097a6061001_proxy]
+type = "file"
+include = [ "/opt/nomad/data/alloc/2cbf80ff-ef2a-e634-33ec-e097a6061001/alloc/logs/proxy*" ]
+line_delimiter = "\n"
+read_from = "beginning"
+
+[transforms.transform_nomad_alloc_2cbf80ff-ef2a-e634-33ec-e097a6061001_proxy]
+type = "remap"
+inputs = ["source_nomad_alloc_2cbf80ff-ef2a-e634-33ec-e097a6061001_proxy"]
+source = '''
+# Store Nomad metadata.
+.nomad_namespace = "default"
+.nomad_node_name = "pop-os"
+.nomad_job_name = "nginx"
+.nomad_group_name = "nginx"
+.nomad_task_name = "proxy"
+.nomad_alloc_id = "2cbf80ff-ef2a-e634-33ec-e097a6061001"
+'''

examples/vector/sink.toml

@@ -0,0 +1,7 @@
+[sinks.stdout_nomad]
+type = "console"
+inputs = ["transform_nomad_alloc*"]
+target = "stdout"
+
+[sinks.stdout_nomad.encoding]
+codec = "json"

go.mod

@@ -1,11 +1,11 @@
-module github.com/mr-karan/nomad-alloc-logger
+module github.com/mr-karan/nomad-vector-logger
 
 go 1.18
 
 require (
 	github.com/hashicorp/nomad/api v0.0.0-20211103234928-04cab9dbecd3
 	github.com/knadh/koanf v1.4.1
-	github.com/mr-karan/nomad-events-sink v0.1.1-0.20220617051258-624503512560
+	github.com/mr-karan/nomad-events-sink v0.2.0
 	github.com/spf13/pflag v1.0.5
 	go.uber.org/zap v1.21.0
 )

go.sum

@@ -97,8 +97,8 @@ github.com/mitchellh/mapstructure v1.4.1/go.mod h1:bFUtVrKA4DC2yAKiSyO/QUcy7e+RR
 github.com/mitchellh/reflectwalk v1.0.0/go.mod h1:mSTlrgnPZtwu0c4WaC2kGObEpuNDbx0jmZXqmk4esnw=
 github.com/mitchellh/reflectwalk v1.0.2 h1:G2LzWKi524PWgd3mLHV8Y5k7s6XUvT0Gef6zxSIeXaQ=
 github.com/mitchellh/reflectwalk v1.0.2/go.mod h1:mSTlrgnPZtwu0c4WaC2kGObEpuNDbx0jmZXqmk4esnw=
-github.com/mr-karan/nomad-events-sink v0.1.1-0.20220617051258-624503512560 h1:u4oy5nydOTf0ppR5O2PErcEtEku2zbZT6IxgZVFv/jE=
-github.com/mr-karan/nomad-events-sink v0.1.1-0.20220617051258-624503512560/go.mod h1:JgAAR7BTiHHlpQ6ktutjkZhdCvPk7585Tt4YyxVknDU=
+github.com/mr-karan/nomad-events-sink v0.2.0 h1:0nyJDfSMCmA600BNerKqO0bYrumLDlxlb243rrBJRjU=
+github.com/mr-karan/nomad-events-sink v0.2.0/go.mod h1:JgAAR7BTiHHlpQ6ktutjkZhdCvPk7585Tt4YyxVknDU=
 github.com/npillmayer/nestext v0.1.3/go.mod h1:h2lrijH8jpicr25dFY+oAJLyzlya6jhnuG+zWp9L0Uk=
 github.com/oklog/run v1.0.0/go.mod h1:dlhp/R75TPv97u0XWUtDeV/lRKWPKSdTuV0TZvrmrQA=
 github.com/pascaldekloe/goe v0.1.0/go.mod h1:lzWF7FIEvWOWxwDKqyGYQf6ZUaNfKdP144TG7ZOy1lc=