|Martin Tournoij b35c86cfbd Fix up go.mod >_<||2 days ago|
|.github||3 weeks ago|
|acme||2 months ago|
|bgrun||1 week ago|
|cfg||2 months ago|
|cmd||3 weeks ago|
|cron||1 week ago|
|db||1 week ago|
|deploy||3 weeks ago|
|docs||1 month ago|
|gctest||1 month ago|
|handlers||2 days ago|
|pack||1 week ago|
|pgstat||2 months ago|
|public||1 week ago|
|tpl||3 weeks ago|
|widgets||1 month ago|
|.editorconfig||5 months ago|
|.gitattributes||2 months ago|
|.gitignore||5 months ago|
|.gogo-release||5 months ago|
|.ignore||10 months ago|
|.travis.yml||1 week ago|
|CHANGELOG.markdown||3 weeks ago|
|LICENSE||2 months ago|
|README.markdown||3 weeks ago|
|admin.go||2 months ago|
|api_token.go||1 month ago|
|db.go||2 months ago|
|export.go||1 month ago|
|export_test.go||1 month ago|
|gen.go||3 weeks ago|
|go.mod||2 days ago|
|go.sum||2 days ago|
|helper.go||1 month ago|
|hit.go||2 weeks ago|
|hit_list.go||1 month ago|
|hit_stats.go||1 month ago|
|hit_test.go||1 month ago|
|kommentaar.conf||2 months ago|
|memstore.go||2 months ago|
|memstore_test.go||2 months ago|
|netlify.toml||8 months ago|
|old.go||2 months ago|
|ref.go||1 month ago|
|refspam.go||1 week ago|
|site.go||1 week ago|
|site_test.go||2 months ago|
|tplfunc.go||1 month ago|
|update.go||2 months ago|
|user.go||1 month ago|
GoatCounter is an open source web analytics platform available as a hosted service (free for non-commercial use) or self-hosted app. It aims to offer easy to use and meaningful privacy-friendly web analytics as an alternative to Google Analytics or Matomo.
There are two ways to run this: as hosted service on goatcounter.com, free for non-commercial use, or run it on your own server (the source code is completely Open Source/Free Software, and it can be self-hosted without restrictions).
See docs/rationale.markdown for some more details on the “why?” of this project.
There’s a live demo at https://stats.arp242.net.
Please consider contributing financially if you’re self-hosting GoatCounter so I can pay my rent :-)
GoatCounter is sponsored by a grant from NLnet’s NGI Zero PET fund.
Easy; if you’ve been confused by the myriad of options and flexibility of Google Analytics and Matomo that you don’t need then GoatCounter will be a breath of fresh air.
Identify unique visits without cookies using a non-identifiable hash (technical details).
Keeps useful statistics such as browser information, location, and screen size. Keep track of referring sites and campaigns.
100% committed to open source; you can see exactly what the code does and make improvements.
Own your data; you can always export all data and cancel at any time.
Integrate on your site with just a single script tag:
Fast: can handle about 800 hits/second on a $5/month Linode VPS using the default settings.
Self-contained binary: everything – including static assets – is in a single ~7M statically compiled binary. The only other thing you need is a SQLite database file or PostgreSQL connection (no way around that).
The release page has binaries for Linux amd64, arm, and arm64. These are statically compiled and should work in pretty much any Linux environment. GoatCounter should run on any platform supported by Go, but there are no binaries for them (yet); you’ll have to build from source for now (it’s not hard, I promise).
Note this README is for the latest master; use the
for the 1.4 README.
Generally speaking only the latest release is supported, although critical fixes (security, data loss, etc.) may get backported to previous releases.
A $5/month Linode is more than enough to run GoatCounter unless you’ve got millions of pageviews. And if you don’t have a Linode account yet then consider using my referral URL and I’ll get some cash back from Linode :-)
Some people have created Dockerfiles. You don’t really need Docker since GoatCounter is a static binary with no external dependencies; it probably creates more problems than it solves IMHO. At any rate, here are some that seem alright at a glance if you must:
Compile from source with:
$ git clone -b release-1.4 https://github.com/zgoat/goatcounter.git $ cd goatcounter $ go build -ldflags="-X main.version=$(git log -n1 --format='%h_%cI')" ./cmd/goatcounter
-ldflags=[..] sets the version; this isn’t strictly required as such,
but it’s recommended as it’s used to “bust” the cache for static files and may
also be useful later when reporting bugs. This can be any string and doesn’t
follow any particular format, you can also set this to the current date or
banana or anything you want really.
Or to build a statically linked binary:
$ go build -ldflags="-X main.version=$(git log -n1 --format='%h_%cI')" \ -tags osusergo,netgo,sqlite_omit_load_extension \ -ldflags='-extldflags=-static' \ ./cmd/goatcounter
You’ll now have a
goatcounter binary in the current directory.
You need Go 1.13 or newer and a C compiler (for SQLite), or compile it with
CGO_ENABLED=0 go build and use PostgreSQL.
It’s recommended to use the latest release as in the above command. The master branch should be reasonably stable but no guarantees, and sometimes I don’t write detailed release/upgrade notes until the actual release.
It’s not recommended to use
go get in GOPATH mode since that will ignore the
dependency versions in go.mod.
You can start a server with:
$ goatcounter serve
The default is to use a SQLite database at
will be created if it doesn’t exist yet. See the
-db flag and
goatcounter help db to customize this.
GoatCounter will listen on port
*:443 by default. You don’t need
to run it as root and can grant the appropriate permissions on Linux with:
$ setcap 'cap_net_bind_service=+ep' goatcounter
Listening on a different port can be a bit tricky due to the ACME/Let’s Encrypt
goatcounter help listen documents this in depth.
You can create new sites with the
$ goatcounter create -email firstname.lastname@example.org -domain stats.example.com
This will ask for a password for your new account; you can also add a password
on the commandline with
-password. If you use a custom DB, you must also pass
-db flag here.
You may need to run the database migrations when updating. Use
-automigrate to always run all pending migrations on startup. This is the
easiest way, although arguably not the “best” way.
goatcounter migrate <file> or
goatcounter migrate all to manually run
migrations; generally you want to upload the new version, run migrations while
the old one is still running, and then restart so the new version takes effect.
goatcounter migrate show to get a list of pending migrations.
Both SQLite and PostgreSQL are supported. SQLite should work well for the vast majority of people and is the recommended database engine. PostgreSQL will not be faster in most cases, and the chief reason for adding support in the first place is to support load balancing web requests over multiple servers. To use it:
Create the database, unlike SQLite it’s not done automatically (you may need
to modify the
$ createdb goatcounter $ psql goatcounter -c ‘\i db/schema.pgsql’ $ goatcounter -db ‘postgresql://dbname=goatcounter’ migrate all
Run with custom
$ goatcounter serve
-db 'postgresql://user=goatcounter dbname=goatcounter sslmode=disable'
See the pq docs for more details on the connection string.
You can compile goatcounter without cgo if you don’t use SQLite:
$ CGOENABLED=0 go build -ldflags=“-X main.version=$(git log -n1 --format=‘%h%cI’)” ./cmd/goatcounter
Functionally it doesn’t matter too much, but builds will be a bit easier and faster as it won’t require a C compiler.
You can start a test/development server with:
$ goatcounter serve -dev
-dev flag makes some small things a bit more convenient for development;
TLS is disabled by default, it will listen on localhost:8081, the application
will automatically restart on recompiles, and a few other minor changes.
See .github/CONTRIBUTING.markdown for more details on how to run a development server, write patches, etc.
Various aggregate data files are available at https://www.goatcounter.com/data