v2.1.2

[ruslandoga]

Warning

This release had problems with login, v2.1.3 has been published with a fix: #4629

The highlights of this release are:

• Automatic IPv6 support for all outgoing connections
• Opt-in HTTPS support and Let's Encrypt integration
• Simplified Docker Compose configuration (no reverse proxy required)
• Various product improvements: https://plausible.io/changelog — up until today's date :)

Upgrade

Update the image used for plausible

_{docker-compose.yml}

plausible:
- image: ghcr.io/plausible/community-edition:v2.1.1
+ image: ghcr.io/plausible/community-edition:v2.1.2

and restart the containers

_console

$ cd hosting
$ docker compose up -d

This will boot up the new version of the app.

Changelog

Following changes have been made since v2.1.1:

Added

• UI to edit goals along with display names
• Support contains filter for goals
• UI to edit funnels
• Add Details views for browsers, browser versions, os-s, os versions, and screen sizes reports
• Add a search functionality in all Details views
• Icons for browsers Add browser icons #4239
• Automatic custom property selection in the dashboard Properties report
• Add contains_not filter support to dashboard
• Traffic drop notifications Implement traffic drop notifications #4300
• Add search and pagination functionality into Google Keywords > Details modal
• ClickHouse system.query_log table log_comment column now contains information about source of queries. Useful for debugging
• New /debug/clickhouse route for super admins which shows information on clickhouse queries executed by user
• Typescript support for /assets
• Testing framework for /assets
• Automatic HTTPS Auto HTTPS in CE #4491
• Make details views on dashboard sortable

Removed

• Deprecate ECTO_IPV6 and ECTO_CH_IPV6 env vars in CE happy tcp #4245
• Remove support for importing data from no longer available Universal Analytics
• Soft-deprecate DATABASE_SOCKET_DIR Improve PG unix sockets configuration #4202

Changed

• Support Unix sockets in DATABASE_URL Improve PG unix sockets configuration #4202
• Realtime and hourly graphs now show visits lasting their whole duration instead when specific events occur
• Increase hourly request limit for API keys in CE from 600 to 1000000 (practically removing the limit) increase api_keys.hourly_request_limit in CE #4200
• Make TCP connections try IPv6 first with IPv4 fallback in CE happy tcp #4245
• is and is not filters in dashboard no longer support wildcards. Use contains/does not contain filter instead.
• bounce_rate metric now returns 0 instead of null for event:page breakdown when page has never been entry page.
• Make TOTP_VAULT_KEY optional Don't require TOTP vault key in CE #4317
• Sources like 'google' and 'facebook' are now stored in capitalized forms ('Google', 'Facebook') Map lowercase tagged sources to capitalized form during ingestion #4417
• DATABASE_CACERTFILE now forces TLS for PostgreSQL connections, so you don't need to add ?ssl=true in DATABASE_URL
• Change auth session cookies to token-based ones with server-side expiration management.
• Improve Google error messages in CE Better Google error messages in CE #4485
• Better compress static assets in CE Add better assets compression to CE #4476
• Return domain-less cookies in CE Return domain-less cookies in CE #4482
• Internal stats API routes now return a JSON error over HTML in case of invalid access.

Fixed

• Fix access to Stats API feature in CE fix ce stats api access #4244
• Fix filter suggestions when same filter previously applied
• Fix MX lookup when using relays with Bamboo.Mua Update bamboo_mua to disable mx lookups when relay is used #4350
• Don't include imports when showing time series hourly interval. Previously imported data was shown each midnight
• Fix property filter suggestions 500 error when property hasn't been selected
• Bamboo.Mua: add Date and Message-ID headers if missing Update bamboo_mua to fix headers issue #4474
• Fix migration order across plausible_db and plausible_events_db databases Run migrations in order across repos #4466

---

This discussion was created from the release v2.1.2.

[Olakeen]

Hello, thank you for this update.

My admin account is no longer working. How do I define the account now ? I don't see Admin_User_Pwd in the wiki ? https://github.com/plausible/community-edition/wiki

[ruslandoga]

Do you have a reverse proxy in front of Plausible? If so, you would need to make sure you have WebSockets-proxying enabled there: https://github.com/plausible/community-edition/wiki/Reverse-Proxy

You can also search for past discussions about that issue:

• Issue with getting Plausible CE working. Register page loads, but everything comes back to registration. I cannot register/login #4274
• Websocket connection failed on community edition #3826
• Registration UI not working #4291
• MySites (/sites) overview with no data after 2.0 to 2.1 Upgrade #4310

[Olakeen]

WebSockets-proxying is correctly configured.

Tested on my phone and login was functionnal so I tried in private mode on my computer and it also worked.
I have deleted cookies and local storage and it now works properly. First time I've had this problem.

[lordcris]

I have deleted cookies and local storage and it now works properly. First time I've had this problem.

Yes, I was using traefik and had the same problem. Deleting the cookies resolved it,

[Tony-MF]

I also had to delete cookies in order to get past the login screen.

[ruslandoga]

#4621 would probably fix the issue and get released as v2.1.3. Sorry for the inconvenience :)

@Olakeen thank you for providing a workaround!

[rootforbid]

Hi,

After upgrading from 2.1.1 as per the instruction, Plausible doesn't boot anymore, getting a bootloop of this error message:

ERROR! Config provider Config.Reader failed with:
** (ArgumentError) errors were found at the given arguments:

  * 1st argument: not an already existing atom

    :erlang.binary_to_existing_atom("1", :utf8)
    /app/releases/0.0.1/runtime.exs:648: (file)
    (elixir 1.17.1) src/elixir.erl:386: :elixir.eval_external_handler/3
    (stdlib 6.0) erl_eval.erl:904: :erl_eval.do_apply/7
    (stdlib 6.0) erl_eval.erl:648: :erl_eval.expr/6
    (stdlib 6.0) erl_eval.erl:271: :erl_eval.exprs/6
    (elixir 1.17.1) src/elixir.erl:364: :elixir.eval_forms/4
    (elixir 1.17.1) lib/module/parallel_checker.ex:112: Module.ParallelChecker.verify/1

Runtime terminating during boot ({badarg,[{erlang,binary_to_existing_atom,[<<"1">>,utf8],[{error_info,#{module=>erl_erts_errors}}]},{elixir_eval,'__FILE__',1,[{file,"/app/releases/0.0.1/runtime.exs"},{line,648}]},{elixir,eval_external_handler,3,[{file,"src/elixir.erl"},{line,386}]},{erl_eval,do_apply,7,[{file,"erl_eval.erl"},{line,904}]},{erl_eval,expr,6,[{file,"erl_eval.erl"},{line,648}]},{erl_eval,exprs,6,[{file,"erl_eval.erl"},{line,271}]},{elixir,eval_forms,4,[{file,"src/elixir.erl"},{line,364}]},{'Elixir.Module.ParallelChecker',verify,1,[{file,"lib/module/parallel_checker.ex"},{line,112}]}]})

Crash dump is being written to: erl_crash.dump...

docker-compose.yml

services:
  plausible_db:
    image: postgres:15-alpine
    container_name: plausible_db
    env_file: ./.env
    user: "${PUID}:${PGID}"
    volumes:
      - ./db:/var/lib/postgresql/data
    restart: always

  plausible_events_db:
    image: clickhouse/clickhouse-server:24.3.3.102-alpine
    container_name: plausible_db-events
    env_file: ./.env
    user: "${PUID}:${PGID}"
    volumes:
      - ./db-events/data:/var/lib/clickhouse
      - ./db-events/logs:/var/log/clickhouse-server
      - ./db-events/clickhouse-config.xml:/etc/clickhouse-server/config.d/logging.xml:ro
      - ./db-events/clickhouse-user-config.xml:/etc/clickhouse-server/users.d/logging.xml:ro
    ulimits:
      nofile:
        soft: 262144
        hard: 262144
    restart: always

  plausible:
    image: ghcr.io/plausible/community-edition:v2.1.2
    container_name: plausible
    env_file: ./.env
    depends_on:
      - plausible_db
      - plausible_events_db
    command: sh -c "sleep 10 && /entrypoint.sh db createdb && /entrypoint.sh db migrate && /entrypoint.sh run"
    restart: always

networks:
    default:
      name: analytics_nw
      external: true

.env

BASE_URL=https://plausible.mysite.com
SECRET_KEY_BASE=******
TOTP_VAULT_KEY=******
DISABLE_REGISTRATION=invite_only
DATABASE_URL=postgres://root:******@plausible_db:5432/plausible
CLICKHOUSE_DATABASE_URL=http://plausible_db-events:8123/plausible_events_db
MAILER_ADAPTER=Bamboo.Mua
MAILER_EMAIL="no-reply@mysite.com"
SMTP_HOST_ADDR="******"
SMTP_HOST_PORT="587"
SMTP_USER_NAME="******"
SMTP_USER_PWD="******"
SMTP_HOST_SSL_ENABLED=1
MAXMIND_LICENSE_KEY=******
MAXMIND_EDITION=GeoLite2-City
GOOGLE_CLIENT_ID=******
GOOGLE_CLIENT_SECRET=******
PUID=1000
PGID=1000

Had to downgrade to 2.1.1 which didn't cause any issue.

[rootforbid]

Thanks. I've been using Plausible since v1.5 and have done the incremental updates at every new release.
It seems that the "SMTP_HOST_SSL_ENABLED" variable is not even used anymore anyway (according to the wiki) so its value had no impact =P
I've removed this variable from the .env file, thanks 👍

[ruslandoga]

I must've removed it by mistake :) Thank you for mentioning that it's missing!

I put it back now: https://github.com/plausible/community-edition/wiki/Configuration#smtp_host_ssl_enabled. It is still being used in some configurations, e.g. #4584 (comment)

[rootforbid]

Even with 2.1.3, I still got the same error bootloop.
Didn't really know the meaning of the error except it was related to "Config".
Went back into my .env file and enclosed every variable values in double-quotes and problem solved!
Not sure why it isn't an issue with 2.1.1 but it now works :)

[ruslandoga]

Nothing changed regarding this environment variable in v2.1.3, so it still needs to be true or false and not an integer.

Went back into my .env file and enclosed every variable values in double-quotes and problem solved!

Quoting shouldn't matter unless the environment variable contains special characters. But note that the container needs to be recreated for the new env_file to take effect in Docker Compose since Docker Compose does not track changes to files in the env_file directive.

Not sure why it isn't an issue with 2.1.1 but it now works :)

In v2.1.1 we didn't read SMTP_HOST_SSL_ENABLED when adapter was set to Bamboo.Mua.

[rootforbid]

This must be it then. I did completely removed the variable because I actually didn't need it, and its default value is false as per the wiki and was still getting the boot loop. But indeed I did not recreate the container.
I will still just use double quotes for every env values in the future. I can never remember which characters are special or not, so why take a chance :)

Thanks for the help, learned a good tip today: when in doubt, recreate container if env file is modified via env_file directive

[DrMaxNix]

After updating from v2.1.1 to v2.1.2, the login was broken. The login correctly redirected me to /sites, but the sites page redirected me back to /login. I tried "clear cookies and site data", which fixed the problem. Login worked flawlessly afterwards!
This problem and fix was also mentioned in #4619 (comment)
Seems to be related to #4621

[ruslandoga]

v2.1.3 with has been published: #4629

It shouldn't have the v2.1.2's login issue.