Add ONCE backup and restore hooks

By default, ONCE pauses the application container when taking a backup.
This guarantees a consistent snapshot of the data, but for a large
installation and/or a slow destination path (like a network mount), that
pause could be a bit disruptive to live traffic.

But ONCE also support a pause-free path: if a `pre-backup` hook is
present in the container, ONCE will run that and skip the pause. So by
supplying necessary hooks, we can avoid any disruption from backups.

- `pre-backup` runs the existing script/admin/prepare-backup, which uses
  SQLite's online backup API to get a consistent snapshop. k

- `post-restore` moves our snaphotted file back into place, and removes
  the orphaned sidecar files

Dockerfile

@@ -66,6 +66,9 @@ ENV HTTP_WRITE_TIMEOUT=300
 COPY --from=build --chown=rails:rails /usr/local/bundle /usr/local/bundle
 COPY --from=build --chown=rails:rails /rails /rails
 
+# Install ONCE backup/restore hooks
+COPY --chmod=755 hooks /hooks
+
 # Set version and revision
 ARG APP_VERSION
 ENV APP_VERSION=$APP_VERSION

README.md

@@ -53,3 +53,7 @@ If you'd rather run the Docker image yourself, you can read more about that in t
 
 You are welcome - and encouraged - to modify Campfire to your liking.
 Please see our [development guide](docs/development.md) for how to get Campfire set up for local development.
+
+## Security
+
+See [SECURITY.md](SECURITY.md) for how to report a vulnerability and a description of our trust model.

SECURITY.md

@@ -0,0 +1,22 @@
+# Security
+
+Report vulnerabilities through our [HackerOne program](https://hackerone.com/basecamp) or at
+https://github.com/basecamp/once-campfire/security. Note that Campfire is in scope for our security
+program, but is not bounty eligible.
+
+## Trust model
+
+Campfire is self-hosted and single-tenant, so the administrator is the server operator,
+with shell, network, and database access already. Anything requiring the administrator
+role grants nothing they do not already have, and is not a vulnerability.
+
+We do want reports of anything a **non-administrator** can reach, and of any credential or
+network path held by the Campfire process but not by the operator's own shell.
+
+## Intentional behavior
+
+Bot webhook URLs are unrestricted: an administrator can point one at any address,
+including internal ones, because operators legitimately wire bots to their own services.
+Link unfurling is different because any member can trigger it by pasting a URL, so it
+validates destinations through `RestrictedHTTP::PrivateNetworkGuard`. The difference is who
+picks the destination.

app/models/webhook.rb

@@ -24,6 +24,8 @@ class Webhook < ApplicationRecord
         Net::HTTP::Post.new(uri, "Content-Type" => "application/json").tap { |request| request.body = payload }
     end
 
+    # No PrivateNetworkGuard, unlike Opengraph::Fetch: only an administrator sets this URL,
+    # and operators legitimately point bots at their own internal services.
     def http
       Net::HTTP.new(uri.host, uri.port).tap do |http|
         http.use_ssl = (uri.scheme == "https")

hooks/post-restore

@@ -0,0 +1,12 @@
+#!/bin/bash
+set -euo pipefail
+
+db_name="${RAILS_ENV:-production}.sqlite3"
+snapshot="/rails/storage/backups/$db_name"
+db="/rails/storage/db/$db_name"
+
+if [ -f "$snapshot" ]; then
+  mkdir -p "$(dirname "$db")"
+  cp "$snapshot" "$db"
+  rm -f "$db-wal" "$db-shm"
+fi

hooks/pre-backup

@@ -0,0 +1,2 @@
+#!/bin/bash
+exec /rails/script/admin/prepare-backup