docker-mailserver/test/test_helper/common.bash
Brennan Kinney b58165762a
fix(changedetector): Use service reload commands instead of supervisorctl restart <service> (#2947)
With `reload` a change detection event during local testing can be processed in less than a second according to logs. Previously this was 5+ seconds (_plus additional downtime for Postfix/Dovecot to become available again_).

In the past it was apparently an issue to use `<service> reload` due to a concern with the PID for wrapper scripts that `supervisorctl` managed, thus `supervisorctl <service> restart` had been used. Past discussions with maintainers suggest this is not likely an issue anymore, and `reload` should be fine to switch to now 👍 

---

**NOTE:** It may not be an issue in the CI, but on _**local systems running tests may risk failure in `setup-cli.bats` from a false positive**_ due to 1 second polling window of the test helper method, and a change event being possible to occur entirely between the two checks undetected by the current approach.

If this is a problem, we may need to think of a better way to catch the change. The `letsencrypt` test counts how many change events are expected to have been processed, and this could technically be leveraged by the test helper too.

---

**NOTE:** These two lines (_with regex pattern for postfix_) are output in the terminal when using the services respective `reload` commands:

```
postfix/master.*: reload -- version .*, configuration /etc/postfix
dovecot: master: Warning: SIGHUP received - reloading configuration
```

I wasn't sure how to match them as they did not appear in the `changedetector` log (_**EDIT:** they appear in the main log output, eg `docker logs <container name>`_).

Instead I've just monitored the `changedetector` log messages, which should be ok for logic that previously needed to ensure Dovecot / Postfix was back up after the `restart` was issued.

---

Commit history:

* chore: Change events `reload` Dovecot and Postfix instead of `restart`

Reloading is faster than restarting the processes.

Restarting is a bit heavy handed here and may no longer be necessary for general usage?

* tests: Adapt tests to support service `reload` instead of `restart`

* chore: Additional logging for debugging change event logs

* fix: Wait on change detection, then verify directory created

Change detection is too fast now (0-1 seconds vs 5+).

Directory being waited on here was created near the end of a change event, reducing that time to detect a change by the utility method further.

We can instead check that the directory exists after the change detection event is completed.

* chore: Keep using the maildir polling check

We don't presently use remote storage in tests, but it might be relevant in future when testing NFS.

This at least avoids any confusing failure happening when that scenario is tested.
2022-12-24 01:57:24 +13:00

316 lines
11 KiB
Bash

#!/bin/bash
load "${REPOSITORY_ROOT}/test/test_helper/bats-support/load"
load "${REPOSITORY_ROOT}/test/test_helper/bats-assert/load"
NAME=${NAME:-mailserver-testing:ci}
# default timeout is 120 seconds
TEST_TIMEOUT_IN_SECONDS=${TEST_TIMEOUT_IN_SECONDS:-120}
NUMBER_OF_LOG_LINES=${NUMBER_OF_LOG_LINES:-10}
# @param ${1} timeout
# @param --fatal-test <command eval string> additional test whose failure aborts immediately
# @param ... test to run
function repeat_until_success_or_timeout {
local FATAL_FAILURE_TEST_COMMAND
if [[ "${1}" == "--fatal-test" ]]; then
FATAL_FAILURE_TEST_COMMAND="${2}"
shift 2
fi
if ! [[ "${1}" =~ ^[0-9]+$ ]]; then
echo "First parameter for timeout must be an integer, received \"${1}\""
return 1
fi
local TIMEOUT=${1}
local STARTTIME=${SECONDS}
shift 1
until "${@}"
do
if [[ -n ${FATAL_FAILURE_TEST_COMMAND} ]] && ! eval "${FATAL_FAILURE_TEST_COMMAND}"; then
echo "\`${FATAL_FAILURE_TEST_COMMAND}\` failed, early aborting repeat_until_success of \`${*}\`" >&2
return 1
fi
sleep 1
if [[ $(( SECONDS - STARTTIME )) -gt ${TIMEOUT} ]]; then
echo "Timed out on command: ${*}" >&2
return 1
fi
done
}
# like repeat_until_success_or_timeout but with wrapping the command to run into `run` for later bats consumption
# @param ${1} timeout
# @param ... test command to run
function run_until_success_or_timeout {
if ! [[ ${1} =~ ^[0-9]+$ ]]; then
echo "First parameter for timeout must be an integer, received \"${1}\""
return 1
fi
local TIMEOUT=${1}
local STARTTIME=${SECONDS}
shift 1
until run "${@}" && [[ $status -eq 0 ]]
do
sleep 1
if (( SECONDS - STARTTIME > TIMEOUT )); then
echo "Timed out on command: ${*}" >&2
return 1
fi
done
}
# @param ${1} timeout
# @param ${2} container name
# @param ... test command for container
function repeat_in_container_until_success_or_timeout() {
local TIMEOUT="${1}"
local CONTAINER_NAME="${2}"
shift 2
repeat_until_success_or_timeout --fatal-test "container_is_running ${CONTAINER_NAME}" "${TIMEOUT}" docker exec "${CONTAINER_NAME}" "${@}"
}
function container_is_running() {
[[ "$(docker inspect -f '{{.State.Running}}' "${1}")" == "true" ]]
}
# @param ${1} port
# @param ${2} container name
function wait_for_tcp_port_in_container() {
repeat_until_success_or_timeout --fatal-test "container_is_running ${2}" "${TEST_TIMEOUT_IN_SECONDS}" docker exec "${2}" /bin/sh -c "nc -z 0.0.0.0 ${1}"
}
# @param ${1} name of the postfix container
function wait_for_smtp_port_in_container() {
wait_for_tcp_port_in_container 25 "${1}"
}
# @param ${1} name of the postfix container
function wait_for_smtp_port_in_container_to_respond() {
local COUNT=0
until [[ $(docker exec "${1}" timeout 10 /bin/sh -c "echo QUIT | nc localhost 25") == *"221 2.0.0 Bye"* ]]; do
if [[ $COUNT -eq 20 ]]
then
echo "Unable to receive a valid response from 'nc localhost 25' within 20 seconds"
return 1
fi
sleep 1
((COUNT+=1))
done
}
# @param ${1} name of the postfix container
function wait_for_amavis_port_in_container() {
wait_for_tcp_port_in_container 10024 "${1}"
}
# TODO: Should also fail early on "docker logs ${1} | egrep '^[ FATAL ]'"?
# @param ${1} name of the postfix container
function wait_for_finished_setup_in_container() {
local STATUS=0
repeat_until_success_or_timeout --fatal-test "container_is_running ${1}" "${TEST_TIMEOUT_IN_SECONDS}" sh -c "docker logs ${1} | grep 'is up and running'" || STATUS=1
if [[ ${STATUS} -eq 1 ]]; then
echo "Last ${NUMBER_OF_LOG_LINES} lines of container \`${1}\`'s log"
docker logs "${1}" | tail -n "${NUMBER_OF_LOG_LINES}"
fi
return ${STATUS}
}
SETUP_FILE_MARKER="${BATS_TMPDIR}/$(basename "${BATS_TEST_FILENAME}").setup_file"
# get the private config path for the given container or test file, if no container name was given
function private_config_path() {
echo "${PWD}/test/duplicate_configs/${1:-$(basename "${BATS_TEST_FILENAME}")}"
}
# @param ${1} relative source in test/config folder
# @param ${2} (optional) container name, defaults to ${BATS_TEST_FILENAME}
# @return path to the folder where the config is duplicated
function duplicate_config_for_container() {
local OUTPUT_FOLDER
OUTPUT_FOLDER=$(private_config_path "${2}") || return $?
rm -rf "${OUTPUT_FOLDER:?}/" || return $? # cleanup
mkdir -p "${OUTPUT_FOLDER}" || return $?
cp -r "${PWD}/test/config/${1:?}/." "${OUTPUT_FOLDER}" || return $?
echo "${OUTPUT_FOLDER}"
}
function container_has_service_running() {
local CONTAINER_NAME="${1}"
local SERVICE_NAME="${2}"
docker exec "${CONTAINER_NAME}" /usr/bin/supervisorctl status "${SERVICE_NAME}" | grep RUNNING >/dev/null
}
function wait_for_service() {
local CONTAINER_NAME="${1}"
local SERVICE_NAME="${2}"
repeat_until_success_or_timeout --fatal-test "container_is_running ${CONTAINER_NAME}" "${TEST_TIMEOUT_IN_SECONDS}" \
container_has_service_running "${CONTAINER_NAME}" "${SERVICE_NAME}"
}
function wait_for_changes_to_be_detected_in_container() {
local CONTAINER_NAME="${1}"
local TIMEOUT=${TEST_TIMEOUT_IN_SECONDS}
# shellcheck disable=SC2016
repeat_in_container_until_success_or_timeout "${TIMEOUT}" "${CONTAINER_NAME}" bash -c 'source /usr/local/bin/helpers/index.sh; _obtain_hostname_and_domainname; cmp --silent -- <(_monitored_files_checksums) "${CHKSUM_FILE}" >/dev/null'
}
# Relies on ENV `LOG_LEVEL=debug` or higher
function wait_until_change_detection_event_completes() {
local CONTAINER_NAME="${1}"
# Ensure early failure if arg is missing:
assert_not_equal "${CONTAINER_NAME}" ""
# Ensure the container is configured with the required `LOG_LEVEL` ENV:
assert_regex \
$(docker exec "${CONTAINER_NAME}" env | grep '^LOG_LEVEL=') \
'=(debug|trace)$'
local CHANGE_EVENT_START='Change detected'
local CHANGE_EVENT_END='Completed handling of detected change' # debug log
function __change_event_status() {
docker exec "${CONTAINER_NAME}" \
grep -oE "${CHANGE_EVENT_START}|${CHANGE_EVENT_END}" /var/log/supervisor/changedetector.log \
| tail -1
}
function __is_changedetector_processing() {
[[ $(__change_event_status) == "${CHANGE_EVENT_START}" ]]
}
function __is_changedetector_finished() {
[[ $(__change_event_status) == "${CHANGE_EVENT_END}" ]]
}
# A new change event is expected,
# If the last event status is not yet `CHANGE_EVENT_START`, wait until it is:
if ! __is_changedetector_processing
then
repeat_until_success_or_timeout 60 __is_changedetector_processing
fi
# Change event is in progress, wait until it finishes:
repeat_until_success_or_timeout 60 __is_changedetector_finished
}
# An account added to `postfix-accounts.cf` must wait for the `changedetector` service
# to process the update before Dovecot creates the mail account and associated storage dir:
function wait_until_account_maildir_exists() {
local CONTAINER_NAME=$1
local MAIL_ACCOUNT=$2
local LOCAL_PART="${MAIL_ACCOUNT%@*}"
local DOMAIN_PART="${MAIL_ACCOUNT#*@}"
local MAIL_ACCOUNT_STORAGE_DIR="/var/mail/${DOMAIN_PART}/${LOCAL_PART}"
repeat_in_container_until_success_or_timeout 60 "${CONTAINER_NAME}" bash -c "[[ -d ${MAIL_ACCOUNT_STORAGE_DIR} ]]"
}
function add_mail_account_then_wait_until_ready() {
local CONTAINER_NAME=$1
local MAIL_ACCOUNT=$2
# Password is optional (omit when the password is not needed during the test)
local MAIL_PASS="${3:-password_not_relevant_to_test}"
run docker exec "${CONTAINER_NAME}" setup email add "${MAIL_ACCOUNT}" "${MAIL_PASS}"
assert_success
wait_until_account_maildir_exists "${CONTAINER_NAME}" "${MAIL_ACCOUNT}"
}
function wait_for_empty_mail_queue_in_container() {
local CONTAINER_NAME="${1}"
local TIMEOUT=${TEST_TIMEOUT_IN_SECONDS}
# shellcheck disable=SC2016
repeat_in_container_until_success_or_timeout "${TIMEOUT}" "${CONTAINER_NAME}" bash -c '[[ $(mailq) == *"Mail queue is empty"* ]]'
}
# Common defaults appropriate for most tests, override vars in each test when necessary.
# For all tests override in `setup_file()` via an `export` var.
# For individual test override the var via `local` var instead.
#
# For example, if you need an immutable config volume that can't be affected by other tests
# in the file, then use `local TEST_TMP_CONFIG=$(duplicate_config_for_container . "${UNIQUE_ID_HERE}")`
function init_with_defaults() {
export TEST_NAME TEST_TMP_CONFIG
# In `setup_file()` the default name to use for the currently tested docker container
# is `${TEST_NAME}` global defined here. It derives the name from the test filename:
# `basename` to ignore absolute dir path and file extension, only extract filename.
TEST_NAME=$(basename "${BATS_TEST_FILENAME}" '.bats')
# In `setup_file()` creates a single copy of the test config folder to use for an entire test file:
TEST_TMP_CONFIG=$(duplicate_config_for_container . "${TEST_NAME}")
# Common complimentary test files, read-only safe to share across containers:
export TEST_FILES_CONTAINER_PATH='/tmp/docker-mailserver-test'
export TEST_FILES_VOLUME="${PWD}/test/test-files:${TEST_FILES_CONTAINER_PATH}:ro"
# The config volume cannot be read-only as some data needs to be written at container startup
# - two sed failures (unknown lines)
# - dovecot-quotas.cf (setup-stack.sh:_setup_dovecot_quotas)
# - postfix-aliases.cf (setup-stack.sh:_setup_postfix_aliases)
# TODO: Check how many tests need write access. Consider using `docker create` + `docker cp` for easier cleanup.
export TEST_CONFIG_VOLUME="${TEST_TMP_CONFIG}:/tmp/docker-mailserver"
# The common default FQDN assigned to the container `--hostname` option:
export TEST_FQDN='mail.my-domain.com'
# Default Root CA cert used in TLS tests with `openssl` commands:
export TEST_CA_CERT="${TEST_FILES_CONTAINER_PATH}/ssl/example.test/with_ca/ecdsa/ca-cert.ecdsa.pem"
}
# Using `create` and `start` instead of only `run` allows to modify
# the container prior to starting it. Otherwise use this combined method.
# NOTE: Forwards all args to the create method at present.
function common_container_setup() {
common_container_create "$@"
common_container_start
}
# Common docker setup is centralized here.
#
# `X_EXTRA_ARGS` - Optional: Pass an array by it's variable name as a string, it will
# be used as a reference for appending extra config into the `docker create` below:
#
# NOTE: Using array reference for a single input parameter, as this method is still
# under development while adapting tests to it and requirements it must serve (eg: support base config matrix in CI)
function common_container_create() {
[[ -n ${1} ]] && local -n X_EXTRA_ARGS=${1}
run docker create --name "${TEST_NAME}" \
--hostname "${TEST_FQDN}" \
--tty \
--volume "${TEST_FILES_VOLUME}" \
--volume "${TEST_CONFIG_VOLUME}" \
"${X_EXTRA_ARGS[@]}" \
"${NAME}"
assert_success
}
function common_container_start() {
run docker start "${TEST_NAME}"
assert_success
wait_for_finished_setup_in_container "${TEST_NAME}"
}