1758: Implement a simpler credential cache (alternative to #1755) r=mergify[bot] a=nextgens
## What type of PR?
Feature: it implements a credential cache to speedup authentication requests.
## What does this PR do?
Credentials are stored in cold-storage using a slow, salted/iterated hash function to prevent offline bruteforce attacks. This creates a performance bottleneck for no valid reason (see the
rationale/long version on https://github.com/Mailu/Mailu/issues/1194#issuecomment-762115549).
The new credential cache makes things fast again.
This is the simpler version of #1755 (with no new dependencies)
### Related issue(s)
- close#1411
- close#1194
- close#1755
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x] In case of feature or enhancement: documentation updated accordingly
- [x] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
1776: optimize generation of transport nexthop r=mergify[bot] a=ghostwheel42
## What type of PR?
bug-fix and enhancement.
## What does this PR do?
Possibly there should be more input validation when editing a relay, but for now this tries to make the best out of the existing "smtp" attribute while maintaining backwards compatibility. When relay is empty, the transport's nexthop is the MX of the relayed domain to fix#1588
```
RELAY NEXTHOP TRANSPORT
empty use MX of relay domain smtp:domain
:port use MX of relay domain and use port smtp:domain:port
target resolve A/AAAA of target smtp:[target]
target:port resolve A/AAAA of target and use port smtp:[target]:port
mx:target resolve MX of target smtp:target
mx:target:port resolve MX of target and use port smtp:target:port
lmtp:target resolve A/AAAA of target lmtp:target
lmtp:target:port resolve A/AAAA of target and use port lmtp:target:port
target can also be an IPv4 or IPv6 address (an IPv6 address must be enclosed in []: [2001:DB8::]).
```
When there is proper input validation and existing database entries are migrated this function can be made much shorter again.
### Related issue(s)
- closes#1588
- closes#1815
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x] In case of feature or enhancement: documentation updated accordingly
- [X] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
Co-authored-by: Florent Daigniere <nextgens@freenetproject.org>
Co-authored-by: Alexander Graf <ghostwheel42@users.noreply.github.com>
1831: Fix roundcube database env configuration r=mergify[bot] a=parisni
## What type of PR?
bug-fix
## What does this PR do?
Both roundcube and mailu admin website can be backed by postgres/mysql. Before this PR, the `DB_FLAVOR` is shared by both services. However, the other roundcube dedicated DB params are prefixed with `ROUNDCUBE_`. (eg: `ROUNDCUBE_DB_NAME`)
There is no reason to share the DB_FLAVOR for both: This PR makes them be considered independently to make things clear and avoid bugs.
Also, the roundcube_db_flavor and db_flavor are made separated in this PR. However for simplicity, the template generator bind them : roundcube_db_flavor = db_flavor. This makes the template generator UI more simple. I considered most of the time people want to have both roundcube and mailu share the same RDBMS.
Also, AFAIK the internal postgresql service is deprecated and will be removed in 1.9. This is why this PR does not integrate roundcube in postgres when the internal DB is choosen: in case of internal postgres, the roundcube is backed with sqlite.
Both documentation and setup website have been updated accordingly.
### Related issue(s)
- Auto close an issue like: closes#1648#1471
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [X] In case of feature or enhancement: documentation updated accordingly
- [X] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
Co-authored-by: parisni <nicolas.paris@riseup.net>
Co-authored-by: Nicolas Paris <nicolas.paris@riseup.net>
1837: Set static hostname for antispam to preserve history. r=mergify[bot] a=Diman0
## What type of PR?
bug-fix
## What does this PR do?
It addresses bug #561. Rspamd stores the history in redis. The key for storing the history contains the hostname. On recreation of the docker container, the hostname changes and for this reason the rspamd history is lost. Setting a fixed hostname resolves this. Upon recreating the antispam container the history is retained.
### Related issue(s)
- Closes#561
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x] In case of feature or enhancement: documentation updated accordingly
- [x] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
Co-authored-by: Dimitri Huisman <diman@huisman.xyz>
1836: Test ci parallel r=Diman0 a=Diman0
## What type of PR?
enhancement
## What does this PR do?
Changes CI workflow to run all tests in parellel. After performing some tests (see #1830 ), I determined that using actions/cache to only cache a tar.gz. file with all build images and use this for all parallel tests is the fasted solution.
With Travis builds took ~30 minutes. Now each build runs for a maximum of 20 minutes (bors test and merge on master).
Bors r+ runs take about ~16/17 minutes.
### Related issue(s)
- Auto close an issue like: closes#1830
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x] In case of feature or enhancement: documentation updated accordingly
- [x] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
Co-authored-by: Dimitri Huisman <diman@huisman.xyz>
Co-authored-by: Dimitri Huisman <52963853+Diman0@users.noreply.github.com>
1746: DNS records for client autoconfiguration (RFC6186) r=Diman0 a=nextgens
## What type of PR?
Feature
## What does this PR do?
Add instructions on how to configure rfc6186 DNS records for client autoconfiguration
### Related issue(s)
- #224
- #498
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x ] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
Co-authored-by: Florent Daigniere <nextgens@freenetproject.org>
Workflow file only triggers if it resides in the branch. If an old PR based on the mailu repo without CI.yml is tested, then the workflow run will not trigger. The merged commit on TESTING/STAGING branch does not contain the required CI.yml workflow file after all. In these cases simply run the workflow manually on the TESTING or STAGING branch,
1694: update compression algorithms for current dovecot r=nextgens a=lub
## What type of PR?
enhancement
## What does this PR do?
This adds additional compression algorithms in accordance with
https://doc.dovecot.org/configuration_manual/zlib_plugin/
### Related issue(s)
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x] In case of feature or enhancement: documentation updated accordingly
- [x] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
Co-authored-by: lub <git@lubiland.de>
1806: Reflect override settings for postfix r=mergify[bot] a=ineiti
## What type of PR?
(documentation)
## What does this PR do?
Fixes path in the documentation setting, and also adds how to change postfix.cf
Co-authored-by: Linus Gasser <linus.gasser@epfl.ch>
1649: Update docs/reverse.rst with Traefik v2+ info r=mergify[bot] a=patryk-tech
## What type of PR?
Documentation
## What does this PR do?
Adds information about using Traefik v2+ as a reverse proxy.
### Related issue(s)
Closes#1503
## Prerequistes
Before we can consider review and merge, please make sure the following list is done and checked.
If an entry in not applicable, you can check it or remove it from the list.
- [x] In case of feature or enhancement: documentation updated accordingly
- [x] Unless it's docs or a minor change: add [changelog](https://mailu.io/master/contributors/guide.html#changelog) entry file.
1673: Remove rspamd unused env var from start script r=mergify[bot] a=cbachert
## What type of PR?
Cleanup
## What does this PR do?
Remove unused environment variable FRONT_ADDRESS in rspamd. FRONT_ADDRESS references were removed with commit 8172f3e in PR #727 like mentioned in chat https://matrix.to/#/!MINuyJjJSrfowljYCK:tedomum.net/$160401946364NGNmI:imninja.net?via=huisman.xyz&via=matrix.org&via=imninja.net
```
Mailu$ grep -r "FRONT_ADDRESS" core/rspamd/
core/rspamd/start.py:os.environ["FRONT_ADDRESS"] = system.get_host_address_from_environment("FRONT", "front")
```
### Related issue(s)
N/A
## Prerequistes
- [x] Documentation updated accordingly: No documentation to update
- [x] Add to changelog: Minor change
Co-authored-by: Patryk Tech <git@patryk.tech>
Co-authored-by: cbachert <cbachert@users.noreply.github.com>
1718: Warn people off of the documentation's K8s recipe. r=mergify[bot] a=c4lliope
Based on a discussion on Matrix (2020.12.19),
the helm charts are the way to go.
1793: Fix Typo in Dutch Translation r=mergify[bot] a=DjVinnii
## What type of PR?
Enhancement
## What does this PR do?
There were 2 typos in the Dutch translation file. These typos are being fixed with this PR.
### Related issue(s)
- None
Co-authored-by: Grace <30454698+c4lliope@users.noreply.github.com>
Co-authored-by: Vincent Kling <vincentkling@msn.com>
- the session key is now generated using
- a hash of the uid seeded by the apps secret_key (size: SESSION_KEY_BITS)
- a random token (size: 128 bits)
- the session's creation time (size: 32 bits)
- redis server side sessions are now refreshed after 1/2 the session lifetime
even if not modified
- the cookie is also updated if necessary
bors[bot]
lub