serverPKI: | Python PKI for internet server infrastructure |
---|---|
Copyright: | Copyright (c) 2015-2020 Axel Rau axel.rau@chaos1.de |
License: | GPLv3 |
Homepage: | https://github.com/mc3/serverPKI |
Documentation: | https://serverpki.readthedocs.io |
serverPKI is a tool to issue, renew and distribute SSL certificates for internet servers. Distribution to target hosts and reloading of server configuration is done via ssh/sftp. Configuration and cert/key data is stored in a relational database.
serverPKI includes support for
- local CA
- LetsEncrypt CA (supports only acme v2 api, see https://letsencrypt.org/docs)
- FreeBSD service jails via ssh access to host
- publishing of DANE RR in DNS, using BIND 9 and TLSA key rollover (see RFC 6698)
- controlling DNS zone info for LetsEncrypt challenges und TLSA RR via dynamic DNS updates (recommended) or via zone files.
- unattended operation via cronjob
- extensive logging
- alerting via mail
- PostgreSQL 12+ server
- The contrib utilities from the PostgreSQL distribution are required (serverPKI needs the citext extension for case insensitive indexes)
- a DB account with super user privileges [dba] or assistance of a DB admin (serverPKI uses a dedicated DB user [pki_op] and a dedicated DB)
- authentication record in pg_hba.conf to allow access of pki_op from local host (client cert authentication recommended)
- PostgreSQL 12+ client installation on local host
- bind 9 DNS server (9.16+ should be used)
- If DNS is handled via zone files,
- serverPKI must be run on the master (hidden primary) DNS server.
- signed Zones being maintained by serverPKI must be run in auto-dnssec maintain + inline-signing operation mode.
- Zone files must be writable by serverPKI process to allow publishing of acme_challenges and TLSA resource records for DANE
- If DNS is handled via zone files,
- Python 3.7+ must be installed (tested with Python 3.8.3)
- Running serverPKI in a Python virtual environment is recommended for ease of upgrading. The author uses virtualenvwrapper.
This project is being developed with the powerful Python IDE PyCharm, which is particularly useful during remote debugging sessions. A professional license has been granted by JetBrains, https://www.jetbrains.com/.