summaryrefslogtreecommitdiffstats
path: root/roles/common
Commit message (Collapse)AuthorAgeFiles
* Firewall: Allow DNS queries over TCP.Guilhem Moulin2018-04-041
|
* APT: use deb.debian.org as archive source.Guilhem Moulin2018-04-041
|
* Perform recipient address verification on the MSA itself.Guilhem Moulin2018-04-042
|
* Upgrade syntax to Ansible 2.5.Guilhem Moulin2018-04-043
|
* Upgrade syntax to Ansible 2.4.Guilhem Moulin2017-11-231
|
* More logcheck-database tweaks.Guilhem Moulin2017-09-143
|
* Fix detection of KVM guests.Guilhem Moulin2017-07-293
|
* rkhunter: Disable remote updates to fix CVE-2017-7480.Guilhem Moulin2017-07-291
|
* Use MariaDB as default MySQL flavor.Guilhem Moulin2017-07-291
|
* Don't install debsecan anymore by default.Guilhem Moulin2017-06-262
| | | | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=789196
* Webmail: don't allow outgoing TCP/993 connections.Guilhem Moulin2017-06-151
| | | | We're going through IPsec to communicate with the IMAP server.
* More logcheck-database tweaks.Guilhem Moulin2017-06-071
|
* postfix-sender-login: wibbleGuilhem Moulin2017-06-051
|
* dovecot: enable user iteration and add a cronjob for `doveadm purge -A`Guilhem Moulin2017-06-051
|
* postfix: enable XFORWARD command from our internal relays.Guilhem Moulin2017-06-021
|
* postfix: don't rate-limit our IPsec subnet.Guilhem Moulin2017-06-022
|
* Don't let authenticated client use arbitrary sender addresses.Guilhem Moulin2017-06-011
| | | | | | | | | | | | | | The following policy is now implemented: * users can use their SASL login name as sender address; * alias and/or list owners can use the address as envelope sender; * domain postmasters can use arbitrary sender addresses under their domains; * domain owners can use arbitrary sender addresses under their domains, unless it is also an existing account name; * for known domains without owner or postmasters, other sender addresses are not allowed; and * arbitrary sender addresses under unknown domains are allowed.
* /lib/systemd/system → /etc/systemd/systemGuilhem Moulin2017-05-315
|
* Also install non-free firmwares on civett.Guilhem Moulin2017-05-302
|
* Change group of executables in /usr/local/{bin,sbin} from root to staff.Guilhem Moulin2017-05-142
|
* MSA: reject null sender address.Guilhem Moulin2017-05-141
|
* Fix Ansible 2.2.0 compatibility of a Jinja2 template.Guilhem Moulin2017-01-141
|
* More logcheck-database tweaks.Guilhem Moulin2016-12-081
|
* Postfix: ensure common aliases are present.Guilhem Moulin2016-09-182
|
* FreshClam: change ownership of /etc/clamav/freshclam.conf.Guilhem Moulin2016-09-181
| | | | | | | | To match the stock version shipped by clamav-freshclam 0.99.2+dfsg-0+deb8u2 ~$ stat -c '%U:%G %a' /etc/clamav/freshclam.conf clamav:adm 444
* Firewall: allow duplicates rules.Guilhem Moulin2016-09-181
|
* More logcheck-database tweaks.Guilhem Moulin2016-08-222
|
* postfix: Remove obsolete templates tls_policy/relay_clientcerts.Guilhem Moulin2016-07-121
|
* Route all internal SMTP traffic through IPsec.Guilhem Moulin2016-07-104
|
* Postfix: avoid hardcoding the instance names.Guilhem Moulin2016-07-101
|
* Postfix: don't share the master.cf between the instances.Guilhem Moulin2016-07-102
|
* Route SMTP traffic from the webmail through IPsec.Guilhem Moulin2016-07-101
|
* More logcheck-database tweaks.Guilhem Moulin2016-07-092
|
* Localize the NTP pool hostnames.Guilhem Moulin2016-07-091
|
* Localize the debian archive hostnames.Guilhem Moulin2016-07-091
|
* ClamAV (FreshClam): use a localized Database Mirror.Guilhem Moulin2016-07-092
| | | | | | As db.local.clamav.net is not always properly localized. Furthermore, our previous Ansiblee script did not ensure ordering of the DatabaseMirror lines.
* IPSec → IPsecGuilhem Moulin2016-06-295
|
* More logcheck-database tweaks.Guilhem Moulin2016-06-293
|
* update-firewall.sh: COMMIT empty iptables rule files.Guilhem Moulin2016-06-291
|
* Use stunnel to secure the connection from the webmail to ldap.fripost.org.Guilhem Moulin2016-06-051
| | | | | We should use IPSec instead, but doing so would force us to weaken slapd.conf's ‘security’ setting.
* typoGuilhem Moulin2016-05-241
|
* IPSec: replace (self-signed) X.509 certs by their raw pubkey for authentication.Guilhem Moulin2016-05-243
| | | | There is no need to bother with X.509 cruft here.
* genkeypair, gendhparam: use -rand /dev/urandom when generating keys or DH ↵Guilhem Moulin2016-05-222
| | | | parameters.
* Tunnel bacula (dir → {fd,sd} and fd → sd) traffic through IPSec.Guilhem Moulin2016-05-226
|
* Tunnel munin-update traffic through IPSec.Guilhem Moulin2016-05-227
|
* Tunnel internal NTP traffic through IPSec.Guilhem Moulin2016-05-222
| | | | | | | More precisely, between our NTP-master (stratum 1) host and the other machines (all stratum 2). Providing authentification and integrity for internal NTP traffic ensures a consistent time within our internal infrastructure.
* Set up IPSec tunnels between each pair of hosts.Guilhem Moulin2016-05-2213
| | | | | | | | | | | | | | | We use a dedicated, non-routable, IPv4 subnet for IPSec. Furthermore the subnet is nullrouted in the absence of xfrm lookup (i.e., when there is no matching IPSec Security Association) to avoid data leaks. Each host is associated with an IP in that subnet (thus only reachble within that subnet, either by the host itself or by its IPSec peers). The peers authenticate each other using RSA public key authentication. Kernel traps are used to ensure that connections are only established when traffic is detected between the peers; after 30m of inactivity (this value needs to be less than the rekeying period) the connection is brought down and a kernel trap is installed.
* postfix: master.cf wibbleGuilhem Moulin2016-05-181
|
* postfix: Update to recommended TLS settings.Guilhem Moulin2016-05-182
| | | | | | | | Following Viktor Dukhovni's 2015-08-06 recommendation http://article.gmane.org/gmane.mail.postfix.user/251935 (We're using stronger ciphers and protocols in our own infrastructure.)
* Move /etc/ssl/private/dhparams.pem to /etc/ssl/dhparams.pem and make it public.Guilhem Moulin2016-05-182
| | | | | | | | | | Ideally we we should also increase the Diffie-Hellman group size from 2048-bit to 3072-bit, as per ENISA 2014 report. https://www.enisa.europa.eu/publications/algorithms-key-size-and-parameters-report-2014 But we postpone that for now until we are reasonably certain that older client won't be left out.