UnrealIRCd 6.1.2 focuses on adding spamfilter features but also contains various other new features and some fixes.
spamfilter::rule crash in 6.1.2. SETNAME when using the SPAMFILTER u target. spamfilter { } blocks and a possible crash on REHASH if you have typos or other errors in the config file. Also fixing ::exclude-security-group not working and it now gives DNSBL lookups some more time.security-group { } block and mask item enhancements:channel "#xyz"; and channel "@#need_ops_here";rule "inchannel('@#main')||reputation()>1000";set::blacklist::recheck-time and can also be set to never if you don't want rechecking.blacklist::recheck to no.set::reputation::score-bump-timer-minimum-channel-members. Setting this to 0 means to bump scores also for people who are in no channels at all, which was the behavior in previous UnrealIRCd versions. Note: this new feature won't work properly when you have any older UnrealIRCd servers on the network (older than 6.1.2), as the older servers will still bump scores for everyone, including users in no channels, and this higher score will get synced back eventually to all other servers.spamfilter { } block improvements:type spamfilter. However, for exempt users no action is taken or logged. This allows it to count normal hits and count hits for except users. The idea is that the hits for except users can be a useful measurement to detect false positives. These hit counts are exposed in SPAMFILTER and STATS spamfilter.spamfilter::rule: with minimal 'if'-like preconditions and functions. If this returns false then the spamfilter will not run at all (no hit).spamfilter::except: this is meant as an alternative to 'rule' and works like a regular except item. If this matches, then the spamfilter will not run at all (no hit).raw (or R on IRC) to match a raw command / IRC protocol line (except message tags), such as LIST*. Naturally one needs to be very careful with these since a wrong filter could cause all/essential traffic to be rejected.action item now supports multiple actions:stop to stop other spamfilters from processing.set to set a TAG on a user, or change the value of one. It also supports changing the reputation score.report to call a spamreport block, see next.set { central-spamfilter { enabled yes; } } to enable.set::central-spamfilter::feed decides which feed to use: fast for early access to spamfilter rules that are new, and standard (the default) for rules that have been in fast for a while.set::central-spamfilter::except defines who will never be affected by central spamfilters. By default it is: users with a reputation score of more than 2016 (7 days online unregged, or 3.5 days as identified user) or having a host of *.irccloud.com. Spam matches for users that fall in this ::except group are counted as false positives and no action is taken or logged.set::spamfilter::utf8 is now on by default:[] in regex.ę then also matches Ę.\p can then be used. For example the regex \p{Arabic} would block all Arabic script. See also this full list of scripts. Please use this new tool with care. Blocking an entire language or script is quite a drastic measure.set { spamfilter { utf8 no; } }set::spamfilter::show-message-content-on-hit you can now configure to hide the message content in spamfilter hit messages. Generally it is very useful to see if a spamfilter hit is correct or not, so the default is 'always', but it also has privacy implications so there is now this option to disable it.include "some-file-or-url" { restrict-config { name-of-block; name-of-block2; } }~flood extended ban. This mode allows you to exempt users from channel mode +f and +F. It was actually added in a previous version (6.1.0) but never made it to the release notes. The syntax is: ~flood:types:mask, where types are the same letters as used in channel mode +f. Example: +e ~flood:t:*!*@*.textflood.example.orgRPL_ISUPPORT via WATCHOPTS=A and then the syntax to actually use this is WATCH A +Nick1 +Nick2 etc.. In UnrealIRCd 6 there was a bug where it would not always correctly inform about the away status, that bug has now been fixed.unknown-users group.RPL_STATSSPAMF): Now includes hits and hits for users that are exempt, two counters inserted right before the last argument (the regex).place_host_ban to take_actionUpdates may require up to 24 hours to propagate to mirrors. If the following command doesn't work, please retry later:
sudo dnf upgrade --refresh --advisory=FEDORA-2023-239e87c78a
Please log in to add feedback.
This update has been submitted for testing by robert.
This update's test gating status has been changed to 'ignored'.
This update has been pushed to testing.
This update has been submitted for stable by bodhi.
This update has been pushed to stable.