Move action.d/mail-whois-common.conf into fail2ban-server
Yes, Hrrrm...
maxlines > 1) together with systemd backend,
now systemd-filter replaces newlines in message from systemd journal with \n (otherwise
multi-line parsing may be broken, because removal of matched string from multi-line buffer window
is confused by such extra new-lines, so they are retained and got matched on every followed
message, see gh-2431).local overwrites options of .conf for config-files
included with before/after)action.d/abuseipdb.conf: switched to use AbuseIPDB API v2 (gh-2302)action.d/badips.py: fixed start of banaction on demand (which may be IP-family related), gh-2390action.d/helpers-common.conf: rewritten grep arguments, now options -wF used to match only
whole words and fixed string (not as pattern), gh-2298filter.d/apache-auth.conf:normal mode (mode-controlled now) (gh-2548);mode - normal (default) and aggressivefilter.d/sshd.conf:Bad protocol version identification in ddos and aggressive modes (gh-2404).Disconnecting ...: Change of username or service not allowed (gh-2239, gh-2279)Disconnected from ... [preauth], preauth phase only, different handling by extra
(with supplied user only) and ddos/aggressive mode (gh-2115, gh-2239, gh-2279)filter.d/mysqld-auth.conf: filter.d/sendmail-reject.conf:mode=extra now captures port IDs of TLSMTA and MSA (defaults for ports 465 and 587 on some distros)files/fail2ban.service.in: fixed systemd-unit template - missing nftables dependency (gh-2313)action.d/mail*: fixed usage with multiple log files (ultimate fix for gh-976, gh-2341)filter.d/sendmail-reject.conf: fixed journal usage for some systems (e. g. CentOS): if only identifier
set to sm-mta (no unit sendmail) for some messages (gh-2385)filter.d/asterisk.conf: asterisk can log additional timestamp if logs into systemd-journal
(regex extended with optional part matching this, gh-2383)filter.d/postfix.conf:errors to match "too many errors" (gh-2439),
also included within modes normal, more (extra and aggressive), since postfix
parameter smtpd_hard_error_limit is default 20 (additionally consider maxretry)filter.d/named-refused.conf:prefregex extended, more selective now (denied/NOTAUTH suffix moved from failregex, so no catch-all there anymore)filter.d/sendmail-auth.conf, filter.d/sendmail-reject.conf :<CIDR> - helper regex to match CIDR (simple integer form of net-mask);<SUBNET> - regex to match sub-net adresses (in form of IP/CIDR, also single IP is matched, so part /CIDR is optional);<ADDR>, <HOST>, <SUBNET>) recognize IP addresses enclosed in square brackets<F-MLFGAINED> for failregex, signaled that the access to service was gained
(ATM used similar to tag <F-NOFAIL>, but it does not add the log-line to matches, gh-2279)logtype (default file for file-backends, and
journal for journal-backends, gh-2387); can be also set to rfc5424 to force filters (which include common.conf)
to use RFC 5424 conform prefix-line per default (gh-2467);logtype can be also used to
select short prefix-line for file-backends too for all filters using __prefix_line (common.conf),
if message logged only with hostname svc[nnnn] prefix (often the case on several systems):[jail]
backend = auto
filter = flt[logtype=short]
filter.d/common.conf: differentiate __prefix_line for file/journal logtype's (speedup and fix parsing
of systemd-journal);filter.d/traefik-auth.conf: used to ban hosts, that were failed through traefikfilter.d/znc-adminlog.conf: new filter for ZNC (IRC bouncer); requires the adminlog module to be loadeddbmaxmatches (fail2ban.conf) and maxmatches (jail.conf) to contol
how many matches per ticket fail2ban can hold in memory and store in database (gh-2402, gh-2118);[Thread] and option stacksize to configure default size
of the stack for threads running in fail2ban (gh-2356), it could be set in fail2ban.local to
avoid runtime error "can't start new thread" (see gh-969);fail2ban-client set <jain> banip <ip1> ... <ipN>fail2ban-client set <jain> unbanip [--report-absent] <ip1> ... <ipN>fail2ban-client set <jail> attempt <ip> [<failure-message1> ... <failure-messageN>]action.d/nftables.conf:nftables-allports supports multiple protocols in single rule nownftables:nftables-common is removed (replaced with single action nftables now)nftables-allports is obsolete, superseded by nftables[type=allports]nftables-multiport is obsolete, superseded by nftables[type=multiport]nftables[type=multiport] action (single set with multiple rules
in chain), following configuration in jail would replace 3 separate actions, see
https://github.com/fail2ban/fail2ban/pull/2254#issuecomment-534684675action.d/badips.py: option loglevel extended with level of summary message,
following example configuration logging summary with NOTICE and rest with DEBUG log-levels:
action = badips.py[loglevel="debug, notice"]fileOptions to set common filter/test options for whole test-file;actionreban or actionban if still not defined in action);actionrepair_on_unban set to true);--no-check-all to avoid check of all regex's (first matched only);-o, --out to set token only provided in output (disables check-all and outputs only expected data).Please log in to add feedback.
This update has been submitted for testing by orion.
This update's test gating status has been changed to 'waiting'.
This update's test gating status has been changed to 'ignored'.
This update has obsoleted fail2ban-0.10.5-1.el7, and has inherited its bugs and notes.
Ready to test it but only the version
fail2ban.noarch 0:0.10.5-1.el7is available for testing in epel-testingDo I am too hurry or I did something wrong
patience, young Jedi. -2 hasn't yet been pushed to testing.
Thank....
This update has been pushed to testing.
I am testing it, the service is running after the update. Please could you hold the release, I can see some log noises and I would like to know if it is our configuration or a fail2ban issue...thank in advance
The log noise come from our configuration, let me keep it some days, I have updated two servers, it seems to run well....
What is the deadline for the release ?
If this update does not receive any negative karma it will get pushed to stable in 14 days, unless it receives 3 positive karma before that.
work smoothly, thank
This update can be pushed to stable now if the maintainer wishes
This update has been submitted for stable by bodhi.
This update has been pushed to stable.