Closed intweed closed 2 months ago
@intweed: Thanks for opening an issue, it is currently awaiting triage.
In the meantime, you can:
Upstream issue: https://github.com/crowdsecurity/grokky/issues/3
Please read comment as to why it difficult to fix until the next release. In future we plan to remove all hard coded patterns and extend them to the normal patterns directory.
Version: 1.6.3
is planned release for monday which will include this fix.
What happened?
The official documentation specifies the following pattern for email parsing:
%{EMAILLOCALPART}@%{HOSTNAME}
where EMAILLOCALPART has the following regex pattern:
[a-zA-Z][a-zA-Z0-9_.+-=:]+
which does not include email addresses starting with a digits (for example, these are acceptable in Gmail). In this case, the parser cannot correctly process the log message and the corresponding scenario is not activated, no decisions are applied.
What did you expect to happen?
Crowdsec correctly processes log messages containing email addresses starting with digits.
How can we reproduce it (as minimally and precisely as possible)?
The above is observed on this configuration:
Crowdsec + Dominic-Wagner/vaultwarden collection + vaultwarden server (logs are collected via docker socket proxy)
Multiple login attempts with wrong credentials (where email address starting with digit) are not registered by Crowdsec.
cscli metrics show acquisition
displays the same value for "Lines read" and "Lines unparsed" for docker:/vaultwarden source. Thus, the bruteforce scenario does not apply.Anything else we need to know?
No response
Crowdsec version
OS version
Enabled collections and parsers
Acquisition config
Config show
Prometheus metrics
Related custom configs versions (if applicable) : notification plugins, custom scenarios, parsers etc.