Bug 1497279
| Summary: | [RFE] Add option to interpret fields in auditd's syslog plugin | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Dylan Gross <dgross> |
| Component: | audit | Assignee: | Steve Grubb <sgrubb> |
| Status: | CLOSED ERRATA | QA Contact: | Ondrej Moriš <omoris> |
| Severity: | medium | Docs Contact: | Mirek Jahoda <mjahoda> |
| Priority: | low | ||
| Version: | 8.0 | CC: | cww, dpal, kfujii, mjahoda, nathan.wahl, omoris, pvrabec, sgrubb, tscherf |
| Target Milestone: | rc | Keywords: | FutureFeature |
| Target Release: | 8.2 | Flags: | pm-rhel:
mirror+
|
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | audit-3.0-0.14.20191104git1c2f876 | Doc Type: | No Doc Update |
| Doc Text: |
See https://bugzilla.redhat.com/show_bug.cgi?id=1757986
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-04-28 16:46:58 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Dylan Gross
2017-09-29 15:53:26 UTC
The problem with this is that it opens you to parsing, TTY, and UTF based text attacks. It is designed this way for a reason. The logs must be translated by ausearch or something else to render them inert and present them safely. CVE-2015-5186 Audit: log terminal emulator escape sequences handling Steve, Understood, at least at a basic level. One suggestion that we mentioned in the RH support case is to create the flag for the audisp syslog plugin, rather than for audit or pam_tty_audit itself. Would this still create a vulnerability? I'm unsure. The end result that our SIEM team is seeking is essentially a different presentation format in /var/log/messages. However, to get there it has to pass through rsyslogd, rather than just printing to the console as it seems aureport and ausearch do. To get to their ArcSight server, I mean. I'll think about that. Maybe if its limited to a plugin that would be acceptable. Upstream commit 4669f76 should address this issue. There is now the possibility to "interpret" as a third argument in the audisp-syslog.conf file. Granting qa_ack+ for RHEL-8.2.0. Acceptance Criteria: * [gating] audit message sent via audisp-syslog plugin to syslog is interpreted when plugin is configured with interpret keyword audit-3.0-0.14.20191104git1c2f876 has been built to address this issue. Steve, I noticed that when interpretation is enabled, second msg field is missing. With 'interpret' keyword in syslog audisp plugin: Nov 11 05:45:19 ci-vm-10-0-137-114 audispd[21334]: type=USER msg=audit(11/11/19 05:45:19) : pid=21374 uid=root auid=root ses=5 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 exe=/usr/sbin/auditctl hostname=ci-vm-10-0-137-114.hosted.upshift.rdu2.redhat.com addr=? terminal=pts/1 res=success Without 'interpret': Nov 11 05:50:28 ci-vm-10-0-137-114 audispd[24237]: type=USER msg=audit(1573469428.411:1238): pid=24277 uid=0 auid=0 ses=5 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 msg='interpretation_test exe="/usr/sbin/auditctl" hostname=ci-vm-10-0-137-114.hosted.upshift.rdu2.redhat.com addr=? terminal=pts/1 res=success' UID="root" AUID="root" Unfortunately, the second msg= is stripped by auparse and I'm not 100% sure it can be reliably added back. Isn't auparse used in ausearch when '-i' option is used (because in ausearch msg is not stripped). In messages table [1] I see that AUDIT_USER is deprecated. Is it possible that msg stripping is issue in AUDIT_USER only? [1] https://github.com/linux-audit/audit-documentation/blob/master/specs/messages/message-dictionary.csv I think that the answer is that we need to do something like text= in USER events. OK, thanks, I file a bug for that and I am moving this one back to ON_QA since the issue is caused by interpreting this particular type of event and not by new functionality in syslog audisp plugin. Upstream commit 93c1354solves the missing text reported in comment #21. Will hold this patch for the moment to see if we need any other fixes. Rebuilt with above patch Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:1812 |