Relationships
#4950 @swamp/cfgmgmt firewall/apt_repository/ssh: follow-ups from #4840 verification review
Opened by stack72 · 11/20/2026· Shipped 11/20/2026
Follow-ups left documented in PR #729 (swamp-club #4840, #4923, #4896, #4873). 1) iptables placement inserts by position: a rule another tool (fail2ban, docker) inserts between apply's re-read of the chain and the iptables -I can leave the new rule one place off, and the own-session guard ran against that listing; the next check reports it misplaced and apply moves it. Consider re-reading the chain after the insert and failing or retrying when the position is off. 2) iptables first counts negated and interface-scoped rules (e.g. -i lo) as conflicts, placing the rule higher than needed; skip rules iptablesRuleMatches would never take. 3) apt_repository checks and repairs only the key's own directory; a closed ancestor (e.g. a key under /root) still leaves apt unable to read it while check reports compliant; at least warn. 4) ufw placement re-reads the second family's listing but does not re-run the guards on it. 5) socketParentIsSafe checks only TMPDIR itself, not its ancestors; note it in the code comment as the README does.
Shipped
Click a lifecycle step above to view its details.
system commented 11/20/2026, 4:57:14 PM
Classified automatically when this issue was filed.
- Source: Extensions
If you feel this classification is incorrect, add a ripple to tell us so.
Sign in to post a ripple.