process limit unlimited (regression)
Bug #391761 reported by
Kees Cook
This bug affects 4 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
pam (Ubuntu) |
Fix Released
|
Low
|
Kees Cook |
Bug Description
AFAIK, RLIMIT_NPROC is determined by the kernel based on available physical memory. In my 512M VMs, I see the following results of "ulimit -u":
dapper: unlimited
hardy, intrepid: 4095
jaunty, karmic: unlimited
The intention is to have this set to in an attempt to reasonably mitigate fork-bombs without getting in the way of intentionally big process collections. Jaunty and Karmic appear to have regressed. I am assuming this is a PAM bug, but it may be a kernel issue. Tracking RLIMIT setting has always eluded me. :)
Related branches
Changed in pam (Ubuntu): | |
importance: | Undecided → High |
status: | New → Confirmed |
Changed in pam (Ubuntu): | |
importance: | High → Low |
security vulnerability: | no → yes |
tags: | added: regression-release |
Changed in pam (Ubuntu): | |
assignee: | nobody → Kees Cook (kees) |
status: | Triaged → Fix Committed |
To post a comment you must log in.
Noted from:
http:// dancingpenguins oflight. com/2009/ 06/watching- a-fork- bomb-explode/