| 1 |
On 1/3/20 3:46 PM, Rich Freeman wrote: |
| 2 |
> If OpenRC contains a vulnerability wouldn't it make more sense to set |
| 3 |
> this as part of OpenRC, |
| 4 |
Indeed. |
| 5 |
|
| 6 |
Furthermore there's a nifty page https://kernsec.org/wiki/index.php/Kernel_Self_Protection_Project/Recommended_Settings |
| 7 |
which yields for me to this /etc/sysctl.d/local.conf : |
| 8 |
|
| 9 |
|
| 10 |
# Restrict potential illegal access via links |
| 11 |
# |
| 12 |
fs.protected_hardlinks = 1 |
| 13 |
fs.protected_symlinks = 1 |
| 14 |
|
| 15 |
# |
| 16 |
# https://kernsec.org/wiki/index.php/Kernel_Self_Protection_Project#CONFIGs |
| 17 |
# |
| 18 |
|
| 19 |
# Try to keep kernel address exposures out of various /proc files (kallsyms, modules, etc). |
| 20 |
kernel.kptr_restrict = 1 |
| 21 |
|
| 22 |
# Avoid kernel memory address exposures via dmesg. |
| 23 |
kernel.dmesg_restrict = 1 |
| 24 |
|
| 25 |
# Block non-uid-0 profiling (needs distro patch, otherwise this is the same as "= 2") |
| 26 |
kernel.perf_event_paranoid = 3 |
| 27 |
|
| 28 |
# Turn off kexec, even if it's built in. |
| 29 |
kernel.kexec_load_disabled = 1 |
| 30 |
|
| 31 |
# Avoid non-ancestor ptrace access to running processes and their credentials. |
| 32 |
kernel.yama.ptrace_scope = 1 |
| 33 |
|
| 34 |
# Disable User Namespaces, as it opens up a large attack surface to unprivileged users. |
| 35 |
user.max_user_namespaces = 0 |
| 36 |
|
| 37 |
# Turn off unprivileged eBPF access. |
| 38 |
kernel.unprivileged_bpf_disabled = 1 |
| 39 |
|
| 40 |
# Turn on BPF JIT hardening, if the JIT is enabled. |
| 41 |
net.core.bpf_jit_harden = 2 |
| 42 |
|
| 43 |
|
| 44 |
-- |
| 45 |
Toralf |
| 46 |
PGP 23217DA7 9B888F45 |
Archives