Skip to content

Add cgroups v2 support for Jammy stemcells #468

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mkocher wants to merge 2 commits into
base: ubuntu-jammy
Choose a base branch
from
+46 −12
Open

Add cgroups v2 support for Jammy stemcells #468

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2c6f904
Add cgroups v2 support for Jammy stemcells
mkocher Jan 28, 2026
2b27253
Prevent systemd-binfmt from running in containers
KauzClay Feb 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,3 +13,4 @@ acceptance-tests/os-conf-release
**/*.log

ci/docker/VMware-ovftool-*.bundle
tmp/
24 changes: 18 additions & 6 deletions stemcell_builder/stages/bosh_monit/assets/monit-access-helper.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,12 +13,24 @@
monit_isolation_classid=2958295041

permit_monit_access() {
net_cls_location="$(cat /proc/self/mounts | grep ^cgroup | grep net_cls | awk '{ print $2 }' )"
net_cls_subproc="$(grep net_cls /proc/self/cgroup | awk -F ":" '{ print $3 }' )"
monit_access_cgroup="${net_cls_location}/${net_cls_subproc}/monit-api-access"
if grep -q '^0::' /proc/self/cgroup 2>/dev/null; then
# cgroupv2 (unified hierarchy)
# Create a sub-cgroup under the current process's cgroup and move into it.
# The iptables rules match on this cgroup path.
cgroup_mount="$(awk '$3 == "cgroup2" { print $2 }' /proc/self/mounts)"
current_cgroup="$(grep '^0::' /proc/self/cgroup | cut -d: -f3)"
monit_access_cgroup="${cgroup_mount}${current_cgroup}/monit-api-access"

mkdir -p "${monit_access_cgroup}"
echo "${monit_isolation_classid}" > "${monit_access_cgroup}/net_cls.classid"
mkdir -p "${monit_access_cgroup}"
echo $$ > "${monit_access_cgroup}/cgroup.procs"
else
# cgroupv1 - use net_cls classid
net_cls_location="$(cat /proc/self/mounts | grep ^cgroup | grep net_cls | awk '{ print $2 }')"
net_cls_subproc="$(grep net_cls /proc/self/cgroup | awk -F ":" '{ print $3 }')"
monit_access_cgroup="${net_cls_location}/${net_cls_subproc}/monit-api-access"

echo $$ > "${monit_access_cgroup}/tasks"
mkdir -p "${monit_access_cgroup}"
echo "${monit_isolation_classid}" > "${monit_access_cgroup}/net_cls.classid"
echo $$ > "${monit_access_cgroup}/tasks"
fi
}
22 changes: 16 additions & 6 deletions stemcell_builder/stages/bosh_monit/assets/restrict-monit-api-access
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,13 +2,23 @@

source /var/vcap/bosh/etc/monit-access-helper.sh

if iptables -t mangle -C POSTROUTING -d 127.0.0.1 -p tcp --dport 2822 \
-m cgroup \! --cgroup "${monit_isolation_classid}" -j DROP
then
/bin/true
if grep -q '^0::' /proc/self/cgroup 2>/dev/null; then
# cgroupv2: dynamically determine the cgroup path for this process.
# The agent calls permit_monit_access() to join the monit-api-access sub-cgroup.
current_cgroup="$(grep '^0::' /proc/self/cgroup | cut -d: -f3)"
cgroup_match="--path ${current_cgroup}/monit-api-access"
else
# cgroupv1: use the classid from monit-access-helper.sh
cgroup_match="--cgroup ${monit_isolation_classid}"
fi

# Add iptables rules if they don't already exist.
# The DROP rule blocks traffic to monit (port 2822) from processes outside the monit cgroup.
# The ESTABLISHED,RELATED rule ensures existing connections aren't broken.
if ! iptables -t mangle -C POSTROUTING -d 127.0.0.1 -p tcp --dport 2822 \
-m cgroup ! ${cgroup_match} -j DROP 2>/dev/null; then
iptables -t mangle -I POSTROUTING -d 127.0.0.1 -p tcp --dport 2822 \
-m cgroup \! --cgroup "${monit_isolation_classid}" -j DROP
-m cgroup ! ${cgroup_match} -j DROP
iptables -t mangle -I POSTROUTING -d 127.0.0.1 -p tcp --dport 2822 \
-m state --state ESTABLISHED,RELATED -j ACCEPT
-m state --state ESTABLISHED,RELATED -j ACCEPT
fi
11 changes: 11 additions & 0 deletions stemcell_builder/stages/bosh_systemd/apply.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,3 +12,14 @@ source $base_dir/lib/prelude_bosh.bash
run_in_chroot $chroot "
echo 'RemoveIPC=no' >> /etc/systemd/logind.conf
"

# Prevent systemd-binfmt from running in containers.
# When running in a privileged container (e.g., Docker CPI on Apple Silicon),
# this service clears the host's binfmt_misc registrations (including Rosetta),
# causing "exec format error" for all subsequent x86_64 processes.
mkdir -p $chroot/etc/systemd/system/systemd-binfmt.service.d

cat > $chroot/etc/systemd/system/systemd-binfmt.service.d/skip-in-container.conf <<EOF
[Unit]
ConditionVirtualization=!container
EOF
Loading