github/containerd
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 2 Wiki Activity

scripts/critest.sh: Prepare for userns tests in runc

Browse Source 
When runc 1.2.0 is released, it will expose support for userns and
therefore the critest suite will run those tests. The thing is, runc
needs to be able to traverse the path to mount the rootfs on itself.

Let's just mark the paths from the BDIR upwards with +x permissions, so
the tests run fine. Containerd already makes sure that the paths below
(the ones it creates) have the right permissions and for the right
group, etc.

I've tested with runc 1.2.0-rc.2 and CI fails without this path, with
this patch it works just fine.

Signed-off-by: Rodrigo Campos <rodrigoca@microsoft.com>
This commit is contained in:
Rodrigo Campos 2024-07-03 16:10:19 +02:00
parent 1117f663e7
commit 89a2cac377
1 changed files with 19 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
script/critest.sh
View File

@ -19,7 +19,26 @@ set -eu -o pipefail
report_dir=$1
mkdir -p $report_dir
function traverse_path() {
local path=$1
cd "$path"
sudo chmod go+rx "$PWD"
while [ $PWD != "/" ]; do
sudo chmod go+x "$PWD/../"
cd ..
done
}
BDIR="$(mktemp -d -p $PWD)"
# runc needs to traverse (+x) the directories in the path to the rootfs. This is important when we
# create a user namespace, as the final stage of the runc initialization is not as root on the host.
# While containerd creates the directories with the right permissions, the right group (so only the
# hostGID has access, etc.), those directories live below $BDIR. So, to make sure runc can traverse
# the directories, let's fix the dirs from $BDIR up, as the ones below are managed by containerd
# that does the right thing.
traverse_path "$BDIR"
function cleanup() {
pkill containerd || true