config: drop /sys/dev/block from masked paths

[giuseppe]

there is no real advantage in masking this path, since this information is already available under /sys/devices. In fact, the files under /sys/dev/block are just symlinks.

[root@953139a2851f /]# ls -l /sys/dev/block/252:0 lrwxrwxrwx. 1 root root 0 Dec 17 12:47 /sys/dev/block/252:0 -> ../../devices/virtual/block/zram0

Fixes: #2277
Fixes: https://issues.redhat.com/browse/RHEL-3115

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: giuseppe

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [giuseppe]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[packit-as-a-service]

Ephemeral COPR build failed. @containers/packit-build please check.

[rhatdan]

I thought this originally was something about lsblk causing issues where apps would think they had access to blk devices and then blew up with permission denied, and this caused tools to no longer think the devices was there. Although I might be misremembering.

Anyways I am fine with this change.

LGTM

[giuseppe]

lsblk works fine after this change:

# podman run --rm -it registry.access.redhat.com/ubi9/ubi:latest lsblk
[no output]
# bin/podman run --rm -it registry.access.redhat.com/ubi9/ubi:latest lsblk
NAME        MAJ:MIN RM   SIZE RO TYPE MOUNTPOINTS
zram0       252:0    0     8G  0 disk [SWAP]
nvme0n1     259:0    0 476.9G  0 disk 
|-nvme0n1p1 259:1    0   600M  0 part 
|-nvme0n1p2 259:2    0     1G  0 part 
`-nvme0n1p3 259:3    0 475.4G  0 part 

[rhatdan]

Right that is my point, the complaint was that it reported blk devices which from the containers point of view, were not actionable, so the code within the container attempts to do something with them, which generates a permission denied. If as in the current behaviour, id sees no blk devices it continues on happily.

[grooverdan]

I totally support the unmasking because of problems exposed in containers/podman#13708.

Its origins where https://bugzilla.redhat.com/show_bug.cgi?id=1772993 related to exposing host storage information inside an unprivileged container. The masking of the symlinks in /sys/dev/block didn't actually achieve this (mentioned previous - #2277, except where lsblk was considered the definitive exposure) because the expose was still present in /sys/devices. With /sys being a ro,nodev mount, the expose isn't prevented.