Absolute paths are being appended together causing an error.

[DeanRM]

Describe the bug
Certain docker commands are appending two container volume paths together causing an error.

To Reproduce
Unbound: v1.22.0
volumes:

• /opt/unbound:/opt/unbound/etc/unbound
• /opt/log:/var/log/unbound:rw

docker inspect unbound | grep -A 15 "Mounts"
"Mounts": [
{
"Type": "bind",
"Source": "/opt/unbound",
"Destination": "/opt/unbound/etc/unbound",
"Mode": "rw",
"RW": true,
"Propagation": "rprivate"
},
{
"Type": "bind",
"Source": "/opt/log",
"Destination": "/var/log/unbound",
"Mode": "rw",
"RW": true,
"Propagation": "rprivate"

Expected behavior
I expect the checkconf to use the absolute path given.

Error messages
Input: docker exec -it unbound unbound-checkconf /opt/unbound/etc/unbound/unbound.conf
Output: /opt/unbound/etc/unbound/var/log/unbound: No such file or directory
[1740683346] unbound-checkconf[122:0] fatal error: logfile directory does not exist

System:

• Unbound version: v1.22.0
• OS: Debian GNU/Linux 12 (bookworm)
• Docker version 20.10.24+dfsg1, build 297e128
• unbound -V output: [1740685756] unbound[688:0] notice: Start of unbound 1.22.0.

Additional information
Add any other information that you may have gathered about the issue here.

[gthess]

This seems like chroot configuration, in that case Unbound (unbound-checkconf) would want to see that the actual logging directory does exist.
Could you share your configuration?