[systemd-pr-build] $ /bin/sh -xe /tmp/hudson2438376146086078148.sh
+ /home/systemd/jenkins-glue.sh
Traceback (most recent call last):
  File "./slave-control.py", line 193, in <module>
    main()
  File "./slave-control.py", line 131, in main
    data = json.loads(json_data)
  File "/usr/lib64/python2.7/json/__init__.py", line 338, in loads
    return _default_decoder.decode(s)
  File "/usr/lib64/python2.7/json/decoder.py", line 365, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib64/python2.7/json/decoder.py", line 383, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded

@zonque , please take a look.
json_data = duffy_cmd("/Node/get", params) is not working

Fedora test is broken, rest is okay.

@systemd-centos-ci
retest this please

All looks good. merging.

@evverx i wonder though whether we can find a fix for the problem you tried to fix with 8492849? I mean, there was a good reason to make that change after all.

I wonder if there's a nice way to detect "foreign" mounts, so that PID 1 in the container could exclude them from umount.target Conflicts= lines. Or alternatively, we could add some logic that makes /tmp a user mount only when it is safe. I.e. basicely reuse your suggested logic, but conditionalize it on the whether there are any mounts configured with /tmp as prefix?

I wonder if there's a nice way to detect "foreign" mounts, so that PID 1 in the container could exclude them from umount.target Conflicts= lines

I think the only reliable way is to ask nspawn via some communication channel. But that's the nspawn-specific way and it doesn't fix something like #4810

I think the only reliable way is to ask nspawn via some communication channel. But that's the nspawn-specific way and it doesn't fix something like #4810

I wonder if /proc/self/mountinfo in some way let's us know this...

Hm, that's a good question. I didn't think about it.