Forward more messages on the sd-notify socket

[alexlarsson]

Several of the standard sd-notify messages are safe to use from a container and are very useful. This commit cleans up the general handling of notify messages and allows forwarding of:

• READY=1
• RELOADING=1
• STOPPING=1
• WATCHDOG=1
• WATCHDOG=trigger
• STATUS=...
• ERRNO=...
• BUSERROR=...
• MONOTONIC_USEC...

See https://www.freedesktop.org/software/systemd/man/latest/sd_notify.html#Well-known%20assignments for documentation of these.

Note: We don't allow unknown messages to be forewarded. For one, all the file-descriptor based ones are currently unsupported (since the forwarding doesn't handle fds), but also some options (current and future) may be security sensitive.

fixes #461
fixes #311

[rhatdan]

LGTM
@giuseppe @haircommander PTAL

[vrothberg]

Please make sure to run this change against Podman's system tests test/system. I think things should work fine but who knows.

[eriksjolund]

This PR also fixes #311

[giuseppe]

LGTM

[alexlarsson]

So, I ran the podman test/system, and we're failing on this:

 ✗ [075] podman exec - cat from stdin
   (from function `bail-now' in file test/system/helpers.bash, line 227,
    from function `is' in file test/system/helpers.bash, line 956,
    in test file test/system/075-exec.bats, line 84)
     `is "${lines[0]}" "3000+0 records in"  "dd: number of records in"' failed
   
   [11:10:24.237041104] $ /vcs/other/podman/bin/podman rm -t 0 --all --force --ignore
   
   [11:10:24.272179180] $ /vcs/other/podman/bin/podman ps --all --external --format {{.ID}} {{.Names}}
   
   [11:10:24.307984108] $ /vcs/other/podman/bin/podman images --all --format {{.Repository}}:{{.Tag}} {{.ID}}
   [11:10:24.339679735] quay.io/libpod/systemd-image:20230531 9984d4cfd1eb
   quay.io/libpod/testimage:20221018 f5a99120db64
   
   [11:10:24.357682655] $ /vcs/other/podman/bin/podman run -d quay.io/libpod/testimage:20221018 top
   [11:10:24.499542609] 76f963b708fe74814e8e078bbc41af84368beafaad2da932dc4ee7c3d6629409
   
   [11:10:24.512104004] $ /vcs/other/podman/bin/podman exec -i 76f963b708fe74814e8e078bbc41af84368beafaad2da932dc4ee7c3d6629409 cat
   [11:10:24.648299147] of9AOlHs62yCSNsgYav5
   1500+0 records in
   1500+0 records out
   1536000 bytes (1,5 MB, 1,5 MiB) copied, 0,00696935 s, 220 MB/s
   
   [11:10:24.670650275] $ /vcs/other/podman/bin/podman exec -i 76f963b708fe74814e8e078bbc41af84368beafaad2da932dc4ee7c3d6629409 dd of=/tmp/bigfile bs=512
   [11:10:24.768570699] 2999+1 records in
   2999+1 records out
   #/vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
   #|     FAIL: dd: number of records in
   #| expected: '3000+0 records in'
   #|   actual: '2999+1 records in'
   #\^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   # [teardown]
   
   [11:10:24.785169774] $ /vcs/other/podman/bin/podman pod rm -t 0 --all --force --ignore
   
   [11:10:24.810581406] $ /vcs/other/podman/bin/podman rm -t 0 --all --force --ignore
   [11:10:24.919517158] 76f963b708fe74814e8e078bbc41af84368beafaad2da932dc4ee7c3d6629409
   
   [11:10:24.921240846] $ /vcs/other/podman/bin/podman network prune --force
   
   [11:10:24.947890495] $ /vcs/other/podman/bin/podman volume rm -a -f

However, this also fails on conmon master, and I bisected it to commit 3605a36

[alexlarsson]

Pushed a nop-change rebase to see if the cio test works now, as that seems unrelated.

[alexlarsson]

I created #470 for the podman test issue.

[eriksjolund]

If you modify the end of the PR description from

This fixes https://github.com/containers/conmon/issues/461 

to something like this

fixes https://github.com/containers/conmon/issues/461
fixes https://github.com/containers/conmon/issues/311

there will be web links to both issues. (I think changing the git commit message does not have any effect on the web links)

[giuseppe]

@haircommander could you please cut a new release?