qvm-prefs --set vmname qrexec_timeout does not allow value 0, and has misleading error message

[yomimono]

Qubes OS version

R4.0

Affected component(s) or functionality
qvm-prefs command in dom0

Brief summary
I want to disable the behavior of QubesOS where VMs will be terminated if they do not respond to qrexec within a certain amount of time. I can't figure out how to do this.

To Reproduce
Steps to reproduce the behavior:
_1. in dom0, attempt to set the qrexec_timeout parameter of a vm to 0 with qvm-prefs --set vmname qrexec_timeout 0. _
_2. Result is usage information for qvm-prefs and the error message qvm-prefs: error: no such property: 'qrexec_timeout'. This is inconsistent as I can see qrexec_timeout with qvm-prefs --get and can set it to a positive nonzero value.

Expected behavior
Successful update of the qrexec_timeout parameter to 0, disabling the shutdown of VMs that don't respond within n seconds.

Actual behavior
qrexec_timeout remains at the original value.

[yomimono]

/cc @linse

[marmarek]

If your VM do not have qrexec installed, to avoid killing it on qrexec timet you can disable qrexec feature. I guess you create TemplateBasedVM, but given alternative kernel, you actually do not use any of the os from its template. If you create StandaloneVM instead, you'll avoid this issue (as qrexec feature is not there by default). Otherwise, you can do that manually with qvm-features tool - set qrexec to empty string to override 1 set on template.

[yomimono]

Ah, OK. So what I want is instead qvm-features appname qrexec "".

[andrewdavidwong]

I guess this should be a doc improvement issue.

[yomimono]

I think the result of qvm-prefs --set vmname qrexec_timeout 0 is rather confusing; there should probably be an error message indicating that 0 is an invalid value, rather than qrexec_timeout.

[HW42]

The technical part should be fixed by QubesOS/qubes-core-admin-client#268 and QubesOS/qubes-core-admin#570. Leaving this open for the documentation part.

[neoniobium]

This appears the be fixed.

[andrewdavidwong]

Closing as completed. If anyone believes this issue is not yet completed, or if anyone is still affected by this issue, please leave a comment saying so, and we'll be happy to reopen it. Thank you.