Dear Maintainer,
after upgrading to Devuan Excalibur (equivalent to Debian Trixie),
pm-suspend was not running any more. Running DEBUG_PM=true pm-suspend
produced an output containing the following lines:
+ command_exists check_suspend.no_molly_guard
+ type check_suspend.no_molly_guard
+ return 1
+ log 'pm-utils does not know how to suspend.no_molly_guard on this system.'
The following change fixed it:
--- /old/pm-action 2025-11-10 17:55:09.078432257 +0100
+++ /usr/lib/pm-utils/bin/pm-action 2025-11-10 17:49:09.814631753 +0100
@@ -26,7 +26,7 @@
# The rule here? Simplicity.
export STASHNAME=pm-suspend
-export METHOD="$(echo ${0##*pm-} |tr - _)"
+export METHOD="$(echo ${0##*pm-} |tr - _|sed 's/\.no_molly_guard$//')"
. "/usr/lib/pm-utils/pm-functions"
# return code tracking for when suspend fails
Alessandro Vesely writes ("Bug#1120480: pm-utils: pm-suspend won't run after molly-guard"):
...
Hi. Thanks for the report. IMO this configuration ought to work.
I agree with you that Debian is the right place for the fix.
I confess I don't fully understand evertthing that's going on here,
but I woinder if pm-utils is the right place to fix this.
I think you have molly-guard installed?
pm-action is documented to iterate through /etc/pm/sleep.d. Does your
system have something in that directory put there by molly-guard?
This encodes specific knowledge of molly-guard into pm-utils. Would
it not be better if molly-guard somehow declared how to "suspend" it?
I'm CCing the molly-guard maintainers to see what they think.
(I see we do have some special code for the benefit of molly-guard,
since 2018, due to usrmerge. #812545. That doesn't seem relevant to
this bug.)
Regards,
Ian.
Hi, Yes, I do. I don't know what pulled it in. I didn't even know it was a separate package. I have now purged it and reinstalled pm-utils. No. I agree the patch makes no sense, given that molly-guard is an separate package. pm-utils works fine as is. This is a molly-guard bug. Perhaps they could patch pm-action in postinst, if they cannot find a better way... Best Ale
Alessandro Vesely writes ("Bug#1120480: pm-utils: pm-suspend won't run after molly-guard"):
Gosh. I definitely wouldn't want us to do that. I'm hoping the
molly-guard maintainer will be able to advise...
Ian.