#11318 NORM 11.3.0: Ad-hoc network is not brought back after idle suspend/resume

[Zarro Boogs per Child]

#11318: Ad-hoc network is not brought back after idle suspend/resume
------------------------------------+---------------------------------------
           Reporter:  erikos        |       Owner:               
               Type:  defect        |      Status:  new          
           Priority:  normal        |   Milestone:  11.3.0       
          Component:  not assigned  |     Version:  not specified
         Resolution:                |    Keywords:               
        Next_action:  diagnose      |    Verified:  0            
Deployment_affected:                |   Blockedby:               
           Blocking:                |  
------------------------------------+---------------------------------------

Comment(by dsd):

 Reproduced, the problem is that the connection failed due to suspending
 while it was being established. NM then marked the connection as broken
 and doesn't try to reconnect.

 powerd trace says:

 {{{
 : @ 1317978712 got event: useridle1, 1317978712, , , .
 : @ 1317978712 general inhibit: 0
 : @ 1317978717 snoozing: until_dim 14400 165 240
 : @ 1317978717 until-sleep_type is until_dim-soft
 powerd-dbus-Message: Device 0x86c9800 state changed: 4
 powerd-dbus-Message: device 0x86c9800 state 4 inhibits suspend
 powerd-dbus-Message: powerd_send_event inhibit-suspend
 : @ 1317978718 s_event-type is inhibit-suspend-soft
 powerd-dbus-Message: Device 0x86c9800 state changed: 5
 powerd-dbus-Message: device 0x86c9800 state 5 inhibits suspend
 powerd-dbus-Message: Device 0x86c9800 state changed: 7
 powerd-dbus-Message: device 0x86c9800 state 7 inhibits suspend
 : @ 1317978724 got wakeup: batterystatechange @ 1317978724, slept 5
 : @ 1317978725 until-sleep_type is until_dim-soft
 : @ 1317978797 got wakeup: keypress @ 1317978797, slept 71
 }}}

 I don't quite understand how this is possible - after the "snooze" message
 somehow enough time passed for 3 NM state changes, including successful
 delivery and reception of the "inhibit suspend" message, before the system
 went to sleep. Is powerd multi threaded?

-- 
Ticket URL: <http://dev.laptop.org/ticket/11318#comment:3>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system