View Issue Details

IDProjectCategoryView StatusLast Update
0019791MMW 5Otherpublic2023-09-28 17:54
Reporterrusty Assigned To 
PriorityurgentSeveritycrashReproducibilityrandom
Status closedResolutionfixed 
Product Version5.0.4 
Target Version5.1Fixed in Version5.1 
Summary0019791: MMW5 crashes while running in background (A3FB7F44)
DescriptionI don't know what caused this, but MMW5 crashed while running in the background (not sure if it occurred before or after the machine resumed from sleep). The UI was in the Music [Browser] view and no tracks were playing.

Crashlog: A3FB7F44

After the crash, the MMW5 UI is no longer responsive and must be force closed.
TagsNo tags attached.
Fixed in build2800

Relationships

related to 0019008 closedmichal MMW 5 Wake from sleep 2x causes Taskbar to "progress" indefinitely and Playback to start by itself 
related to 0015000 closedrusty MMW v4 Sync: Device scan do not find MMW server 
related to 0019063 closedrusty MMW 5 Crash when resuming from sleep on systems using 'Modern standby' 

Activities

Ludek

2023-02-06 13:23

developer   ~0071099

Last edited: 2023-02-06 13:52

A3FB7F44 is freeze log where main UI thread frozen after 12 hours of running.
Last callstack on the UI thread was

Callstack: Script: file:///helpers/mediaSync.js ; Func: _registerNewProfiles ; Row: 2446 ; Col: 13
Script: file:///helpers/mediaSync.js ; Func: eval ; Row: 2453 ; Col: 18

This happens once list of available servers changes (a media server is added or removed) and MM5 checks whether there isn't MMS server on the network to register it as sync profile.
The question is why it has frozen there, I'll need to add more debug strings for the next version...

EDIT: Analyzing the log further the strange thing is that r_pc_wakeup message was actually received before the PC wake up (at 40028.892s ) , but the PC woke up 3000 seconds later at 43264.048s
EDIT2: The servers were enumerated because of 0015000 -- i.e. after PC wake up UPnP service is restarted as workaround for issue 0015000

Ludek

2023-02-06 13:46

developer   ~0071100

image.png (66,845 bytes)   
image.png (66,845 bytes)   

Ludek

2023-02-06 13:50

developer   ~0071101

Last edited: 2023-02-06 13:59

Based on further analyzing it seems related to 0019008 / 0019063 where DisableAntiFreeze; on the line 39643 was called once _after_ the wake up and not before the wakeup.
See the screenshot from AppInitCode.pas above.

This can be seen from debug log messages including "lastPowerState" in the DbgView part of the log A3FB7F44
Thus Eureka was not deactivated and falsely detected the sleep as "freeze" and generated the freeze log..

michal

2023-02-06 15:09

developer   ~0071102

Last edited: 2023-02-06 15:10

We catch suspend event and then disable antifreeze. But in this case, the suspend event arrived _after_ resume event. Currently I have no idea, how to handle this if even possible, as it seems Windows paused thread, which handles suspend event, before the suspend event arrived to this thread (!). So we could not do anything before sleep/suspend.

rusty

2023-02-06 15:57

administrator   ~0071103

Last edited: 2023-02-06 15:59

fyi, this morning I observed crashlog A3FB0414 upon resuming from sleep.

What's really strange is that I had never observed this issue on this laptop until recently. The only thing that is different recently is that this machine only had MM in portable mode until last week. Last week, I installed and ran MM in 'regular' mode on this machine in order to test MicroPlayer functionality/issues on Windows 11. The issue is occurring since then even though I'm running MM in portable mode.

Could it be that the regular installation and/or the fact that I'd tried to enable the MicroPlayer on Windows 11 be causing this?

michal

2023-02-13 19:12

developer   ~0071152

Last edited: 2023-02-13 19:12

Hopefully fixed in build 2800. I cannot reproduce, as it is related to Standby mode S0 and I do not have such device. But I tried to add some handling based on the log, which could help.

rusty

2023-02-19 17:41

administrator   ~0071175

fyi: I just experienced crash A3FB8314 on resume from sleep on 5.0.4. I realize that the issue may be fixed, but I'm noting it here in case it provides any additional clues.

michal

2023-06-21 15:33

developer   ~0072261

Cannot be reproduced already, resolving.

peke

2023-09-28 17:54

developer   ~0072939

Unable to replicate in 2816 too. Closing