It is currently March 28th, 2024, 5:47 pm

4.5.16-3687 crashing, faulting module Rainmeter.dll

Report bugs with the Rainmeter application and suggest features.
User avatar
nmdelrio
Posts: 74
Joined: June 28th, 2015, 12:51 pm
Location: Paranaque City, Philippines

4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by nmdelrio »

I updated Rainmeter from 4.5.14-3663 to 4.5.16-3587. Update was automatic after being prompted by the application on startup.

Rainmeter-4.5.16-3687 crashing intermittently, twice or more daily. Having a 5-monitor setup, it sometimes resets my monitor configuration. Cold restarting the PC fixes the problem, but Rainmeter crashes after some time.

I checked Event Viewer and found this ...

Code: Select all

Faulting application name: Rainmeter.exe, version: 4.5.16.3687, time stamp: 0x632a289e
Faulting module name: Rainmeter.dll, version: 4.5.16.3687, time stamp: 0x632a289c
Exception code: 0xc0000005
Fault offset: 0x000000000013351d
Faulting process id: 0x4b04
Faulting application start time: 0x01d8d4e1b178f83f
Faulting application path: C:\Program Files\Rainmeter\Rainmeter.exe
Faulting module path: C:\Program Files\Rainmeter\Rainmeter.dll
Report Id: db6d2875-69e2-41f8-99e5-0c59e303591b
Faulting package full name: 
Faulting package-relative application ID:
I since then reverted to 4.5.14-3663, which had no problems with my loads of self-made skins. Rainmeter hasn't crashed since.

Any ideas?

All is good for me as there is no more crashing. Just wanted to share the information.
User avatar
Brian
Developer
Posts: 2673
Joined: November 24th, 2011, 1:42 am
Location: Utah

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by Brian »

Could you post your version information?
About dialog -> Version tab -> Click "Copy to Clipboard"
https://docs.rainmeter.net/manual/user-interface/about/#VersionTab
nmdelrio wrote: October 8th, 2022, 12:01 am ... self-made skins.
Could you narrow down which skin is causing the crash, and then post the skin code here? You may need to include any external files like images, scripts, etc. You can create a .rmskin file if needed. If you are more advanced at debugging problematic skin code, please post just the necessary code to reproduce your issue - if not, that is okay - just post what you can.

If you are interacting with a skin or Rainmeter itself (like the dialogs or context menus), post what steps you are doing to cause the crash.

Lastly, if you could turn on logging and "debug mode". Restart Rainmeter and once the crash happens, post your Rainmeter.log file?
https://docs.rainmeter.net/manual/user-interface/manage/#SettingsTab

-Brian
User avatar
nmdelrio
Posts: 74
Joined: June 28th, 2015, 12:51 pm
Location: Paranaque City, Philippines

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by nmdelrio »

HI Brian,

So sorry for not replying sooner, not my intention, but strangely I did not get an email notification of replies, forgot to check the Notify me when a reply is posted checkbox, been a while since I posted in the forums, anyway...

I have loads of skins (18 total, yes I counted :D ), mostly made by me, and some with scrips 'borrowed' from other author's skins, same of my self-made skins and been modified to suit my new hardware. Skins are mostly system monitoring, BTW.

I have since then built a new PC and new install of Rainmeter. I still have 3663 installed. Let n]me try to install the latest version and see if it still crashes and get back to you. For all I know it could be my old system that's causing the crash.

Meanwhile, here's my desktop with 18 skins spread over 5 monitors. Sorry, just can't live without Rainmeter.
Attachments
rain-desktop-resized.png
User avatar
nmdelrio
Posts: 74
Joined: June 28th, 2015, 12:51 pm
Location: Paranaque City, Philippines

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by nmdelrio »

Just an update, Rainmeter stopped crashing, it mush have been something in my Windows 10 installation. After doing a clean re-install of Windows 10 and all my applications, Rainmeter haven't crashed since. :thumbup:
User avatar
rlnoonan
Posts: 26
Joined: July 30th, 2018, 10:37 pm

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by rlnoonan »

I've been using 4.5.16.3687 for some time with no issues until about a week and a half ago when it began crashing in an almost identical fashion to what the OP described. After some amount of time, always when I'm not at the computer and often when nothing is running, rainmeter crashes. I can start it back up with no problems. Here is the event view information from each crash:

Code: Select all

Faulting application name: Rainmeter.exe, version: 4.5.16.3687, time stamp: 0x632a289e
Faulting module name: Rainmeter.dll, version: 4.5.16.3687, time stamp: 0x632a289c
Exception code: 0xc0000005
Fault offset: 0x00000000000df8e0
Faulting process id: 0x2ba8
Faulting application start time: 0x01d8f87d144c2daa
Faulting application path: C:\Program Files\Rainmeter\Rainmeter.exe
Faulting module path: C:\Program Files\Rainmeter\Rainmeter.dll
Report Id: dd9dec3a-68f4-4af1-bab5-0b8f0a8fc765
Faulting package full name: 
Faulting package-relative application ID:
Here is the version information from Rainmeter:

Code: Select all

Rainmeter 4.5.16.3687 (64-bit)
Language: English (1033)
Build time: 2022-09-20 16:53:41
Windows 10 Home 21H2 (build 19044.2251) 64-bit - English (1033)
Path: C:\Program Files\Rainmeter\
SkinPath: C:\Users\rlnoo\OneDrive\Documents\Rainmeter\Skins\
SettingsPath: C:\Users\rlnoo\AppData\Roaming\Rainmeter\
IniFile: C:\Users\rlnoo\AppData\Roaming\Rainmeter\Rainmeter.ini
I don't have any self-made skins and have made no recent changes to what skins are running, so I don't think it is related to a particular skin. However, right when this issue began occurring I did upgrade my monitor to a 4K monitor. So this likely changed my video driver (Windows automatically changed configuration to use 4K resolution) and I have changed from using my VGA output to now using my DisplayPort output. I'm sure this is somehow related, since the crashing seemed to start immediately after this. I just can't imagine how it is related.

I'm going to follow Brian's advice and turn on logging and "debug mode". I'll update with that information once I get it. In the meantime, if there is anything else I should do in order to collect helpful information, please let me know. Thanks.
User avatar
SilverAzide
Rainmeter Sage
Posts: 2588
Joined: March 23rd, 2015, 5:26 pm
Contact:

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by SilverAzide »

Huh... I guess I've seen this too, but didn't bother to report it. In my case, I got a new Win 11 PC where the power settings were set to turn off the monitor after some period of inactivity. When I'd come back to the PC and wake it up, Rainmeter would be dead and gone. No error window or anything. It was not something I could reproduce easily, so I let it slide. (The Rainmeter debug log just ends, no errors noted.)

At some point later I changed my power settings to set the monitor to never sleep, and -- poof -- no more crashing.

Could something in Windows be telling Rainmeter there are ZERO active display devices and causing it to crash?
User avatar
nmdelrio
Posts: 74
Joined: June 28th, 2015, 12:51 pm
Location: Paranaque City, Philippines

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by nmdelrio »

Just some additional info... before I re-installed and change hardware (motherboard, CPU, RAM), I also noticed my monitor configuration randomly changing (i.e. primary display being changed, messing up my wallpapers on all 5 monitors, messing up Rainmeter skins.) It's like when you do a new clean install of Windows when you have to tell Windows which is primary, and where each monitor goes to which position. I also don't know if this has something to do with Rainmeter (I don't see a reason Rainmeter resetting monitor configuration), but it definitelky crashes Rainmeter when it happens, which I would understand.
User avatar
rlnoonan
Posts: 26
Joined: July 30th, 2018, 10:37 pm

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by rlnoonan »

Interesting, I think there is something to the idea that Windows display is changing and that is causing the crash. Rainmeter stayed up all night and didn't crash until this morning when I was using it again. The log shows everything going along smoothly, with the weather skin regularly fetching weather information. The right when Rainmeter crashed I got this:

Code: Select all

NOTE (09:23:22.171) : System: Display settings changed
NOTE (09:23:22.301) : System: Display settings changed
DBUG (09:23:22.581) : ------------------------------
DBUG (09:23:22.581) : * EnumDisplayDevices / EnumDisplaySettings API
DBUG (09:23:22.582) : \\.\DISPLAY1
DBUG (09:23:22.584) :   Name     : Generic PnP Monitor
DBUG (09:23:22.584) :   Adapter  : NVIDIA Quadro K2100M
DBUG (09:23:22.585) :   Flags    : ACTIVE PRIMARY (0x00000005)
DBUG (09:23:22.585) :   Handle   : 0x0000000000010001
DBUG (09:23:22.586) :   ScrArea  : L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (09:23:22.586) :   WorkArea : L=0, T=0, R=2560, B=1400 (W=2560, H=1400)
DBUG (09:23:22.587) : \\.\DISPLAY2
DBUG (09:23:22.588) :   Adapter  : NVIDIA Quadro K2100M
DBUG (09:23:22.589) :   Flags    : (0x00000000)
DBUG (09:23:22.589) : \\.\DISPLAY3
DBUG (09:23:22.591) :   Adapter  : NVIDIA Quadro K2100M
DBUG (09:23:22.591) :   Flags    : (0x00000000)
DBUG (09:23:22.592) : \\.\DISPLAY4
DBUG (09:23:22.593) :   Adapter  : NVIDIA Quadro K2100M
DBUG (09:23:22.593) :   Flags    : (0x00000000)
DBUG (09:23:22.594) : ------------------------------
DBUG (09:23:22.594) : * EnumDisplayMonitors API
DBUG (09:23:22.595) : \\.\DISPLAY1
DBUG (09:23:22.595) :   Flags    : PRIMARY (0x00000001)
DBUG (09:23:22.596) :   Handle   : 0x0000000000010001
DBUG (09:23:22.597) :   ScrArea  : L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (09:23:22.597) :   WorkArea : L=0, T=0, R=2560, B=1400 (W=2560, H=1400)
DBUG (09:23:22.598) : ------------------------------
DBUG (09:23:22.598) : * METHOD: EnumDisplayDevices + EnumDisplaySettings Mode
DBUG (09:23:22.599) : * MONITORS: Count=4, Primary=@1
DBUG (09:23:22.599) : @0: Virtual screen
DBUG (09:23:22.599) :   L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (09:23:22.600) : @1: \\.\DISPLAY1 (active), MonitorName: Generic PnP Monitor
DBUG (09:23:22.600) :   L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (09:23:22.601) : @2: \\.\DISPLAY2 (inactive), MonitorName: 
DBUG (09:23:22.601) : @3: \\.\DISPLAY3 (inactive), MonitorName: 
DBUG (09:23:22.602) : @4: \\.\DISPLAY4 (inactive), MonitorName: 
DBUG (09:23:22.602) : ------------------------------
One thing that is important to note is I have two computers sharing my KVM via a switch. This crash occurred (and I think all other crashes) when I switched away from this computer. It doesn't crash every time I switch away, but it seems to happen on most switches. Looking at the log information, it looks like when I switched away, the video information changed to 2560x1440. My actual resolution is 3840x2160.

One other thing that is interesting is that after I turned on logging and "debug mode", I did successfully switch away and back once or twice. The log contains nothing for those successful switches. So I think that occasionally Windows sees the switch as a change in display hardware and that causes the change in display information.

Last note, I run Rainmeter on the other computer that I switch between (exact same skins, etc) and that computer never gets a crash with Rainmeter. My guess is that it has something to do with the hardware handling the switch away better. Both computers are laptops. The one that does not crash uses a port replicator and is native HDMI (KVM switch and monitor are HDMI). The one that crashes is older and uses a docking station that is native DisplayPort that I use a converter to output HDMI. I would imagine that has something to do with Windows seeing the display change when I switch away.

I'll keep logging to see if I get any further information. If there is anyone has any ideas, something I should try, or data you want me to try to gather, please let me know.

Thanks.
User avatar
rlnoonan
Posts: 26
Joined: July 30th, 2018, 10:37 pm

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by rlnoonan »

I little more info. When I restarted Rainmeter after my last post, I provided this information about the display at startup:

Code: Select all

NOTE (10:23:45.111) : Rainmeter 4.5.16.3687 (64-bit)
NOTE (10:23:45.112) : Language: English (1033)
NOTE (10:23:45.113) : Build time: 2022-09-20 16:53:41
NOTE (10:23:45.113) : Windows 10 Home 21H2 (build 19044.2251) 64-bit - English (1033)
NOTE (10:23:45.114) : Path: C:\Program Files\Rainmeter\
NOTE (10:23:45.114) : SkinPath: C:\Users\rlnoo\OneDrive\Documents\Rainmeter\Skins\
NOTE (10:23:45.115) : SettingsPath: C:\Users\rlnoo\AppData\Roaming\Rainmeter\
NOTE (10:23:45.115) : IniFile: C:\Users\rlnoo\AppData\Roaming\Rainmeter\Rainmeter.ini
DBUG (10:23:45.118) : ------------------------------
DBUG (10:23:45.119) : * EnumDisplayDevices / EnumDisplaySettings API
DBUG (10:23:45.120) : \\.\DISPLAY1
DBUG (10:23:45.122) :   Name     : Generic PnP Monitor
DBUG (10:23:45.123) :   Adapter  : NVIDIA Quadro K2100M
DBUG (10:23:45.123) :   Flags    : ACTIVE PRIMARY (0x00000005)
DBUG (10:23:45.124) :   Handle   : 0x0000000000010001
DBUG (10:23:45.125) :   ScrArea  : L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (10:23:45.125) :   WorkArea : L=0, T=0, R=2560, B=1400 (W=2560, H=1400)
DBUG (10:23:45.126) : \\.\DISPLAY2
DBUG (10:23:45.128) :   Adapter  : NVIDIA Quadro K2100M
DBUG (10:23:45.128) :   Flags    : (0x00000000)
DBUG (10:23:45.129) : \\.\DISPLAY3
DBUG (10:23:45.131) :   Adapter  : NVIDIA Quadro K2100M
DBUG (10:23:45.131) :   Flags    : (0x00000000)
DBUG (10:23:45.132) : \\.\DISPLAY4
DBUG (10:23:45.134) :   Adapter  : NVIDIA Quadro K2100M
DBUG (10:23:45.134) :   Flags    : (0x00000000)
DBUG (10:23:45.134) : ------------------------------
DBUG (10:23:45.135) : * EnumDisplayMonitors API
DBUG (10:23:45.135) : \\.\DISPLAY1
DBUG (10:23:45.136) :   Flags    : PRIMARY (0x00000001)
DBUG (10:23:45.136) :   Handle   : 0x0000000000010001
DBUG (10:23:45.137) :   ScrArea  : L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (10:23:45.137) :   WorkArea : L=0, T=0, R=2560, B=1400 (W=2560, H=1400)
DBUG (10:23:45.138) : ------------------------------
DBUG (10:23:45.138) : * METHOD: EnumDisplayDevices + EnumDisplaySettings Mode
DBUG (10:23:45.139) : * MONITORS: Count=4, Primary=@1
DBUG (10:23:45.139) : @0: Virtual screen
DBUG (10:23:45.140) :   L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (10:23:45.140) : @1: \\.\DISPLAY1 (active), MonitorName: Generic PnP Monitor
DBUG (10:23:45.140) :   L=0, T=0, R=2560, B=1440 (W=2560, H=1440)
DBUG (10:23:45.141) : @2: \\.\DISPLAY2 (inactive), MonitorName: 
DBUG (10:23:45.141) : @3: \\.\DISPLAY3 (inactive), MonitorName: 
DBUG (10:23:45.142) : @4: \\.\DISPLAY4 (inactive), MonitorName: 
DBUG (10:23:45.142) : ------------------------------
Interesting that it shows 2560x1440, when I was actively using the monitor in 3840x2160. Is it possible that it holds the old video information from the last crash or something? The other thing that was interesting is that when it ended up crashing again (after the second or third switch away) it reported this:

Code: Select all

NOTE (16:16:19.335) : System: Display settings changed
NOTE (16:16:19.399) : System: Display settings changed
DBUG (16:16:19.826) : ------------------------------
DBUG (16:16:19.827) : * EnumDisplayDevices / EnumDisplaySettings API
DBUG (16:16:19.828) : \\.\DISPLAY1
DBUG (16:16:19.829) :   Name     : Generic PnP Monitor
DBUG (16:16:19.830) :   Adapter  : NVIDIA Quadro K2100M
DBUG (16:16:19.831) :   Flags    : ACTIVE PRIMARY (0x00000005)
DBUG (16:16:19.831) :   Handle   : 0x0000000000010001
DBUG (16:16:19.832) :   ScrArea  : L=0, T=0, R=3840, B=2160 (W=3840, H=2160)
DBUG (16:16:19.832) :   WorkArea : L=0, T=0, R=3840, B=2100 (W=3840, H=2100)
DBUG (16:16:19.833) : \\.\DISPLAY2
DBUG (16:16:19.834) :   Adapter  : NVIDIA Quadro K2100M
DBUG (16:16:19.835) :   Flags    : (0x00000000)
DBUG (16:16:19.836) : \\.\DISPLAY3
DBUG (16:16:19.838) :   Adapter  : NVIDIA Quadro K2100M
DBUG (16:16:19.838) :   Flags    : (0x00000000)
DBUG (16:16:19.839) : \\.\DISPLAY4
DBUG (16:16:19.841) :   Adapter  : NVIDIA Quadro K2100M
DBUG (16:16:19.841) :   Flags    : (0x00000000)
DBUG (16:16:19.842) : ------------------------------
DBUG (16:16:19.843) : * EnumDisplayMonitors API
DBUG (16:16:19.843) : \\.\DISPLAY1
DBUG (16:16:19.844) :   Flags    : PRIMARY (0x00000001)
DBUG (16:16:19.845) :   Handle   : 0x0000000000010001
DBUG (16:16:19.845) :   ScrArea  : L=0, T=0, R=3840, B=2160 (W=3840, H=2160)
DBUG (16:16:19.846) :   WorkArea : L=0, T=0, R=3840, B=2100 (W=3840, H=2100)
DBUG (16:16:19.848) : ------------------------------
DBUG (16:16:19.849) : * METHOD: EnumDisplayDevices + EnumDisplaySettings Mode
DBUG (16:16:19.849) : * MONITORS: Count=4, Primary=@1
DBUG (16:16:19.850) : @0: Virtual screen
DBUG (16:16:19.850) :   L=0, T=0, R=3840, B=2160 (W=3840, H=2160)
DBUG (16:16:19.850) : @1: \\.\DISPLAY1 (active), MonitorName: Generic PnP Monitor
DBUG (16:16:19.851) :   L=0, T=0, R=3840, B=2160 (W=3840, H=2160)
DBUG (16:16:19.851) : @2: \\.\DISPLAY2 (inactive), MonitorName: 
DBUG (16:16:19.852) : @3: \\.\DISPLAY3 (inactive), MonitorName: 
DBUG (16:16:19.853) : @4: \\.\DISPLAY4 (inactive), MonitorName: 
DBUG (16:16:19.853) : ------------------------------
Now it sees 3840x2160 correctly, but it still crashed. Not what I was expecting, but I have no knowledge of what is going on beneath the hood in Rainmeter. Can anyone make sense of this?
User avatar
Brian
Developer
Posts: 2673
Joined: November 24th, 2011, 1:42 am
Location: Utah

Re: 4.5.16-3687 crashing, faulting module Rainmeter.dll

Post by Brian »

Hmmm. The only display-related "known" bug that I know of is when your graphics driver updates, it *can* crash Rainmeter. It's usually happens if a skin is currently being redrawn while the drivers are in the middle of an update.

But, most of the crashes described here seem to be something else.

First let me address some of the questions:
SilverAzide wrote: November 15th, 2022, 1:48 am Could something in Windows be telling Rainmeter there are ZERO active display devices and causing it to crash?
I doubt this is the problem but you never know. Since you said you can't easily reproduce it (I can't either). Nothing significant has changed with monitor detection - at least on the Rainmeter end. Maybe a Windows update changed something with its monitor detection... :confused:

rlnoonan wrote: November 15th, 2022, 10:27 pm Is it possible that it holds the old video information from the last crash or something?
All that information is provided by Windows just before Rainmeter logs it. Rainmeter asks Windows for monitor information when the application starts, and if any of the display settings (or work area) changes. Rainmeter doesn't retain any of this information when Rainmeter closes, it just gets it from Windows at start up (or changes).

If those values are wrong at the time when Rainmeter logs it, then something has to be wrong on the Windows side of things.

But there is something about your logs that intrigues me...there are 2 System: Display settings changed messages back to back (literally a few milliseconds apart). Normally, when this message happens Rainmeter will get all the monitor information again (basically to update the monitor variables, and move any skins that are no longer on a monitor). Since these messages happen rapidly, there might be a timing issue where the first message is in the process of clearing the "old" monitor values and setting the new ones - and then the second message clears them again before the first message is finished.

But this is all speculation.



Could both of you revert back to Rainmeter 4.5.14.3663 and check if Rainmeter is crashing in the same manner as you described?

You will probably want to uncheck "automatic updating" first, so that Rainmeter won't update to the latest again.
https://docs.rainmeter.net/manual/user-interface/manage/#SettingsTab

I would also leave "Debug mode" on, and it would be helpful if you could just run the standard "illustro" skins to test with (to eliminate the possibility that a 3rd party skin or plugin is causing this in some way). You can use Rainmeter safe start to load the standard skins.


If Rainmeter version 4.5.14 doesn't cause any crashes, I can look for other issues that might cause crashing, but if the crashes still happen, then I fear some sort of Windows/driver update might be causing issues.

-Brian
Post Reply