[EN] Vostro 3700 - Random freezing when on battery (SOLVED)

[dlalias2k1]

So I've had my laptop freeze a few times randomly when on battery as opposed to outlet power. At first it just seemed totally random, but in experimenting with it it does only seem to happen when I'm on battery. I'm not sure how to begin troubleshooting the problem, I assume it has something to do with some sort of power daemon.

Thoughts?

Thanks!

[dlalias2k1]

Found a thread on KDE.org
Guy running KDE with the exact same model is experiencing the same problem with multiple distros.

http://forum.kde.org/viewtopic.php?f=22&t=101781&start=15

He claims through trial and error that he has determined it's a cooling issue and that the problem was resolved by installing and configuring thinkfan.

I'm gonna try it and see if it works.

I'll update if it works.

[der_bud]

Hi dlalias2k1, you wrote that you have problems with "your laptop" an that another one has the same issues with "exact same model".

Perhaps no answers here because you didn't tell which model of laptop you have ;-). Always a good idea is to post the output of inxi -v3 as a starting point, which shows some basic info to start with troubleshooting.

I use thinkfan myself on my Thinkpad T60, if it works for you would be nice to post your config.

[dlalias2k1]

Yea, sorry about that, I was in a rush to post before it froze and i left out pertinent details. It's a Dell Vostro 3700.

System:    Host: mwp Kernel: 3.8-5.towo-siduction-amd64 x86_64 (64 bit, gcc: 4.7.2)
           Desktop: KDE 4.8.4 (Qt 4.8.2) Distro: siduction 12.2.0 Riders on the Storm - kde - (201212092131)
Machine:   Mobo: Dell model: 0R6NH5 version: A10 serial: .1LKF3N1.CN7016612F000N. Bios: Dell version: A10 date: 10/25/2010
CPU:       Quad core Intel Core i7 CPU Q 740 (-HT-MCP-) cache: 6144 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 13832.1
           Clock Speeds: 1: 931.00 MHz 2: 931.00 MHz 3: 931.00 MHz 4: 931.00 MHz 5: 1730.00 MHz 6: 931.00 MHz 7: 931.00 MHz 8: 931.00 MHz
Graphics:  Card: NVIDIA GT216 [GeForce GT 330M] bus-ID: 02:00.0
           X.org: 1.12.4 driver: nvidia tty size: 183x46 Advanced Data: N/A for root
Network:   Card-1: Realtek RTL8111/8168B PCI Express Gigabit Ethernet controller
           driver: r8169 ver: 2.3LK-NAPI port: d000 bus-ID: 13:00.0
           IF: eth0 state: down mac: 78:2b:cb:cb:06:5d
           Card-2: Broadcom BCM4313 802.11b/g/n Wireless LAN Controller driver: bcma-pci-bridge bus-ID: 12:00.0
           IF: wlan0 state: up mac: 68:a3:c4:10:1c:59
Drives:    HDD Total Size: 500.1GB (28.5% used) 1: model: SAMSUNG_HM500JJ
Info:      Processes: 194 Uptime: 10 min Memory: 681.7/5910.7MB Runlevel: 5 Gcc sys: 4.7.2
           Client: Shell (bash 4.2.45) inxi: 1.8.45


free:
             total       used       free     shared    buffers     cached                                                                                                              
Mem:       6052552    1195432    4857120          0      40848     463232
-/+ buffers/cache:     691352    5361200
Swap:      7340028          0    7340028


I've identified two separate issues. One is random freezing on battery , this usually occurs shortly after switching to battery from the power supply. When starting the machine up from a cold boot, I haven't yet had the laptop lockup if it just stays on battery the whole time. So I'm unsure if it's a power daemon issue, or a temperature issue.

The other issue is that attemping Hibernate seems to create a Kernel panic. The machine goes black and the two of the Lock buttons flash together until a force the machine off.

Haven't had a chance to configure thinkfan, but i'm trying that next.

Thanks!

[DeepDayze]

Sure you have enough space in your swap for hibernating? Hibernation dumps all memory state to the swap file and your swap needs to be large enough to hold all of RAM plus extra space for the control data so in theory 1.5x RAM is a good rule of thumb for swap partition size (oh yes that's Windows-think )

[dlalias2k1]

resized my swap to 10.5 gigs. Still getting the kernel panic on hibernate.

             total       used       free     shared    buffers     cached
Mem:       6052552    1015304    5037248          0      41096     332384
-/+ buffers/cache:     641824    5410728
Swap:     10412028          0   10412028

trying to research thinkfan at the moment.

[dlalias2k1]

Thinkfan didn't seem to make a difference, I think the fans were operately properly without it, so i got rid of it..

In doing a general search about this problem, some distros are reporting "pm-utils" and "Laptop-mode-tools" can conflict and cause such hangs.
Somehow both are on my machine. Is there a known issue on siduction?

[dibl]

The KDE forum poster said it was a thermal issue, but have you checked your CPU temp on the Vostro -- can you confirm a thermal problem?

Did you originally install with the nouveau driver?  Were the same problems present with that driver?  If you didn't give it a good test, you might want to try it with that driver for awhile.

Check Debian forums for anything like this -- your siduction system is Debian.

I have two Dell laptops, but they're not Vostros.  They run trouble free.

[dlalias2k1]

I'm not a master of troubleshooting Thermal issues. But for comparison, I went ahead and booted windows and ran a hearty game.

My i7 Core 740QM
HD Temp is 100F
Claims it's Maximum operating threshold is: 212 F
Idling my cores are around 125 F (120, 120, 125, 127)
In a Heavy Game session for extended periods my cores max out around 182 (174,180,189,185)

That's all under windows, No crashing.

Idling under SiductionKDE on power supply
My Cores average around 130 F  (129.2, 130.1, 125.6, 127.7, 131)
Heavy Game session with same game under wine (182, 184.2, 187, 181)

Waiting for the machine to go back to idle temp before I try it all on battery.

[dibl]

130F = 54C, which is not dangerous, but it is warmer than I would expect for normal idling.

I would advise you to remove the nvidia driver and install the nouveau driver, and try it that way for a few days.  Unfortunately the instructions are still in German (I should do a translation), but the code to remove Nvidia and then install xserver-xorg-video-nouveau is pretty obvious -- follow the code in the box, and then reboot.  Do all of this in a tty console in init 3.

[dlalias2k1]

Yea, with thinkfan on it would idle at 118F roughly. But even with thinkfan on, the random freezes still happened.

As to my earlier mention of other distros. I think I found the " "pm-utils" and "Laptop-mode-tools" can conflict and cause such hangs." on a LinuxMintDebianEdition forum which is based on debian testing, so that's why I asked if it was a known issue in Siduction as well.

I tried turning the powermizer to max performance instead of adaptive to see if maybe the battery switch was triggering the issue on battery.

I'm gonna try to disable power management first, cause I'd like to keep the Nvidia driver for wine gaming if I can. If that doesn't have an impact, then I'll revert to Nouveau and see if the issue is resolved.

Thanks for everyones help, by the way.

[dibl]

Here is a Dell Latitude D620 running the proprietary Nvidia driver -- as you can see, it has pm-utils installed but not laptop-mode-tools.  It has been running very nicely for almost a year now -- my wife uses it daily.  It gets pretty warm to hold on one's lap, up to 55C or so (probably needs the fan cleaned), so she got a fan-cooled USB pad to hold it on.

Code: [Select]

root@latitude-D620:/home/susan# apt-cache policy pm-utils
pm-utils:
  Installed: 1.4.1-9
  Candidate: 1.4.1-9
  Version table:
 *** 1.4.1-9 0
        500 http://http.debian.net/debian/ unstable/main amd64 Packages
        100 /var/lib/dpkg/status
root@latitude-D620:/home/susan# apt-cache policy laptop-mode-tools
laptop-mode-tools:
  Installed: (none)
  Candidate: 1.61-2
  Version table:
     1.63-1 0
          1 http://http.debian.net/debian/ experimental/main amd64 Packages
     1.61-2 0
        500 http://http.debian.net/debian/ unstable/main amd64 Packages

root@latitude-D620:/home/susan# inxi -v3
System:    Host: latitude-D620 Kernel: 3.8-6.towo-siduction-amd64 x86_64 (64 bit, gcc: 4.7.2)
           Desktop: KDE 4.10.2 (Qt 4.8.2) Distro: siduction 12.1.1 Desperado Reloaded - kde - (201206241901)
Machine:   System: Dell product: Latitude D620 serial: 587F1D1
           Mobo: Dell model: 0KX350 serial: .587F1D1.CN1296175N2438. Bios: Dell version: A10 date: 05/16/2008
CPU:       Dual core Intel Core2 CPU T7200 (-MCP-) cache: 4096 KB flags: (lm nx sse sse2 sse3 ssse3 vmx) bmips: 7989.14
           Clock Speeds: 1: 2000.00 MHz 2: 1000.00 MHz
Graphics:  Card: NVIDIA G72M [Quadro NVS 110M/GeForce Go 7300] bus-ID: 01:00.0
           X.org: 1.12.4 driver: nvidia tty size: 80x39 Advanced Data: N/A for root
Network:   Card-1: Broadcom NetXtreme BCM5752 Gigabit Ethernet PCI Express driver: tg3 ver: 3.128 bus-ID: 09:00.0
           IF: eth0 state: down mac: 00:18:8b:d6:2c:8f
           Card-2: Intel PRO/Wireless 3945ABG [Golan] Network Connection driver: iwl3945 ver: in-tree:s bus-ID: 0c:00.0
           IF: wlan0 state: up mac: 00:1b:77:37:e7:b5
Drives:    HDD Total Size: 100.0GB (5.4% used) 1: model: Hitachi_HTS72101
Info:      Processes: 158 Uptime: 2 min Memory: 596.4/3267.4MB Runlevel: 5 Gcc sys: 4.7.2
           Client: Shell (bash 4.2.45) inxi: 1.8.45

root@latitude-D620:/home/susan# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1:        +37.5°C  (crit = +126.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0:       +37.0°C  (high = +100.0°C, crit = +100.0°C)
Core 1:       +37.0°C  (high = +100.0°C, crit = +100.0°C)


[dlalias2k1]

Well, Reverted to nouveau and the laptop still freezes on battery. I'm not sure what the next step is.

[dibl]

One question in my mind is whether it is the entire computer, or only X that is freezing. One thing you could try is to open a terminal and run top or htop, and see whether that continues to update while X is frozen.

[dlalias2k1]

switching to console via crtl-altf2 thru f6 fails to respond, I just sit at the frozen screen. I've even tried the reisub hotkeys to see if those would accomplish anything, that fails to respond as well.

caps locks and other lock keys don't respond