On 03/15/2017 12:37 AM, Marc Chamberlin wrote:
On 03/14/2017 08:27 PM, Carlos E. R. wrote:
On 2017-03-15 04:05, David C. Rankin wrote:
On 03/14/2017 09:46 PM, Carlos E. R. wrote:
One way to have both events so near is noise around the trip point. No hysteresis cycle designed in the detection. It would cause a storm of events, and in fact, there is such:
2017-03-14T11:11:33.429398-07:00 marcslaptop mcelog[816]: mcelog: Too many trigger children running already Hmm...,
I don't claim to understand the silicon level, but if the trigger is fluctuating that wildly, it smells like a voltage problem at the chip. Power supply issue? The log doesn't mention the temperature it reads, so we can't know if it fluctuates wildly. Perhaps those details are in the missing email.
Interesting stuff Carlos, David... I am following along but not sure yet what to do... But will add a couple tidbits of info related to your thoughts...
1. I rebooted my laptop into Windoz10 and let it run Boinc and a few other automatic tasks overnight. No problems with Windoz10 and laptop was running fine this morning... So if it is a hardware problem Windoz10 is keeping quiet about it.. Dunno if there is someplace specifically I should look...
2. Executed mail and mail -u root, both responded by saying there was no mail....
Marc...
Marc, Open top, or htop or the like and look for runaway processes continually pegging your CPU at 100+% to eliminate it being an actual heating problem. (even though you should be able to run at 100% continually without issue, as cooling degrades due to build-up of dust that itself can push temps over the limit. Again checking/cleaning the fan screens and cooling passages is a cheap quick test/fix if dealing with laptop overtemp issues (if you can peg the fan with a opened paperclip or thumbtack inserted through the a grill opening, you can use a shopvac to suck out the cooling passages and usually remove a bulk of the duff -- but don't use the shopvac directly over the cooling passage unless you peg the fan, you can wildly overspeed the fan causing it to come apart) The other thought is to check any BIOS temperature limit settings (usually under the System Status menu (or Power or Temperature or something similar) Some BIOS allow you to set limits in 5 degree increments over a 50-80 degree range appropriate for your processor. The reason I also suggested a power issue is if you are getting fluctuating bus voltages that can cause havoc with the processor reporting as well. That about exhausts my "what to check for list" without knowing more. When did this first start, and were there any changes you made to suse before it began? What do you have running (other than kde/konsole) when the errors occur. Does it seem like the messages are in response to something you were doing at the time? Was you mail reader running, and how many mail accounts do you have it pulling from?, etc..., etc..., etc..... Errors on Win not on Linux or vice-versa are not really that unexpected. Booth do better/worse in some areas, and it's possible it is happening on windoze, but it's just not sophisticated enough to warn/e-mail you about the temperature excursions. As Carlos said, post whatever additional information you have that might help us help you narrow it down :) -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org