Try replacing the trackpad or the magsafe / usb board, I couldn't believe my issues were coming from an issue with the magsafe / usb, I really thought it was the logic board.
After my magsafe / usb board was replaced i've been kernel panic free.
The technician said this problem usually occurs from a failing trackpad connection.
I started having this problem a few months after purchasing my 15" MacBook Pro w/Retina. It was extremely frustrating and the only way to temporarly fix the issue was to do a forced shutdown (hold down the power button) of my machine. I would see this issue happen every few weeks.
Then, all of a sudden, the problem stopped. I did nothing on my end. I have no idea what caused it, but I am just glad that it stopped.
A POSSIBLE SOLUTION:
I have a Macbook Air from late 2010, running Leopard 10.6.8. I had the issue of Kernel_Task running wild (>150%), fan running between 6000 and 6500 rpms for long periods, but no heat issue, and the system was, obviously, lagging like crazy. Looking at the console told me that there was an error occurring several times a second — bad access and sigsegv — but I do not know enough about this to really understand just from that. I tried everything I found from this thread and from googling, but I was hesitant to remove any kexts or plist files, so I held off. I'm glad I did.
I was also all set to take apart my MBA, disconnect the battery, and try and clean out the fan, when I ended up doing this:
1) I downloaded smcFanControl
2) Ran the app, found that my machine was at about 90 deg F, fan running like I said earlier. Setting the fan to 2000 rpms did nothing to alter the fan activity.
3) I turned off my machine and did yet another smc reset.
4) After booting up, I ran smcFanControl again, and again set my fan to 2000 rpms (lowest possible setting).
5) This did the trick! My machine has not had the same issue (yet).
The conclusion I draw from this is that this error is NOT a fault temp sensor, but rather an issue with the software. Likely reinstalling OSX would solve the problem. Also, this is not a fix — just a way of making your machine more usable. Thus far I have not had any issues with overheating.
Because of this, I do not recommend removing any kexts or plist files, especially if you do anything requiring a lot of CPU power on your machine. You risk damage to the CPU. At least this way, you can manually up the fan control if your machine does begin to overheat. Removing kexts or plists on your machine could make it difficult to fix in the future and might not even make the machine usable in the meantime.
As I was wisely told, taking apart your machine opens it up to all kinds of issues and can cause many problems and solve none. It is also difficult to find the proper screwdriver for removing the screws from a MBA (pentalobe head, can be bought for like $15 from iFixit if you really need to do this). Using an inappropriate tool to pull the screws means you risk stripping the screwheads and therefore making things even more difficult for yourself in the long run.
- DO NOT remove any kext or plist unless you've tried the steps I listed above
- DO NOT take apart your MBA unless you have the right tools and have taken apart delicate machines before
- DO NOT just send it off for the 'geniuses' to fix, because they will likely just stick another logic board in it. If you are like me, you are past warranty and the cost of a new LB is half the cost of a MBA.
- DO run smcFanControl and do an SMC reset and try running smcFanControl again. It should work.
- DO run a hardware test. I didn't find anything and I ran it twice, but you never know.
- DO try disabling all third-party extensions, as I wonder if updates from different applications mess with OSX (I saw people solved this issue by getting rid of DropBox, etc).
SOLUTION! Owner of the late 2013 macbook pro 13" - suffered from the Kernal Task countless times & by random i have found the solution. Place a password on wake up from ''sleep''. Boom probem solved.. As my first ever post on apple suuport community i needed to be sure on what i am saying and to provide a real solution backed up with evidence. Here it goes: 2 out of 3 times that i put my laptop to sleep, or if it goes on sleep by its own due to inactivity - once i wake it up - kernal task hits >100% CPU usuage. Now though, after enabling the password upon wake up, i have for 2 weeks straight and again i repeat for 14 days = countless wake ups from ''sleep' - i have never, not even once experienced the kernal task running north of 2% after wake up, or at any other time. Please try doing so and let me know of your results. Peace & pass-'the'-word!