Tuesday, January 12, 2016

"Critical process died": More misadventures with Windows 10


MY HP Envy, after Windows 10 installation in September 2015, still has some of the intermittent stability problems that I had with Windows 8.1.  The Geek Squad installed W10, and had to redo it as a fresh install (without affecting the data) because Windows File Explorer “Quick Access” would not populate thumbnails properly.

The pattern is still that if I start Windows 10 by simply powering up, it sometimes will freeze within a few minutes.  Hitting the power button always recovers the machine (Chrome resumes where it was, and Office goes back to the last copy I had saved).   Restarting the machine makes it more stable.

But sometimes, after working a long time, Windows will interrupt me with a forced restart over a “Critical Process Died” and then throw the blue “hard disk error”.  The power button always reboots the machine.  If I get two such crashes in a short time, the machine is usually stable for a long time after one more restart.



The HP disk diagnosis tool apparently is not available to Windows 10, as these screenshots show.

But on a couple of occasions, HP has run a disk check (scandisk) before starting the machine normally.  That would take around 5 minutes.  It would repair a sector at around 8%, and then again between 48-50%.

Microsoft doesn’t seem to have a clean explanation.  The forums are all over the map, for example.

It is possible that a hard disk error did cause the critical process to die.  But HP’s own auto disk repair, which has run a couple of times, ought to have fixed it.

It’s not practical to rebuild machines and operating systems at home all the time.  There is too much to do.  It’s also irresponsible for Microsoft to expect customers to do “free” upgrades that are likely to be unstable, when consumers have everyday work to do.

I do use Carbonite for backups (and check it).  I also make manual backups (on USB thumbs) of the most important files (several different copies), usually leave at least one in a safe deposit box, and also load the important files to another machine (not networked) with W8 that also has its own carbonite.  And I do have a few Seagate backups lying around.

Below are some typical errors that I see on my own event log.

“The User Data Access_1b440ca service terminated unexpectedly.  It has done this 1 time(s).  The following corrective action will be taken in 10000 milliseconds: Restart the service.
“The application-specific permission settings do not grant Local Activation permission for the COM Server application with CLSID
{D63B10C5-BB46-4990-A94F-E40B9D520160}
 and APPID
{9CA88EE3-ACB7-47C8-AFC4-AB702511C276}
 to the user NT AUTHORITY\SYSTEM SID (S-1-5-18) from address LocalHost (Using LRPC) running in the application container Unavailable SID (Unavailable). This security permission can be modified using the Component Services administrative tool.

“The User Data Access_47e87 service terminated unexpectedly.  It has done this 1 time(s).  The following corrective action will be taken in 10000 milliseconds: Restart the service.”

Another machine, a Toshiba satellite, is still on Windows 8.1, and will sometimes throw a forced restart blue screen (less frequently), but never gets the Hard Disk Error.  
But ever since Windows 8, Microsoft has simply gotten less stable for home users.


Update: Jan 18, 2016

After a couple days of stabililty, I got an "unexpected store exception" right after a fast start, and a fake hard disk error.  A push button start worked as usual.  Here's a Blogger (Kunmi's Space) post on it. 

1 comment:

Steve Madden said...

Good Day I Am So Happy I Found Your Website, I
Found You By Mistake, While I Was Searching On Aol For
Something Else, Anyhow I Am Here Now And Would Just Like To Say Many Thanks For
A Remarkable Post And An All Round Entertaining Blog
I Don’t Have Time To Read It, All At The Moment But I Have BookMarked It
And Also Added In Your RSS Feeds, So When I Have Time I Will Be Back To Read More,
Please Do Keep Up The Fantastic Job. You can also check; CRITICAL_PROCESS_DIED