event ID 4013
event ID 5781
Jun 12, 2011
Jun 11, 2011
Unsuccessful POST attempt(s)
Building your own PC is not as trivial as it seems at the first glance. Especially if you're going to be an early adopter of Intel technology.
Here is what I used recently:
The next day I wanted to tweak the BIOS to make it boot a little faster. An ordinary POST took about a minute! So I enabled all the Hyper-Boot options. That was the point where the following error appeared:
The error popped up irregulary. Most of the time I could boot after entering BIOS by choosing the 'Y' option and exiting BIOS with saving. But sometimes neither 'Y' nor 'N' worked. I had absolutely no clue where the error came from and suspected the recent BIOS changes as the source of that problem. After canceling those changes the error still persisted.
I called Intel support. The first thing the technician on the line wanted to know was the brand and the designation of the memory I used. Since the memory controller was moved into the CPU with the Core iX-Series, using the "wrong" memory became fatal. Where wrong means a too high memory voltage. I knew this, that's why I chose the KHX1600C93K3/12GX, which were tested with the 58SO2 by Intel. After checking the remaining components the technician couldn't see anything wrong. He advised me to remove all but one memory module and to see if the error went away.
I thoroughly tested the memory and couldn't find any defects. The unsuccessful POST error seemed to disappear after I removed and reinstalled the modules. But it didn't last long. Some reboots later it was there again.
Meanwhile I did some research on the internet and found some hints that the memory clock could be too high. My memory was running with 1333 MHz and 1.5V. I tried to reduce the clock, to no avail. In the options there are 3 values indicating the memory speed: Default, Proposed, Active. The only value I could change was the default one. After reducing the default value to 1066 MHz saving, shutting down and powering on again, the current speed was still shown as 1333 MHz.
Again I called Intel. This time I wanted to know how to reduce the memory speed. Guess what, the technician couldn't tell me! Instead he offered me to replace the mainboard but for the moment I declined, keeping the option to replace it later on.
I decided to replace the memory. Since the memory tested by Intel with the 58SO2 was hard to come by, except of the modules I already had, I chose the Kingston KVR1333D3N9K3/12G.
I went on with my research and found an interesting statement in the "Performance Tuning Guide for Intel® Desktop Board DP55KG and Intel® Desktop Board DP55SB Extreme Series" by Intel.
On the next day the replacement memory arrived. After exchanging it, at first it looked very promising but some reboots later the unsuccessful POST was there again.
My conclusion is that it must be some kind of memory incompatibility. After all I didn't find a solution. If you experienced similar problems, please leave me a comment with your configuration.
Here is what I used recently:
- Mainboard: Intel 58SO2
- CPU: Core i7-970
- RAM: Kingston KHX1600C93K3/12GX
- Grafics: Asus GTX560 Ti
The next day I wanted to tweak the BIOS to make it boot a little faster. An ordinary POST took about a minute! So I enabled all the Hyper-Boot options. That was the point where the following error appeared:
The error popped up irregulary. Most of the time I could boot after entering BIOS by choosing the 'Y' option and exiting BIOS with saving. But sometimes neither 'Y' nor 'N' worked. I had absolutely no clue where the error came from and suspected the recent BIOS changes as the source of that problem. After canceling those changes the error still persisted.
I called Intel support. The first thing the technician on the line wanted to know was the brand and the designation of the memory I used. Since the memory controller was moved into the CPU with the Core iX-Series, using the "wrong" memory became fatal. Where wrong means a too high memory voltage. I knew this, that's why I chose the KHX1600C93K3/12GX, which were tested with the 58SO2 by Intel. After checking the remaining components the technician couldn't see anything wrong. He advised me to remove all but one memory module and to see if the error went away.
I thoroughly tested the memory and couldn't find any defects. The unsuccessful POST error seemed to disappear after I removed and reinstalled the modules. But it didn't last long. Some reboots later it was there again.
Meanwhile I did some research on the internet and found some hints that the memory clock could be too high. My memory was running with 1333 MHz and 1.5V. I tried to reduce the clock, to no avail. In the options there are 3 values indicating the memory speed: Default, Proposed, Active. The only value I could change was the default one. After reducing the default value to 1066 MHz saving, shutting down and powering on again, the current speed was still shown as 1333 MHz.
Again I called Intel. This time I wanted to know how to reduce the memory speed. Guess what, the technician couldn't tell me! Instead he offered me to replace the mainboard but for the moment I declined, keeping the option to replace it later on.
I decided to replace the memory. Since the memory tested by Intel with the 58SO2 was hard to come by, except of the modules I already had, I chose the Kingston KVR1333D3N9K3/12G.
I went on with my research and found an interesting statement in the "Performance Tuning Guide for Intel® Desktop Board DP55KG and Intel® Desktop Board DP55SB Extreme Series" by Intel.
Figure 18 shows what happens when you push the system too far either in Memory Performance or Processor Performance.
On the next day the replacement memory arrived. After exchanging it, at first it looked very promising but some reboots later the unsuccessful POST was there again.
My conclusion is that it must be some kind of memory incompatibility. After all I didn't find a solution. If you experienced similar problems, please leave me a comment with your configuration.
Deleting your Facebook Account
This has been described countless times already. Nevertheless, since it is so well hidden here we go, one more time:
Congratulations, you have successfully deleted your Facebook Account. At least Mark says so ..
If you think that this instruction was helpful, don't forget to
- Log in with your Facebook Username
- Follow this Link
- Now you should be presented with an Account Deletion Dialog. Confirm it.
- Don't log in for the next 14 days!
Congratulations, you have successfully deleted your Facebook Account. At least Mark says so ..
If you think that this instruction was helpful, don't forget to
Jun 7, 2011
Server 2008 R2 "Unidentified Network"
After setting up a virtual domain environment on only one physical Server I encountered the following situation: My Domain Controller's NIC was listed as an "Unidentified Network" instead of a "Domain Network" as it should be.
The downside of an "Unidentified Network" is a resulting quite restrictive firewall profile.
The configuration is as follows:
Server 1: Hyper-V, host of several VMs and Domain Controller
Server 2: VM of Server 1 and TMG-Firewall (Gateway of Server1)
The problem in that configuration is a deadlock-situation. "Domain Network" is only assigned if the gateway which is Server 2 is reachable during the boot process. Since it is not, because the Hyper-V Host yet needs to load the gateway-VM, the network is being marked as an "Unidentified Network".
After some research on the topic I found out that you can manually change the network-type to "Work" or "Home" or "Public" but you can not set it to "Domain Network" at will. The only way I could achieve this was either by plugging out and in the cable or by deactivating and reactivating the NIC (within ncpa.cpl).
The workaround was to automatically de- and reactivate the NIC using the netsh command. To automatically do it on start up just create a batch file like this:
netsh interface set interface "LAN 1" disabled
netsh interface set interface "LAN 1" enabled
Make sure that you replace "LAN 1" with the name of your NIC and run the batch file as Administrator or another user with sufficient rights.
Then just hook it up to the task scheduler about 15 minutes after start up.
The downside of an "Unidentified Network" is a resulting quite restrictive firewall profile.
The configuration is as follows:
Server 1: Hyper-V, host of several VMs and Domain Controller
Server 2: VM of Server 1 and TMG-Firewall (Gateway of Server1)
The problem in that configuration is a deadlock-situation. "Domain Network" is only assigned if the gateway which is Server 2 is reachable during the boot process. Since it is not, because the Hyper-V Host yet needs to load the gateway-VM, the network is being marked as an "Unidentified Network".
After some research on the topic I found out that you can manually change the network-type to "Work" or "Home" or "Public" but you can not set it to "Domain Network" at will. The only way I could achieve this was either by plugging out and in the cable or by deactivating and reactivating the NIC (within ncpa.cpl).
The workaround was to automatically de- and reactivate the NIC using the netsh command. To automatically do it on start up just create a batch file like this:
netsh interface set interface "LAN 1" disabled
netsh interface set interface "LAN 1" enabled
Make sure that you replace "LAN 1" with the name of your NIC and run the batch file as Administrator or another user with sufficient rights.
Then just hook it up to the task scheduler about 15 minutes after start up.
Subscribe to:
Posts (Atom)