 |
Did you verify the checksum of the iso prior to writing it to the USB stick? Did you try the "Test this media and install" option (I assume you did, but I have to verify) ? If you did, did the media check pass ? |
|
|
Yes, I did both and both passed. The install progressed further when I selected the "Test this media and install" option. When I just did the normal install, the installer would crash almost immediately. I also let memtest86 run for a while and no issues were reported. I have the full install iso downloaded, so I may try that and see what happens. |
|
|
That would be a good test, please do it. We might find out if it's an issue restricted to the netinstall or maybe your motherboard is not supported. To be honest random crashes usually rhyme with defective hardware, an error in the program usually manifest always in the same place so I am tempted to not blame anaconda yet. Mind that for conclusive results, memtest needs to be run for more than "a while". And even then, running a PC in dual-channel mode might mask memory problems. Happened to me once, a system passed all memtest tests with flying honors until I left a single DIMM inside. As soon as I did that, memtest detected it as being defective. |
|
|
I have the same problem with ASRock AD2550B-ITX (Intel D2550 CPU with Intel GMA 500). This system is very similar to yours. I think is a problem with X driver. Installing with Anaconda GUI produce a Kernel Panic on shutdown and at bootloader install, but using Anaconda CLI installation works fine You are using EFI Boot, right? |
|
|
I tried the full install ISO with the same result. I also tried using a serial port console to log the kernel messages and didn't get anything useful, both for normal and text mode installation. Crashes seemed completely random and switching to text mode didn't fix the issue. Sometimes I would see a kernel panic other times there would be a python error. I also thought hardware problems initially, but I'm not convinced now. I let memtest86 run overnight (4 passes) with no errors reported. Single channel memory setup. CentOS 6.5 and Ubunto both run fine. I'm using EFI boot. |
|
|
the same happened with us on j1900 and it's randomly crash after the first few minutes of use (usuably just after we login into X) |
 |
I would not completely eliminate the possibility of bad memory. In at least one incident in my case, running memtest overnight did not find any error. But it eventually found some errors after I extended the running time to > 24 hours. |
|
|
it's not a hardware problem. and since this kernel line option: intel_pstate=disable solve the problem for us i'm sure it's a kernel or intel's kernel module bug. |
|
|
Perhaps, the OP and others reporting the issue may want to try and see if the intel_pstate=disable kernel option that worked for lfarkas also fixes the problem for them. |
|
|
Congratulations! Worked on ASRock Q1900M motherboard with Intel J1900 CPU after adding intel_pstate=disable kernel option. Thank you very much! |
|
|
Ooops! Random kernel crashes again after installed whether intel_pstate=disable or not. |
|
|
Confirmed, that's another bug for Logitech Unify receiver. intel_pstate=disable works fine! |
|
|
Same problem. intel_pstate=disable allows the install to continue, but after the install, rebooting hangs at "tsc : Fast TSC calibration failed". Turning off precision timer in bios has no effect. Machine requires a hard reset at this point. I filed bug report http://bugs.centos.org/view.php?id=7948 on 11-29-14 about this. Hardware is: Gigabyte GA-J1800N-D2H Motherboard Intel Dual-Core Celeron J1800 SoC (2.41Ghz) processor. 8GB RAM 480GB Crucial SSD Drive |
- Anonymous
