July 1, 20179 yr Hi Guys, I have been running UNRAID 6.3.5 for the last few weeks without any issues. Last night the machine locked up and I needed to hard reset it. When I rebooted; very early into the boot sequence, pretty much straight after loading the boot.img file, the kernel posted a kernel panic. I originally thought that the USB key was corrupted...and it was.....so I performed a chkdsk in Windows and repaired it however this did not fix the problem. I have since backed up the USB key and performed a fresh install with and without my /config folder however as soon as the boot img file is loaded I receive the following error. The USB key is fine, I tried another key and a fresh install and it comes up the same error. I have run memtest for.....2 passes for 6 hours without any errors. I am going to run it again overnight but honestly don't think I will encounter any issues. The BIOS has the very latest update....dated May 2017. I am unable to boot in any mode, shell, GUI, safemode with no plugins etc. As a sanity check I reverted to an older BIOS and also reflashed the current BIOS. Everything is at its default clocks, voltages etc...... The error is consistent and always the same..... I have also pulled two sticks of memory and also swapped memory sticks and banks..... I am at a complete loss and have been trying to diagnose this now for the better part of a day. I am unable to access UNRAID in any way so am stuck.
July 1, 20179 yr Author 7 hours ago, jonathanm said: As a hail mary, try disconnecting all drives and see if you can get a login prompt. Thanks for the advice. I disabled all the drives and controllers in the BIOS; I realise this isn't the same as electrically disconnecting them however its a real pain to disconnect all the drives in this machine there are over 24 HDD's. I would have to individually tag and trace each SATA cable. I booted the machine and it still came up with the same error. I think it may be memory related memory; memtest froze overnight after 6 hours ironically with no errors reported. Also I found this article from Red Hat that seems to indicate a 'kernel panic on hrtimer_run_queues' error is hardware RAM related. Unfortunately I don't have a second set of memory to validate. https://access.redhat.com/solutions/56530 Until put a second set of RAM sticks in this machine I won't really know if that is the root cause.
July 2, 20179 yr 17 hours ago, Enver said: in this machine there are over 24 HDD's. I would have to individually tag and trace each SATA cable Completely up to you, but that's a recipe for disaster. With that many drives, you should definitely invest in hotswap bays, as SATA signal / power cables aren't known for always making robust connections. If you ever have to swap out a drive, the odds are extremely good that you're going to disturb another drive(s) connectors which could impact the ability to rebuild the drive.
July 2, 20179 yr 20 minutes ago, Squid said: Completely up to you, but that's a recipe for disaster. With that many drives, you should definitely invest in hotswap bays, as SATA signal / power cables aren't known for always making robust connections. If you ever have to swap out a drive, the odds are extremely good that you're going to disturb another drive(s) connectors which could impact the ability to rebuild the drive. Absolutely! This is the one I recommend. SuperMicro CSE-M35T-1B (Newegg) SuperMicro CSE-M35T-1B (eBay) And read more below about reasons to use the cages:
July 13, 20178 yr Author Thanks for the advice everyone. I must have been having a blonde moment; I do have HDD cages in this machine with hotswap capability. I have tried pulling all the drives; I have also tried disabling all the peripherals in BIOS; I even pulled SATA cables thinking it might be a bad cable. I have managed to get a loan of some memory and unfortunately I am still having the exact same problem. This has been going on for almost two weeks now; I really need advice from someone on this forum who has had experience with this sort of thing. My friends are telling me to bin uNRAID and install Windows!! I am actually considering this as I have no idea if its something peculiar(bug) to the uNRAID kernel? I have tried a clean install of the uNRAID USB key; safe mode etc......the error is always the same. Can someone please take a long close look at the screen capture......
July 13, 20178 yr Author Can someone supply uNRAID 6.4 RC6 on a zip file please? Obviously I am unable to down this via plugin.
July 13, 20178 yr Wish I could be more help. Feels hardware related to me. Might look at this link. https://en.m.wikipedia.org/wiki/Kernel_panic Perhaps @johnnie.black will take a look and comment on your issue. At these times helps to have a functioning computer so you have an ability to swap in known functioning parts. When you only have a broken system, isolation is hard and frustrating. Seeing your post. I do not have. Your 'previous' folder on the USB stick will have the previous version of unRaid that you could revert to. Not sure what version that was. Good luck!
July 13, 20178 yr Author 6 minutes ago, bjp999 said: Wish I could be more help. Feels hardware related to me. Might look at this link. https://en.m.wikipedia.org/wiki/Kernel_panic Perhaps @johnnie.black will take a look and comment on your issue. To diagnose it'd try to get to the most basic config - motherboard, PSU, one stick of RAM, video card and keyboard. No add in controllers, no disks (not even powered). From there make sure you can get into bios. If you can get to that level of functioning, try the USB for unRaid. If that fails, try a different stick of RAM. If no luck, try some other booting usb. Like Windows install. Our different stick with different version of unRaid. See if you can get anything to boot. At these times helps to have a functioning computer so you have an ability to swap in known functioning parts. When you only have a broken system, isolation is hard and frustrating. Seeing your post. I do not have. Your 'previous' folder on the USB stick will have the previous version of unRaid that you could revert to. Not sure what version that was. Good luck! Hi bjp999, So getting into the BIOS is not a problem; it appears stable there. I am yet to install Windows but right now when I boot off the USB key vanilla unRAID 6.3.5 or my configuration it goes into kernel panic mode. I had been running .6.3.5 for a few weeks without issue or so it seems; it just froze one night and i have been in this situation ever since. I was reluctant to try an earlier version of uNRAID, I am unsure what that will do the configuration or parity etc......but I guess I don't have much of a choice right now.
July 13, 20178 yr Community Expert 10 minutes ago, bjp999 said: Feels hardware related to me. Agree, if it was previously booting with the same unRAID release and the OP already tried a new flash drive with a fresh install.
July 13, 20178 yr @Enver I don't see much risk going back to a prior version, unless you have added an obnoxious number of disks recently (recent versions upped the max drive count).
July 13, 20178 yr Author 1 hour ago, bjp999 said: @Enver I don't see much risk going back to a prior version, unless you have added an obnoxious number of disks recently (recent versions upped the max drive count). Here is what I get with a fresh/different USB key and unRAID 6.3.4. Its a different error message; CPU?
July 13, 20178 yr My extremely limited experience has been that the motherboard rather than CPU is more likely to fail.
July 16, 20178 yr Author Ok, I have narrowed it down to Core 2 on my 5960x. If I disable that core in the BIOS, unRAID boots and works as expected. If I enable it it kernel panics. I have tried to install Windows 10 natively via bootable USB key with Core 2 enabled, and it BSOD before the Windows Logo is even loaded. The problem is the same and repeatable. I know this could still be motherboard, but I disabled Core 2 and booted the Intel Linux CPU Diagnostics Util, it boots off a Fedora USB key and ran some tests, both CPU(Core 2 disabled) and PCH (on mobo) passed. When I tried to boot off the Intel Linux Diagnostics CPU Util USB with Core 2 enabled the system froze very early into the boot process. So for now it does look like it's a failed Core. Will RMA CPU and report back.
Archived
This topic is now archived and is closed to further replies.