Hey all. This bug looks to be known (but on a different kernel in the Debian report), but from what I can tell it was supposed to be fixed and yet I'm seeing it. It's very easy to replicate (unlike the bug reports where I'm assuming they weren't using ZFS and had to create the high load...
Memory is 256GB, as shown by TOM. Or at least, that's what should be shown, sometimes the registers seem to stay on a hot reboot.
Not sure why it says kernel module is mpt3sas for the 2308, but otherwise this is all the additional PCIe devices as seen from the system report (with the vGPUs cut...
And that did it. I made the edit to the kernel cmdline and did a reboot, but it was no good. Nothing seemed to change. Following another power cycle, the hot reboots are now working. Additionally, I see the correct entries in /proc/mtrr, and I noticed another little something during boot...
Yes, I have checked. No BIOS updates for the mainboard (v1.1 is latest, mainboard is at v1.1). Been trying to research the MTRR thing. I've run across the MTRR cleanup option but I wanted to see how cat /proc/mtrr displayed, if it was different from a cold boot to a hot one. I have the answer to...
Well, there is a definite difference in the dmesg/journal between the two. Initial comparison between the files so far I only see that in the reboot set of logs the Tonga adapters are not mentioned (at all, again). The PCI sections are just blank in reboot compared to power cycle. Attaching...
It's a Supermicro H11SSL-NC. I have PCIE Spread Spectrum off. Not sure what else it could be.
The card is installed in Slot 4 (hence why 3 is disabled, since it's unusable).
Tried to edit but apparently message is too long.
Here's the system report with regards to the PCI devices:
After a power cycle:
# lspci -nnk
00:00.0 Host bridge [0600]: Advanced Micro Devices, Inc. [AMD] Family 17h (Models 00h-0fh) Root Complex [1022:1450]
Subsystem: Advanced Micro...
Hey, @dcsapak have you noticed any unusual behaviour on reboot of the host for your card? I've found that a poweroff and poweron will show the expected lspci with all the various Tonga virtual adapters, but if I reboot the machine (ie by typing reboot at prompt) the machine will come back up...
That did indeed help.
Additionally, for those that find this thread via Google, I was using a Supermicro H11SSL-NC board. Booting from the SATA connectors or the x4 PCI slot works without issue, the onboard SAS 3008 chipset is the issue here. I'm using the 846XA chassis so the distance between...
So found this article (link) on StackExchange that talks about changing the ZFS_INITRD_POST_MODPROBE_SLEEP='0' line. I attempted to edit both it and the preceding line:
# Wait for this many seconds in the initrd pre_mountroot?
# This delays startup and should be '0' on most systems.
# Only...
Hey all,
Running into some problems getting the system to work properly now that release is out and I want to make a ZFS Mirror instead of just a single drive. For clarity, the system showed no issue booting using ext4 or xfs on NVME or SATA SSDs (if used by themselves) but now that I'm...
@dcsapak
So made this compile successfully. There is a 5.0 kernel fix / patch on pull request for the MxGPU repo (can't post link, but only 3 pull requests and it's labeled well enough). Just checkout the branch and it'll build.
Here's the DKMS file I used:
PACKAGE_NAME=amd_gim...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.