brandonhowlett Posted September 24, 2022 Share Posted September 24, 2022 (edited) This is one of a handful of challenges I'm facing since upgrading from 6.10.3 to 6.11. I'm starting with what I hope is low hanging fruit. Since upgrading, my log is flooded with "kernel: EDID block 0 is all zeroes" I have a supermicro X11SSH-F motherboard with an e3-1285v6. I hardware transcode with the i915 igpu for plex. So far, I've tried recommendations in posts to blacklist the i915 driver (I don't know if this would disable HW transcoding in plex, but it don't work anyway) e.g., echo "blacklist i915" > /boot/config/modprobe.d/i915.conf Reboot. Same errors. I've also unplugged the kvm and replaced it with a vga dummy plug. Reboot. Same. Any other ideas? bukadoza-diagnostics-20220924-1154.zip Edited September 24, 2022 by brandonhowlett Added diagnostic file Quote Link to comment
JorgeB Posted September 24, 2022 Share Posted September 24, 2022 Please post the diags, you might also have an AST GPU. Quote Link to comment
brandonhowlett Posted September 24, 2022 Author Share Posted September 24, 2022 Sorry, diagnostic file added. Thanks for the reminder, @JorgeB. Quote Link to comment
JorgeB Posted September 24, 2022 Share Posted September 24, 2022 Yep, there is an AST GPU (for the IPMI) and it's loading the driver, instead of blacklisting the iGPU (which is not used for video out with that board) try blacklist the AST GPU: echo "blacklist ast" > /boot/config/modprobe.d/ast.conf Quote Link to comment
brandonhowlett Posted September 24, 2022 Author Share Posted September 24, 2022 Unfortunately, that didn't resolve the error. Quote Link to comment
JorgeB Posted September 24, 2022 Share Posted September 24, 2022 Please post new diags. Quote Link to comment
brandonhowlett Posted September 24, 2022 Author Share Posted September 24, 2022 bukadoza-diagnostics-20220924-1453.zip Quote Link to comment
Tony325 Posted September 25, 2022 Share Posted September 25, 2022 me too. 6.10.3 to 6.11.0mediacenter-diagnostics-20220925-1316.zip Quote Link to comment
JorgeB Posted September 25, 2022 Share Posted September 25, 2022 Did you try blacklisting both the iGPU and the AST GPU? Quote Link to comment
JorgeB Posted September 25, 2022 Share Posted September 25, 2022 3 hours ago, Tony325 said: me too. See here for some things to try: Quote Link to comment
brandonhowlett Posted September 25, 2022 Author Share Posted September 25, 2022 (edited) @JorgeB, that was the post I started with. With both the iGPU and AST GPUs blacklisted the error persists. And with the addition of a dummy vga plug, as well. bukadoza-diagnostics-20220925-0145.zip Edited September 25, 2022 by brandonhowlett Forgot dummy plug Quote Link to comment
JorgeB Posted September 26, 2022 Share Posted September 26, 2022 Sorry, no more ideas then other than try a real monitor. Quote Link to comment
Hoopster Posted September 26, 2022 Share Posted September 26, 2022 I have a VGA monitor connected to my server. It does not stop the errors which hit the log every 2 seconds Sep 26 09:59:28 MediaNAS kernel: EDID has corrupt header Sep 26 09:59:28 MediaNAS kernel: EDID block 0 is all zeroes I have an AST (IPMI) on the server and iGPU in the E-2288G processor. I can't blacklist either as I use iGPU for transcoding and AST for KVM access to server via IPMI. Hopefully this gets fixed soon. The problem was not present in 6.10.3 which was running previously on the server. Same issue on my backup server which also has an iGPU and ASpeed chip for IPMI. No monitor is plugged into that server as it is in a closet in the basement. I guess I could try a dummy plug there but an attached monitor on main server does not prevent the errors. Quote Link to comment
brandonhowlett Posted September 26, 2022 Author Share Posted September 26, 2022 I can also confirm that attaching a monitor does not fix it. Thanks anyway, @JorgeB. Quote Link to comment
limetech Posted September 27, 2022 Share Posted September 27, 2022 8 hours ago, brandonhowlett said: I can also confirm that attaching a monitor does not fix it. Thanks anyway, @JorgeB. Please go to Tools/System Devices and check the checkbox next to the ast device and then reboot and see if issue persists. If so, try same thing with the Intel display adapter. Checking the box for a device completely isolates it from the Linux kernel and we should be able to at least determine which is the culprit. 1 Quote Link to comment
Hoopster Posted September 27, 2022 Share Posted September 27, 2022 (edited) 1 hour ago, limetech said: Please go to Tools/System Devices and check the checkbox next to the ast device and then reboot and see if issue persists. If so, try same thing with the Intel display adapter. Checking the box for a device completely isolates it from the Linux kernel and we should be able to at least determine which is the culprit. iGPU bound to VFIO = no EDID has corrupt header/EDID block 0 is all zeroes errors in syslog. ASpeed bound to VFIO = EDID has corrupt header/EDID block 0 is all zeroes errors in syslog. The iGPU in the case of my servers appears to be the culprit. Edited September 27, 2022 by Hoopster 1 Quote Link to comment
JorgeB Posted September 27, 2022 Share Posted September 27, 2022 59 minutes ago, Hoopster said: The iGPU in the case of my servers appears to be the culprit. Yep, and probably for all having these issues, also I believe that since your board has IPMI the monitor output will use the AST GPU, so connecting a monitor won't help. 1 Quote Link to comment
Hoopster Posted September 27, 2022 Share Posted September 27, 2022 2 minutes ago, JorgeB said: since your board has IPMI the monitor output will use the AST GPU, so connecting a monitor won't help. Yep, that is the case. Monitor output is via VGA connector which is controlled by the ASpeed chip. 1 Quote Link to comment
wirenut Posted September 27, 2022 Share Posted September 27, 2022 4 hours ago, Hoopster said: Yep, that is the case. Monitor output is via VGA connector which is controlled by the ASpeed chip. I can confirm my results are the same as @Hoopster I also have a monitor connected via VGA controlled by the ASpeed chip. isolating the iGPU from the kernel and the error does not appear. Quote Link to comment
brandonhowlett Posted September 27, 2022 Author Share Posted September 27, 2022 @limetech, I can confirm, it's the intel igpu. Same result as @Hoopster. Quote Link to comment
Solution limetech Posted September 27, 2022 Solution Share Posted September 27, 2022 We created a kernel patch in Unraid 6.11.1 to silence these messages. 3 3 Quote Link to comment
wirenut Posted September 27, 2022 Share Posted September 27, 2022 58 minutes ago, limetech said: We created a kernel patch in Unraid 6.11.1 to silence these messages. Good news! Thank you @limetech team! Quote Link to comment
brandonhowlett Posted September 27, 2022 Author Share Posted September 27, 2022 Great, Thank you! Quote Link to comment
qwrty Posted December 1, 2022 Share Posted December 1, 2022 Hello everyone, I'm facing the same issue on my Debian... after upgrading the kernel from 5.17.x to 5.19.17 I know it's not the same OS, but how did you patch the kernel to silent this log ? I only have this king of log: Dec 1 15:26:32 kernel: [21750.563946] EDID block 0 is all zeroes Dec 1 15:26:33 kernel: [21750.795510] EDID block 0 is all zeroes Dec 1 15:26:55 kernel: [21773.559925] EDID block 0 is all zeroes Dec 1 15:26:56 kernel: [21773.790095] EDID block 0 is all zeroes Dec 1 15:26:56 kernel: [21773.824563] EDID block 0 is all zeroes Thank you for any helps Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.