BipBip1981 Posted December 19, 2023 Posted December 19, 2023 (edited) Hi, I use Helios64 from the begin and this board is very unstable according to my experience and many posts here... After many dialogue here and settings shared about Frequency and Governor Kernel... maybe i find the problem and i ask anyone return of experiences about the procedure and setting belong: - Connect USB Wireless Dongle or USB(-c) Ethernet Plug - Configure it - Disconnect all ethernet cable to the board - Use with last time crash configuration and do a endurance test your helios64 board Have a good day Edited December 19, 2023 by BipBip1981 0 Quote
BipBip1981 Posted December 27, 2023 Author Posted December 27, 2023 Hi, not crash again 😞 i try step by step "configuration crash" (btrfs check my file system and mdadm check raid at same time, 2 days by pass) with my custom fancontrol file and step by step up cpu frequency with performance governor to not use auto scale frequency. For moment, it stable at 1000mhz, i start a minimum 7 days ago Keep in touch result Bye. 0 Quote
bunducafe Posted December 30, 2023 Posted December 30, 2023 Which kernel are you on? I am running my machine now for over 80 days without any issues whatsoever. See screenshot. I started from the last official image and then updated manually to the 5.15.93 kernel. Then rebooted and froze further kernel updates. Then I updated all the rest. Since then I have a solid machine and I won't go any further kernelwise. 0 Quote
BipBip1981 Posted January 2 Author Posted January 2 Hi, Game Over for me... again... crash with CPU @400-1200MHz with "on demand" Governor Kernel... in less than 24h... I retry same test with "schedutil" and the last main build by me... I not hope many thing... Return next week Bye 0 Quote
bunducafe Posted January 2 Posted January 2 What are you running the helios64 with? OMV? Did you do any manual changes within the system? I once had a faulty battery. Once I ripped that out everthing worked again... What does the log tell you once the machine goes down? I can feel the frustration but I believe if configured correctly the machine might run reasonably. 0 Quote
BipBip1981 Posted January 2 Author Posted January 2 (edited) Hi, It's running with Armbian Brookworm build by me with official script. Two different crash arrive, one with fault led red system flashing, second with all blue led are fix, lost network and lost console access with usb-c cable. In log, they are nothing important, crash append but i not found information in log (My log isn't in ramlog, i disable this shit feature...). I have 4 8TO raid 10 with BTRFS over Luks, i reproduce crash by run btrfs check file system with checksum and mdadm check raid at same time. My brother have helios64 and he crash his Helios simply crash on waiting with docker container active and do nothing and without any HD drive with all last official Kobol Armbian build or with my own build. I will test next week with this image system (https://dl.slarm64.org/slackware/images/helios64/) simply to know if armbian specific problem or hardware problem. Have a good day. Edited January 3 by BipBip1981 0 Quote
BipBip1981 Posted January 11 Author Posted January 11 (edited) Hi, My return, best result and best stabilty is this setting: 1200Mhz max performance with upper frequency random crash... some time just when reconnect to network (ex: restart my router, many samba access) or launch docker container... or just unlock drive with cryptosetup... or during "btrfs check" Have a good day Edited January 13 by BipBip1981 0 Quote
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.