August 25, 201510 yr Hi i am a longtime unraid user and decided to upgrade my unraid, my system was running stable before the upgrade i got myself an ASUS ROG Crossblade Ranger + AMD A10-7870K Black Edition, 4x 3.90GHz + 16GB DDR3 RAM i used Version 6.0beta12 first but ran into kernel panic, upgraded to 6.0.1, got into kernel panic again, i could not scroll upwards and syslog is gone after reboot, so no real information about the error, this happend between 1-4hours uptime so i tried 6.0.1 in safemode and at some point ksyslogd was using over 100% usage, i think 100% means 1 core so i looked into the syslog and found following Aug 24 22:41:25 Tower kernel: Call Trace: Aug 24 22:41:25 Tower kernel: [<ffffffff815ff789>] dump_stack+0x4c/0x6e Aug 24 22:41:25 Tower kernel: [<ffffffff815fb708>] __schedule_bug+0x46/0x54 Aug 24 22:41:25 Tower kernel: [<ffffffff8160188f>] __schedule+0x66/0x676 Aug 24 22:41:25 Tower kernel: [<ffffffff81601f0f>] schedule+0x70/0x7f Aug 24 22:41:25 Tower kernel: [<ffffffff8160215a>] schedule_preempt_disabled+0x10/0x19 Aug 24 22:41:25 Tower kernel: [<ffffffff8106e0bd>] cpu_startup_entry+0x2d3/0x2da Aug 24 22:41:25 Tower kernel: [<ffffffff8102f040>] start_secondary+0x122/0x140 Aug 24 22:41:25 Tower kernel: BUG: scheduling while atomic: swapper/1/0/0x00000201 Tower kernel: Modules linked in: md_mod fam15h_power igb i2c_algo_bit ptp ahci pps_core libahci i2c_piix4 mvsas libsas k10temp scsi_transport_sas acpi_cpufreq [last unloaded: md_mod] Aug 24 22:41:25 Tower kernel: CPU: 1 PID: 0 Comm: swapper/1 Tainted: G W 4.0.4-unRAID #5 Aug 24 22:41:25 Tower kernel: Hardware name: System manufacturer System Product Name/CROSSBLADE RANGER, BIOS 0507 07/01/2014 Aug 24 22:41:25 Tower kernel: 0000000000000001 ffff88040d58fe38 ffffffff815ff789 000000000000f482 Aug 24 22:41:25 Tower kernel: ffff88041ec93ac0 ffff88040d58fe48 ffffffff815fb708 ffff88040d58fe98 Aug 24 22:41:25 Tower kernel: ffffffff8160188f ffff88040d58fe48 ffffffff81995520 ffff88040d58fe78 Aug 24 22:41:25 Tower kernel: Call Trace: Aug 24 22:41:25 Tower kernel: [<ffffffff815ff789>] dump_stack+0x4c/0x6e Aug 24 22:41:25 Tower kernel: [<ffffffff815fb708>] __schedule_bug+0x46/0x54 Aug 24 22:41:25 Tower kernel: [<ffffffff8160188f>] __schedule+0x66/0x676 Aug 24 22:41:25 Tower kernel: [<ffffffff81601f0f>] schedule+0x70/0x7f Aug 24 22:41:25 Tower kernel: [<ffffffff8160215a>] schedule_preempt_disabled+0x10/0x19 Aug 24 22:41:25 Tower kernel: [<ffffffff8106e0bd>] cpu_startup_entry+0x2d3/0x2da Aug 24 22:41:25 Tower kernel: [<ffffffff8102f040>] start_secondary+0x122/0x140 Aug 24 22:41:25 Tower kernel: BUG: scheduling while atomic: swapper/1/0/0x00000201 Aug 24 22:41:25 Tower kernel: Modules linked in: md_mod fam15h_power igb i2c_algo_bit ptp ahci pps_core libahci i2c_piix4 mvsas libsas k10temp scsi_transport_sas acpi_cpufreq [last unloaded: md_mod] Aug 24 22:41:25 Tower kernel: CPU: 1 PID: 0 Comm: swapper/1 Tainted: G W 4.0.4-unRAID #5 system crashed after around 8hours in safemode any opinions? i guess my hardware is to new for this kernel? or does this look like a hardware-failure? unRAID OS Version: 6.0beta12 & 6.0.1 Description:kernel panic got a new message in my console after 25minutes runtime Message from syslogd@Tower at Aug 25 21:42:17 ... kernel:page:ffffea000ce6a5c0 count:1 mapcount:0 mapping: (null) index:0xca Message from syslogd@Tower at Aug 25 21:42:17 ... kernel:flags: 0x200000000000000()
August 25, 201510 yr Author system uses around 280Watt peak at bootup and 160watt while parity check 12V single rail operation http://www.bequiet.com/en/powersupply/386
August 25, 201510 yr Under technical specifications of your PSU it states: 12V single rail operation: yes +12V1 (A): 62 I think we can rule out your PSU as a possible source of the issue.
August 26, 201510 yr Author Is this completely-stock bare-metal unRaid install? If so, please try 6.1-rc6. no it's not completly stock, thats the thing i like on unRaid that i have a linux-server where i can run some stuff, i didn't have the time to checkout if docker other VM hit my usability wishes, i run my system that way since Unraid 4, but i had the time to play with BIOS settings, and there are a lot of settings in my BIOS System has now 10hours uptime, that is a new record so here is what i changed: 1)forced the PCIe to 8x since my controllers are 8x 2)set the "Core Leveling Mode" to "One core per processor" my guess is that it has to do with the "Core Leveling Mode", one core is really not much, so if it runs stable until the weekend i will step up to "One core per Compute Unit" i will keep you updated
August 28, 201510 yr Author after more the 2 days uptime i updated to 6.1-rc6 and switch bios cpu settings to "automatic" started array, after 5 minutes i lost 1 disk reboot switching to "One core per Compute Unit" after boot i lost the seconed disk..... switched back to "One core per processor" after boot all disks were present, the seconed disk i lost was back with a green light, the first one with a red cross (guess because there was some write access when the array was up and the disk was missing) could not rebuild that on that disk changed the disk to rebuild
September 4, 201510 yr Author after more then 2 days of stable uptime i can tell you that the issue was cause by an BIOS-Version that was not uptodate and not ready for my cpu sorry for that i still have drives that my sas-controller finds on bootup but unraid doesn't, i hotswap them out and in, then unraid finds them
Archived
This topic is now archived and is closed to further replies.