September 17, 201312 yr Hello I was curious if there was a way to run a memory test while unRaid is loaded (or running)? it looks like only one memory stick is bad (OCZ ... so ... ) Sep 16 19:40:01 storage kernel: INFO: rcu_sched self-detected stall on CPU { 5} (t=6001 jiffies g=380020 c=380019 q=19626) Sep 16 19:40:01 storage kernel: Pid: 9523, comm: unraidd Not tainted 3.9.6p-unRAID #23 Sep 16 19:40:01 storage kernel: Call Trace: Sep 16 19:40:01 storage kernel: [<c1062ba6>] print_cpu_stall+0xbc/0x107 Sep 16 19:40:01 storage kernel: [<c1062e36>] __rcu_pending+0x4f/0x12a Sep 16 19:40:01 storage kernel: [<c1062f84>] rcu_check_callbacks+0x73/0x9b Sep 16 19:40:01 storage kernel: [<c1032ead>] update_process_times+0x2d/0x53 Sep 16 19:40:01 storage kernel: [<c1055167>] tick_sched_timer+0x77/0xa1 Sep 16 19:40:01 storage kernel: [<c1040d86>] ? __remove_hrtimer+0x25/0x7a Sep 16 19:40:01 storage kernel: [<c1040ec9>] __run_hrtimer+0x45/0xaf Sep 16 19:40:01 storage kernel: [<c1041231>] hrtimer_interrupt+0xf1/0x1e7 Sep 16 19:40:01 storage kernel: [<c101c43a>] smp_apic_timer_interrupt+0x6d/0x7f Sep 16 19:40:01 storage kernel: [<c14033f9>] apic_timer_interrupt+0x2d/0x34 Sep 16 19:40:01 storage kernel: [<c12fb86b>] ? scsi_request_fn+0x33a/0x371 Sep 16 19:40:01 storage kernel: [<c1246f7b>] __blk_run_queue+0x28/0x31 Sep 16 19:40:01 storage kernel: [<c12472bd>] blk_run_queue+0x1b/0x2c Sep 16 19:40:01 storage kernel: [<c12fb0b6>] scsi_run_queue+0xe4/0x151 Sep 16 19:40:01 storage kernel: [<c12fba35>] scsi_next_command+0x28/0x34 Sep 16 19:40:01 storage kernel: [<c12fbf50>] scsi_end_request+0x66/0x70 Sep 16 19:40:01 storage kernel: [<c12fc274>] scsi_io_completion+0x1b0/0x421 Sep 16 19:40:01 storage kernel: [<c12fc00d>] ? scsi_device_unbusy+0x7c/0x82 Sep 16 19:40:01 storage kernel: [<c12f70a8>] scsi_finish_command+0x91/0x97 Sep 16 19:40:01 storage kernel: [<c12fc5b9>] scsi_softirq_done+0xc5/0xcd Sep 16 19:40:01 storage kernel: [<c104a251>] ? enqueue_task_fair+0x132/0x13a Sep 16 19:40:01 storage kernel: [<c124ca5a>] blk_done_softirq+0x4a/0x57 Sep 16 19:40:01 storage kernel: [<c102e99f>] __do_softirq+0x94/0x151 Sep 16 19:40:01 storage kernel: [<c1044ca1>] ? ttwu_do_wakeup+0xf/0xaa Sep 16 19:40:01 storage kernel: [<c102eabc>] irq_exit+0x33/0x6c Sep 16 19:40:01 storage kernel: [<c100367e>] do_IRQ+0x87/0x9b Sep 16 19:40:01 storage kernel: [<c1403d2c>] common_interrupt+0x2c/0x31 Sep 16 19:40:01 storage kernel: [<c125e261>] ? memcmp+0x1a/0x25 Sep 16 19:40:01 storage kernel: [<f957ccf3>] handle_stripe+0xa53/0xcf6 [md_mod] Sep 16 19:40:01 storage kernel: [<c1044ebb>] ? __wake_up+0x3b/0x42 Sep 16 19:40:01 storage kernel: [<f957d007>] unraidd+0x71/0xb5 [md_mod] Sep 16 19:40:01 storage kernel: [<f9579bf4>] md_thread+0xd3/0xea [md_mod] Sep 16 19:40:01 storage kernel: [<c103efb5>] ? wake_up_bit+0x5b/0x5b Sep 16 19:40:01 storage kernel: [<c103eb75>] kthread+0x90/0x95 Sep 16 19:40:01 storage kernel: [<f9579b21>] ? import_device+0x166/0x166 [md_mod] Sep 16 19:40:01 storage kernel: [<c1403837>] ret_from_kernel_thread+0x1b/0x28 Sep 16 19:40:01 storage kernel: [<c103eae5>] ? kthread_freezable_should_stop+0x4a/0x4a Sep 16 19:46:24 storage kernel: INFO: rcu_sched self-detected stall on CPU { 5} (t=6000 jiffies g=380487 c=380486 q=1455) Sep 16 19:46:24 storage kernel: Pid: 9523, comm: unraidd Not tainted 3.9.6p-unRAID #23 Sep 16 19:46:24 storage kernel: Call Trace: Sep 16 19:46:24 storage kernel: [<c1062ba6>] print_cpu_stall+0xbc/0x107 Sep 16 19:46:24 storage kernel: [<c1062e36>] __rcu_pending+0x4f/0x12a Sep 16 19:46:24 storage kernel: [<c1062f84>] rcu_check_callbacks+0x73/0x9b Sep 16 19:46:24 storage kernel: [<c1032ead>] update_process_times+0x2d/0x53 Sep 16 19:46:24 storage kernel: [<c1055167>] tick_sched_timer+0x77/0xa1 Sep 16 19:46:24 storage kernel: [<c1040d86>] ? __remove_hrtimer+0x25/0x7a Sep 16 19:46:24 storage kernel: [<c1040ec9>] __run_hrtimer+0x45/0xaf Sep 16 19:46:24 storage kernel: [<c1041231>] hrtimer_interrupt+0xf1/0x1e7 Sep 16 19:46:24 storage kernel: [<c101c43a>] smp_apic_timer_interrupt+0x6d/0x7f Sep 16 19:46:24 storage kernel: [<c14033f9>] apic_timer_interrupt+0x2d/0x34 Sep 16 19:46:24 storage kernel: [<c1402e53>] ? _raw_spin_unlock_irqrestore+0x8/0xa Sep 16 19:46:24 storage kernel: [<c1248252>] blk_end_bidi_request+0x4a/0x53 Sep 16 19:46:24 storage kernel: [<c124829a>] blk_end_request+0x12/0x14 Sep 16 19:46:24 storage kernel: [<c12fbf09>] scsi_end_request+0x1f/0x70 Sep 16 19:46:24 storage kernel: [<c12fc274>] scsi_io_completion+0x1b0/0x421 Sep 16 19:46:24 storage kernel: [<c12fc00d>] ? scsi_device_unbusy+0x7c/0x82 Sep 16 19:46:24 storage kernel: [<c12f70a8>] scsi_finish_command+0x91/0x97 Sep 16 19:46:24 storage kernel: [<c12fc5b9>] scsi_softirq_done+0xc5/0xcd Sep 16 19:46:24 storage kernel: [<c12fc274>] ? scsi_io_completion+0x1b0/0x421 Sep 16 19:46:24 storage kernel: [<c124ca5a>] blk_done_softirq+0x4a/0x57 Sep 16 19:46:24 storage kernel: [<c102e99f>] __do_softirq+0x94/0x151 Sep 16 19:46:24 storage kernel: [<c1048335>] ? sched_clock_cpu+0x3f/0x13f Sep 16 19:46:24 storage kernel: [<c102eabc>] irq_exit+0x33/0x6c Sep 16 19:46:24 storage kernel: [<c100367e>] do_IRQ+0x87/0x9b Sep 16 19:46:24 storage kernel: [<c101b00f>] ? native_smp_send_reschedule+0x3f/0x41 Sep 16 19:46:24 storage kernel: [<c1044c2b>] ? resched_task+0x57/0x5a Sep 16 19:46:24 storage kernel: [<c1403d2c>] common_interrupt+0x2c/0x31 Sep 16 19:46:24 storage kernel: [<c1243e3f>] ? xor_sse_5_pf64+0xc5/0x32c Sep 16 19:46:24 storage kernel: [<c124333e>] xor_blocks+0x74/0x7c Sep 16 19:46:24 storage kernel: [<f957c186>] check_parity+0x96/0xcc [md_mod] Sep 16 19:46:24 storage kernel: [<f957cccf>] handle_stripe+0xa2f/0xcf6 [md_mod] Sep 16 19:46:24 storage kernel: [<c1044ebb>] ? __wake_up+0x3b/0x42 Sep 16 19:46:24 storage kernel: [<f957d007>] unraidd+0x71/0xb5 [md_mod] Sep 16 19:46:24 storage kernel: [<f9579bf4>] md_thread+0xd3/0xea [md_mod] Sep 16 19:46:24 storage kernel: [<c103efb5>] ? wake_up_bit+0x5b/0x5b Sep 16 19:46:24 storage kernel: [<c103eb75>] kthread+0x90/0x95 Sep 16 19:46:24 storage kernel: [<f9579b21>] ? import_device+0x166/0x166 [md_mod] Sep 16 19:46:24 storage kernel: [<c1403837>] ret_from_kernel_thread+0x1b/0x28 Sep 16 19:46:24 storage kernel: [<c103eae5>] ? kthread_freezable_should_stop+0x4a/0x4a Someone said that would be caused by plugins in another thread ... I rebooted and ran memtest and I'm getting errors around 4150MB to 4260MB I've tried ram speeds from ddr3-1333 down to ddr3-800 and the slower I run the memory the fewer the errors ... but I don't know how to slow it down more. Not sure how to figure out all the kludged memory settings. I was curious if I could use the mem=2047 parameter in the syslinux.cfg file to keep to the better stick (or better area of memory) without having to remove the bad stick. If I did this I would like to be able to test memory to make sure I'd be using the good memory. I have to disconnect all my HTPC connections to get into my unRaid currently ... I'd only like to have to do that once when I get the new memory and I still have to wait a few days for replacement memory ... Thanks for any time and help, Bobby
September 29, 201312 yr I'm experiencing the same issue: http://lime-technology.com/forum/index.php?topic=29616.msg265711#msg265711
September 29, 201312 yr I rebooted and ran memtest and I'm getting errors around 4150MB to 4260MB You need to replace the defective module. Remove all but one memory module; boot to Memtest; and then test the module individually until you isolate the defective one -- then replace it.
October 1, 201312 yr You could try to increase the voltage to your ram modules. That sometimes helps, though most of times replacing the defective module is the only solution.
October 3, 201312 yr Author I got the new memory and replaced it ... all of it I had way too much trouble with the OCZ memory (bought one batch for my HTPC and returned to store bad ... bought another batch for my unRaid and returned to store bad ... tried to RMA the second batch that I returned a year ago when the first module of the three died now another of that batch is gone ... and I'm still waiting) the old memory was rated 1.65V and I had up to 1.75V at 800 Mhz and still erroring out above 4,100MB I was just hoping there was a lazy way as I have server cases stacked up without a rack and unRaid is on the bottom If I could have checked the first dimm and limited to that ... or first few GB ... Would the mem=2047 parameter work the same for memtest as it would for unRaid? (grab the same chunk of memory?) I've got to get these in a rack ... just looking for some inexpensive 20"+ shelves with a 45 pound capacity that would fit in a 30" deep 4 post rack ... fixed would be good on rails would be spectacular ... Thanks for everyone's time, Bobby
October 3, 201312 yr I have server cases stacked up without a rack and unRaid is on the bottom Get a rack !! As Prostuff noted, check Craigslist -- there are often listings for racks from out-of-business companies, or companies that are upgrading to larger racks. But in any event, it's worth springing for a proper rack for your gear.
Archived
This topic is now archived and is closed to further replies.