avalonea Posted November 28, 2011 Share Posted November 28, 2011 Hi ! Having a bit of a drama here with unraid beta5 14, This problem has also happened on 11/13, with or without additional packages. It randomly strikes, maybe at least once every 24 hours. Forcing me to have to do a hard power cycle on the machine. I have attached a syslog, That I got after the problem happened a few mins after boot. I believe this is the line 1327 : kernel: irq 18: nobody cared Is there a command to tell my server that I care ? I have attempted talking to it, but I didn't enable the sound card in the bios.... So we are having problems communicating. I would value any input anyone may have! Joshua. Ps. I did attempt to use version 4.7 , although due to my choice of hardware, it wasn't an option. syslog.txt Quote Link to comment
bcbgboy13 Posted November 28, 2011 Share Posted November 28, 2011 If "nobody cared" for IRQ18 one should take care and flash the BIOS to the latest available, then load the default values, then disabling any unused hardware features and then check out if this has gone away. Quote Link to comment
ubuntuaddicted Posted November 28, 2011 Share Posted November 28, 2011 If this is an unRAID server then disable the sound card in your bios, save and exit. Should fix it. I'm not sure unRAID doesn't like it and actually makes you hard reboot the machine merely cause a sound card isnt disabled BUT that's probably another thread Tapatalk is tha shizzle Quote Link to comment
avalonea Posted November 30, 2011 Author Share Posted November 30, 2011 Thanks for your input guys! I have updated the bios of the motherboard, and reconfigured to make sure that sound card/serial ports/ other unused stuff is all disabled. Although this didn't fix the issue. I also had a better look at the sata card im using Si3114, and after moving it between pci slots, saw that the IRQ problem followed it ( moved to IRQ17 ) . There was a update for the bios, and after some painful attempts to update it, I have it at the latest version. Fingers cross that fixed it , but Ill check back in after 48 hours again Quote Link to comment
avalonea Posted November 30, 2011 Author Share Posted November 30, 2011 So it happened again. I believe this is the section of the log that's relevant. Nov 30 14:52:44 avalonea-FS kernel: irq 17: nobody cared (try booting with the "irqpoll" option) Nov 30 14:52:44 avalonea-FS kernel: Pid: 0, comm: swapper Not tainted 3.1.1-unRAID #1 Nov 30 14:52:44 avalonea-FS kernel: Call Trace: Nov 30 14:52:44 avalonea-FS kernel: [<c104fa8c>] __report_bad_irq+0x1f/0x95 Nov 30 14:52:44 avalonea-FS kernel: [<c104fc39>] note_interrupt+0x137/0x1a8 Nov 30 14:52:44 avalonea-FS kernel: [<f84e5832>] ? sil_interrupt+0x1d/0x67 [sata_sil] Nov 30 14:52:44 avalonea-FS kernel: [<c104e776>] handle_irq_event_percpu+0xef/0x100 Nov 30 14:52:44 avalonea-FS kernel: [<c1050152>] ? handle_edge_irq+0xcb/0xcb Nov 30 14:52:44 avalonea-FS kernel: [<c104e7ab>] handle_irq_event+0x24/0x3b Nov 30 14:52:44 avalonea-FS kernel: [<c1050152>] ? handle_edge_irq+0xcb/0xcb Nov 30 14:52:44 avalonea-FS kernel: [<c10501bb>] handle_fasteoi_irq+0x69/0x82 Nov 30 14:52:44 avalonea-FS kernel: <IRQ> [<c1003566>] ? do_IRQ+0x37/0x90 Nov 30 14:52:44 avalonea-FS kernel: [<c130c669>] ? common_interrupt+0x29/0x30 Nov 30 14:52:44 avalonea-FS kernel: [<c11ddd89>] ? acpi_idle_enter_bm+0x22a/0x25e Nov 30 14:52:44 avalonea-FS kernel: [<c12734f0>] ? cpuidle_idle_call+0x75/0xbd Nov 30 14:52:44 avalonea-FS kernel: [<c1001a5f>] ? cpu_idle+0x39/0x5a Nov 30 14:52:44 avalonea-FS kernel: [<c12fbd40>] ? rest_init+0x58/0x5a Nov 30 14:52:44 avalonea-FS kernel: [<c145172d>] ? start_kernel+0x28c/0x291 Nov 30 14:52:44 avalonea-FS kernel: [<c14510b0>] ? i386_start_kernel+0xb0/0xb7 Nov 30 14:52:44 avalonea-FS kernel: handlers: Nov 30 14:52:44 avalonea-FS kernel: [<f84e5815>] sil_interrupt Nov 30 14:52:44 avalonea-FS kernel: Disabling IRQ #17 Quote Link to comment
prostuff1 Posted November 30, 2011 Share Posted November 30, 2011 so try the irqpoll option like it suggests in the syslinux.cfg file on the flash drive. mine looks like this: default menu.c32 menu title Lime Technology LLC prompt 0 timeout 50 label unRAID OS menu default kernel bzimage append initrd=bzroot irqfixup rootdelay=10 label FreeDOS kernel memdisk append iso initrd=fdbasecd.iso label Memtest86+ kernel memtest notice the irqfix up that I put in mine. on my board I get a disabling of IRQ 18 and IRQ19. One is my PCI-e x16 slot with a SASLP card in it and the other is my GB Ethernet port. When I get the disabling message for those the server still works but EVERYTHING grinds to painful slowness. Quote Link to comment
avalonea Posted November 30, 2011 Author Share Posted November 30, 2011 Thanks for that ! I have changed it and rebooted the server. Fingers crossed this works! Quote Link to comment
avalonea Posted November 30, 2011 Author Share Posted November 30, 2011 Ah , no joy! It happened again! Nov 30 17:41:10 avalonea-FS kernel: irq 17: nobody cared (try booting with the "irqpoll" option) Nov 30 17:41:10 avalonea-FS kernel: Pid: 0, comm: swapper Not tainted 3.1.1-unRAID #1 Nov 30 17:41:10 avalonea-FS kernel: Call Trace: Nov 30 17:41:10 avalonea-FS kernel: [<c104fa8c>] __report_bad_irq+0x1f/0x95 Nov 30 17:41:10 avalonea-FS kernel: [<c104fc39>] note_interrupt+0x137/0x1a8 Nov 30 17:41:10 avalonea-FS kernel: [<c104e776>] handle_irq_event_percpu+0xef/0x100 Nov 30 17:41:10 avalonea-FS kernel: [<c1050152>] ? handle_edge_irq+0xcb/0xcb Nov 30 17:41:10 avalonea-FS kernel: [<c104e7ab>] handle_irq_event+0x24/0x3b Nov 30 17:41:10 avalonea-FS kernel: [<c1050152>] ? handle_edge_irq+0xcb/0xcb Nov 30 17:41:10 avalonea-FS kernel: [<c10501bb>] handle_fasteoi_irq+0x69/0x82 Nov 30 17:41:10 avalonea-FS kernel: <IRQ> [<c1003566>] ? do_IRQ+0x37/0x90 Nov 30 17:41:10 avalonea-FS kernel: [<c130c669>] ? common_interrupt+0x29/0x30 Nov 30 17:41:10 avalonea-FS kernel: [<c11ddd89>] ? acpi_idle_enter_bm+0x22a/0x25e Nov 30 17:41:10 avalonea-FS kernel: [<c12734f0>] ? cpuidle_idle_call+0x75/0xbd Nov 30 17:41:10 avalonea-FS kernel: [<c1001a5f>] ? cpu_idle+0x39/0x5a Nov 30 17:41:10 avalonea-FS kernel: [<c12fbd40>] ? rest_init+0x58/0x5a Nov 30 17:41:10 avalonea-FS kernel: [<c145172d>] ? start_kernel+0x28c/0x291 Nov 30 17:41:10 avalonea-FS kernel: [<c14510b0>] ? i386_start_kernel+0xb0/0xb7 Nov 30 17:41:10 avalonea-FS kernel: handlers: Nov 30 17:41:10 avalonea-FS kernel: [<f84e4815>] sil_interrupt Nov 30 17:41:10 avalonea-FS kernel: Disabling IRQ #17 Quote Link to comment
avalonea Posted November 30, 2011 Author Share Posted November 30, 2011 Seems I can get this to happen, by spinning down the drives, and then spinning them up again. Quote Link to comment
prostuff1 Posted November 30, 2011 Share Posted November 30, 2011 can we get a complete hardware breakdown please? and also the output of: cat /proc/interrupts Quote Link to comment
ubuntuaddicted Posted November 30, 2011 Share Posted November 30, 2011 Almost sounds like a hardware problem or bios passing off hardware to unRAID problem, Tapatalk is tha shizzle Quote Link to comment
avalonea Posted November 30, 2011 Author Share Posted November 30, 2011 Morning! So good news/Bad news scenario. Since my server was stupidly unstable. I have yanked the SATA controller ( sil 3114 on the compatible hardware list) , and setup the raid again, minus the drives that were on the controller, Lucky for me no data had made it onto any of them yet. This problem does seem localised to the sil 3114, Although to help rule it out, I'm going to put this into a friends unraid server, and see if we can get it to fault. Im also going to use his super micro 8 port card in my server, to see if i can get it to fail. As I get what was said above, that it should be a hardware ( say mobo/chipset ) error being masked by something else, I would like to single it out as the card if it is the problem. On that, with a quick look on google, im not the only one that's had issues with the card. Sometimes a bios update to the motherboard or card fixed it, sometimes it didn't. For me, the updated firmware on the sata device, only made problems worse! not better. and the legacy firmware wouldnt allow my system to boot ( although it was from 2003 or something.. So im not surprised.) I appreciate the assistance! Quote Link to comment
RobJ Posted December 9, 2011 Share Posted December 9, 2011 It does sound like buggy firmware on the card. Quote Link to comment
Hoopster Posted January 13, 2012 Share Posted January 13, 2012 I have this exact issue on a Biostar motherboard and a PCIe SATA controller with the Marvell 88XS7042 chipset. I have updated the Biostar BIOS to the latest. There is no BIOS update for the controller (or even any way to update the BIOS on this controller). In my case it is IRQ #16. It gets disabled right after boot and happens to be the IRQ assigned to the SATA controller. The disks on that controller suffer horrible performance degradation as a result. http://lime-technology.com/forum/index.php?topic=17823.0 Did you ever resolve this issue avalonlea? 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.