mcai3db3

After removing the dirty cache drive, it hasn't crashed yet. Been about 4 days. Still evaluating. Gonna add a cache drive (a different drive) soon, see if issues arise then. Not touched voltages yet. Scared.

haha I think you're grossly overestimating my knowledge on this kinda thing! Right now the voltages will be whatever the default was. I'll give them a look tonight and let you know what they say.

My ram is Kingston HyperX 8GB (4 x 2GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) Desktop Memory Model KHX1600C9AD3K2/4G I'll take a look at the BIOS tonight and see what my settings are, I believe I already updated the BIOS when I first received the components. I'm not sure how easy it is to change all of this stuff though, will have to take a look. I'm apprehensive to change anything until it crashes again though, thanks for your help.

Ok thanks. Well for now I've removed my cache driving that was giving me grief. I've got sab/sick installed on a regular disk, which is far from ideal, but will do for a week or so. I'm basically just going to wait it out and see if it crashes again. At some point I'll open her up and dust off the various connection cables & make sure they're all kosher. Also, I have a spare 2TB drive, and one of my array disks is 1TB 7200, so I may just swap the array disk out for my 2TB and then make that disk my new cache. If my theory of crappy cache drive is actually anywhere near correct. Last note: I think my syslog that I posted wasn't helpful because it just showed the problems caused by the random reboot, rather than the cause of the reboot. This was why the user folder was not empty, because UnRAID had issues after the reboot, it did not create all the shares, only created one of them (and then it couldn't get rid of that 1 share when I tried to stop the array - thus it wasn't empty).

I've attached a screenshot of my unMenu main screen - I was confused by the sda and sda1 that exists there for the cache drive. Is this normal?

My entire Go script as requested: #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & /boot/unmenu/uu #sleep for 30 seconds sleep 30 #Start SABnzbd installpkg /mnt/cache/.custom/SABnzbdDependencies-2.1-i486-unRAID.tgz python /mnt/cache/.custom/sabnzbd/SABnzbd.py -f /mnt/cache/.custom/sabnzbd.ini -d -s 192.168.1.2:1066 python /mnt/cache/.custom/sickbeard/SickBeard.py --daemon cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c

Ok, first of all, here is a recent error I just got: Feb 10 17:24:19 SERVER kernel: irq 16: nobody cared (try booting with the "irqpoll" option) (Errors) Feb 10 17:24:19 SERVER kernel: Pid: 0, comm: swapper Not tainted 2.6.32.9-unRAID #8 (Errors) Feb 10 17:24:19 SERVER kernel: Call Trace: (Errors) Feb 10 17:24:19 SERVER kernel: [<c10451cf>] __report_bad_irq+0x2e/0x6f (Errors) Feb 10 17:24:19 SERVER kernel: [<c1045305>] note_interrupt+0xf5/0x13c (Errors) Feb 10 17:24:19 SERVER kernel: [<c1045a14>] handle_fasteoi_irq+0x5f/0x9d (Errors) Feb 10 17:24:19 SERVER kernel: [<c1004a82>] handle_irq+0x1a/0x24 (Errors) Feb 10 17:24:19 SERVER kernel: [<c1004285>] do_IRQ+0x40/0x96 (Errors) Feb 10 17:24:19 SERVER kernel: [<c1002f29>] common_interrupt+0x29/0x30 (Errors) Feb 10 17:24:19 SERVER kernel: [<c10085f9>] ? mwait_idle+0x4c/0x52 (Errors) Feb 10 17:24:19 SERVER kernel: [<c1001a14>] cpu_idle+0x3a/0x4e (Errors) Feb 10 17:24:19 SERVER kernel: [<c129c662>] start_secondary+0x195/0x19a (Errors) Feb 10 17:24:19 SERVER kernel: handlers: Feb 10 17:24:19 SERVER kernel: [<c11ead61>] (usb_hcd_irq+0x0/0x5b) (Drive related) Feb 10 17:24:19 SERVER kernel: [<c11ead61>] (usb_hcd_irq+0x0/0x5b) (Drive related) Feb 10 17:24:19 SERVER kernel: [<f8247022>] (ahci_interrupt+0x0/0x3df [ahci]) (Drive related) Feb 10 17:24:19 SERVER kernel: Disabling IRQ #16

First thing I noticed was that instead of 'installpkg' you have 'installpgk'...

Thanks for your help Joe. I'll take a look at this further this evening, but my first comments are: RE: Time sync - I have no clue what any of that means RE: Add-ons - off the top of my head I have sabnzbd, sickbeard and the add-on for printer drivers. I'm going to take my cache offline for now and thus remove sabnzbd and sickbeard. I will uninstall any other add-ons I have and see how it runs for a few days. The go script DOES have a 30 seconds wait before installing sab/sick. When I get home I'll post my exact goscript. It's worth noting - that syslog was taken after my system had done one of it's random reboots, at this point the system tries to parity check (I stopped it). At that point when you connect to the server most of the shares do not appear, so I tried to stop my array and start it again. It would not stop, then I powered down. This makes me think the syslog in question may have errors that would not normally appear in day-to-day use, and maybe a product of the shares not being available.

Towards the end it's me telling the system to stop the array and the cache drive not wanting to unmount, can't remember the exact steps but I just tried to shut down all sab/sickbeard etc and tried to close everything I could think of... it didn't help, I ended up just using powerdown in UnMENU. Also the cache drive temperature does not display in unRAID, so thats one of the errors (it just shows 0C - but it works fine in unMENU). sda is my cache btw.

Thanks for your help. Off to work right now so I'll continue on the testing tonight. I did post a syslog from unMENU in my OP, but I was confused as to how useful they'd be, given they begin at "Feb 9 18:27:50 SERVER syslogd 1.4.1: restart.".

Smart report on my cache drive - the only drive to return any errors. I'm suspicious of this drive.

7 4 WD20EARS Green Drives 1 500GB Cache Drive 1 random 1TB drive 1 2TB Seagate barracuda w/updated firmware parity drive I have actually added 1 or 2 of those drives since I first encountered this issue. I don't really know how to troubleshoot this. I don't have any spare components and really can't buy a random PSU/Mobo on the off-chance that the problem is being caused that way.

Oh jeez sorry I misread the post. It's a PSU CORSAIR|CMPSU-550VX 550W RT.

Ooohh, wait, capacitors I guess. I've not looked. I'd be most annoyed if it was, the thing is about a month old. I guess it's possible. Update: capacitors look fine.

It's a 3.06Ghz i3, I don't think it's underpowered... it's also not overclocked. Aaand I don't know what bad caps means

So my server keeps deciding to reboot itself. It's driving me mad. It has happened every few days and I really don't know why. Once it happened whilst streaming a video from the server, and on multiple occasions it has happened whilst I've been on the web management pages, messing around with settings etc. I ran a memtest and it came up with no issues, so I throw it out to you all now. What information should I upload? I have included a syslog but I doubt it's of any use. http://www.2shared.com/document/dkdSeI5G/syslog-2011-02-09.html The most recent crash was when I 'connected to server' from my Mac and it was spinning up one of my drives. My server is a 3.06Ghz i3, 8 gig ram, gigabyte GA-H57M-USB3, I have 7 drives in there - 1 500GB WD cache, my 2tb parity, 3 * 2tb WD20EARS and a random 1TB hard drive.

I bought one. I'm a HDD buying addict. I need help.

Are these still going strong for you? I just ordered two from eBay (which are going to be a week delayed due to Chinese New Year), and was hoping that my $16 has not been wasted I can only add 3 more drives in addition to the 7 mobo SATA ports, due to space issues, so this was a decent option given that I've got 2 PCI-E ports. If I find myself in the market for a humongous case any time soon I can always buy something fancier, it's not like the $16 has broken the bank.

Also bought me a Seagate Barracuda, now to check out if there are any WD20EARSesque special rules before I put this in a server! Altho, no worry of that anytime soon, given the lack of decent 8sata h55/7 lga1156 mobos.

Yea your right. My blindness continues today. The board layout is pretty similar though, maybe yours is based off of this one with some upgrades. Well I'll just need to do a bit more research on the asus boards available here. The closest thing I've found is the ASUS P7H57D-V EVO, but you're looking at $150+.

I figured, lol. If the difference between the CPUs is really only $12, then I agree, the 540 is likely worth it. However, does the 540 have the same power saving capabilities as the 530? I don't know off hand. I do know that the 530 idles at about the same rate as an Atom, which is most impressive. I'm no expert, but my googling seems to show the 530 and 540 as comparable in the power consumption stakes. Here are some pretty pictures stolen from elsewhere... Anyway... I feel I've digressed from the topic at hand...

It's going on my spreadsheet too, and I'm going to make it my goal in life to get exactly what you're getting on your new server, but just a little bit better. You get the 530, I'm getting the 540. You get the 690II, I'm getting the 690II Advanced, 4GB of Ram? 8GB of ram. Mmmm no financial contraints. My logic for getting the slightly faster i3 was simply just because I wanted to make my server have the longest lifespan attainable, so every little bit of processing power is welcome, assuming it's not going to break the bank. It's only about a $12 difference, so I don't really see any harm in it. p.s. I do actually know UTT IIRL, so I'm not just on a vendetta against a random internet foe.