January 3, 201016 yr Today while watching a movie I got a long pause. Didn't know why so I checked my unRaid menu and saw I had errors on a drive and the ball was red. I should have grabbed a report but I didn't. I shut the box down and proceeded to change the SATA cable w/ a fresh one. Then I powered the box back up and still had the red ball. I then ran a SMART test and the drive passed. No problems at all. So I was going to do the Trust my Array prodedure since I'm sure there is nothing wrong w/ the drive. I went back into the menu and clicked Restore. Everything went blue but my cache disk of course. I then telnet into the box and typed the command mdcmd set invalidslot 99 and got: cmdOper=set cmdResult=failed As of right now I have not done nothing else until I get a response on what to do next.
January 3, 201016 yr Author Actually I did everything right except I had a typo when I typed the original command "invalitslot" LOL Its running a parity check which won't be done until tomorrow morning. Started with 92 errors but I'm pretty sure that they will fix themselves and everything will be straight.
January 5, 201016 yr Author I'm getting errors now after running parity check twice. First I had 92 errors. I ran it again after that and got 0. So I assumed everything was ok. I've attached my syslog. http://filebin.ca/rxcoqz/syslog-2006-01-05.txt
January 13, 201016 yr Wow, that is one seriously troubled machine! Several serious problems, but the most important is something seriously wrong systemically with the machine that when instructed to spin multiple drives down, many or most of the SATA links go down! That is not supposed to happen(!), and because of other really odd behaviors, my first recommendation is to try another and better power supply. I can't think of any other way to explain what is happening. The motherboard also may be defective, as there is something wrong with IRQ19. It assigns some USB ports to IRQ19, then has to disable it. It then assigns it to your SiI3132 card, but has to disable it AGAIN (so any drives that might have been connected to it would have been disabled). It then assigns IRQ19 to your motherboard SATA ports, and although I don't see any further problems with IRQ19, I don't trust it either. It is possible that a Boot Code might produce better behavior, such as pci=nomsi or pci=routeirq or irqpoll. You also have a bad SATA cable (the most likely explanation) connected to sdo (WDC_WD10EADS-00L5B1_WD-WCAU49150622). I don't understand why unRAID tries over and over to spin all of the drives down. Each round of spin downs results in numerous exceptions, errors, resets, and speed limiting, and once finally resolved, starts all over again with another set of spin down commands, perhaps because unRAID detected that one or more of the drives did not respond correctly that it was spun down. These sequences of spin down commands and the resultant exceptions ultimately result in 4 of the drives being disabled. I think Tom may want to take a look at your syslog, not because he can fix your machine, but because I don't think unRAID responded correctly here, in a very unusual situation. It's also possible that your motherboard is not fully compatible with Linux, but there may be one or more of those boot codes that will enable it to run well enough.
Archived
This topic is now archived and is closed to further replies.