Hi all,

We've been away on vaca for several days, and upon arriving home & attempting to upload the photos to the server I've discovered that something appears to have gone awry while we were gone.

1st I noticed that my shares appeared to have no data, so went to the dashboard to see what I could find out.

My disk 6 is currently marked as faulty, and the log is pretty much fully red. Unfortunately it looks like the log doesn't go back as far as it needs to?

 

I have not rebooted yet, and have attached the Diag zip.

version = 6.1.9

installed plugins = powerdown package 2.22, community applications, fix common problems, preclear disks, dynamix webgui, unraid server OS

 

Not sure if I should just try a rebuild and see how it goes?

Anything else I should gather up before rebooting?

 

thanks all!

tower-diagnostics-20160907-1055.zip

there was another log file that it looks like the diag collector didn't grab. I've attached here.

 

I've looked a bit into the "program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO" error and have seen mentions that maybe this could be PSU related?

Does that bear looking into?

 

Thanks!

syslog-20160807-132412.zip

Updating to add list of installed dockers:

MumbleServer

SABnzbd

Sonarr

4 disks drooped offline, looks like a controller/cable issue.

 

Disk3, 5 and 6 and the unassigned WD 3TB.

 

Powerdown, check cables/controller.

How is it that you had to "discover" this problem instead of already knowing about it from Notifications?

Thanks johnnie. I'm in the process of doing the rebuild on disk 6.

It looked like all the disks you mentioned were on the controller, so I'll keep an eye on that. Is there any way to determine if the controller itself is going bad? I'll go ahead and pick up another cable for the controller just in case.

 

Trurl,

as for notifications, I've just never used them. I'll look into setting them up, although I'm guessing they wouldn't have helped at all here since I was on vaca with the wife & kids and was basically offline for the week. Thanks for the tip though.

 

ms_1

Trurl,

as for notifications, I've just never used them. I'll look into setting them up, although I'm guessing they wouldn't have helped at all here since I was on vaca with the wife & kids and was basically offline for the week. Thanks for the tip though.

 

ms_1

Notifications are one of the most important new features of V6 but not hyped as much as dockers and VMs unfortunately. Probably because they are so easy to set up so don't need a lot of forum discussion.

 

There are recent examples of V5 users who just set it and forget it only to "discover" that one disk had already failed some time ago, unRAID had been emulating that disk for all that time, and then they "discovered" they couldn't access some of their data when another disk failed. Two failed disks means none can be rebuilt of course.

Thanks trurl, I think I have notifications all set up now. The test went through OK

 

Looks like my data rebuild on disk 6 failed though along with read errors on disks 3 & 5.

Would the next step be to replace the controller cable, and then the controller itself if that didn't do the trick?

 

 

tower-diagnostics-20160907-1630.zip

Same 4 disks dropped offline, probably a bad controller, SASLP?

 

If don't have another controller or cable to test you can try using the other miniSAS port and a different PCIe slot if available.

Yep, SASLP. I actually did swap over to the other port when I was checking connections.

I've just swapped the controller over to the 2nd PCIe slot on the Z77 Pro3 & trying data rebuild again.

 

Assuming this doesn't fix the issue, I'll plan on picking up a new controller tomorrow.

I'll dig around the forums a bit, but is the SASLP-MV8 still a solid controller choice?

 

Thanks!

I think the SASLP is a great 4 port controller, because it's bandwidth challenged I don't recommend it if you need more than 4 ports (6 tops).

 

If you plan to add more disks in the future and if parity check/disk rebuild speeds are important for you, then I'd recommend a LSI based controller like the Dell H310 or IBM M1015, but these have to be flashed to IT mode to work with unRAID.