Several drives failing, webui locking up

[astartz]

I am running 6.1.7 and for the last week i have had unraid UI keep locking up. Loading a page will take 10 seconds some times and a second later it will never load and just time out.

When this happens the Dockers i have running keep working so its just the main UI that has this issue.

 

Unraid also keep reporting bad drives and errors.

I have replaced 2 drives due to unraid saying they are bad, once the data rebuild completes everything will work for a hour or so and then unraid sees more drives as bad. I did a clean shutdown, rebooted and now most drives are bad, reboot all are back to working, except for one, that is listed a unmountable.

 

I attached my sys log and i have SMART reports ( 8 drives to many to attach all the logs),  does anyone have any ideas on what is happening? 

 

Thanks,

 

 

August

syslog.txt

[wreck]

Probably best to attach a full diagnostics report (tools -> diagnostics). I would also suggest running memtest for 24+ hours to eliminate that as an issue. What PSU do you have? How old is it?

[astartz]

The power supply is a BFG 450 watt.

I can get the full model if needed its just hard to read it where the server is right now.

I thought it was a higher watt model, i would think 450 is enough for 8 hard drive but i would love for that to be the root cause.

 

Here is the full hardware info for the server from the UI:

Model: Custom

M/B: Gigabyte Technology Co., Ltd. - P55A-UD4P

CPU: Intel® Core™ i7 CPU @ 2933

HVM: Enabled

IOMMU: Disabled

Cache: 64 kB, 8192 kB

Memory: 8192 MB (max. installable capacity 32 GB)

Network: eth0: 1000Mb/s - Full Duplex

Kernel: Linux 4.1.15-unRAID x86_64

OpenSSL: 1.0.1q

Uptime:0 days, 00:51:36

 

I attached a full diagnostics report.

 

xbmc-diagnostics-20160123-2352.zip

[Squid]

The power supply is a BFG 450 watt.

I can get the full model if needed its just hard to read it where the server is right now.

I thought it was a higher watt model, i would think 450 is enough for 8 hard drive but i would love for that to be the root cause.

 

Here is the full hardware info for the server from the UI:

Model: Custom

M/B: Gigabyte Technology Co., Ltd. - P55A-UD4P

CPU: Intel® Core™ i7 CPU @ 2933

HVM: Enabled

IOMMU: Disabled

Cache: 64 kB, 8192 kB

Memory: 8192 MB (max. installable capacity 32 GB)

Network: eth0: 1000Mb/s - Full Duplex

Kernel: Linux 4.1.15-unRAID x86_64

OpenSSL: 1.0.1q

Uptime:0 days, 00:51:36

 

I attached a full diagnostics report.

Considering that 12v2 (18a) is strictly for the CPU, that means that 12v1 is handling everything else (motherboard, hard drives, any addon cards etc).  Since round figures a hard drive draws up to 2a while spinning up, that means that worst case your hard drives alone draw 16a against a rail that can supply up to 18a.  Add to that power supply aging, the fact that the BFG isn't exactly the best quality power supply to begin with, etc etc etc.

 

So, no your supply is not sufficient for 8 drives

 

[jonp]

Single rail PSU is highly recommended. It always surprises me how often folks go cheap on the PSU to save a few bucks. I'm sure it's because folks just don't understand the importance, but it is EXTREMELY important.