Hey everyone,

I've been running into a few issues in the last week with my unraid server - this issue has cropped up before and I did manage to resolve it. 

About a month ago I added an Nvidia 1660ti alongside a 710 and hadn't had any unusual issues.

 

Last week encountered a read error on a disk (caused by a glitchy vm) and when resolving that issue I had Unraid crash on the line:

"Starting haveged entropy daemon: /sbin/haveged" during startup. 

 

After reseating the RAM, rearranging the graphics cards what felt like 300 times, unraid booted happily and I resolved the read error by rebuilding the disk.

Fast forward to today where the server had been up and running smoothly for several days, until it unexpectedly crashed - and now presents the:

"Starting haveged entropy daemon: /sbin/haveged" error before crashing.  

 

I've attempted everything I can think of:

 - trying the graphics cards in every slot configuration

 - reseating RAM

 - creating a whole new unraid usb drive

 - resetting the BIOS

 - booting into safe mode

none of which has resolved the issue. 

 

I'm tearing my hair out at this point. I would really hope that both of my GPU's aren't fried - I'm able to get video output from both right up until the system shuts down. 

Thank in advance for any help!

 

Note: I have been able to intermittently boot successfully with the 710 in the primary slot and leaving the 1660ti disconnected - but that setup isn't much use to me and will fail on reboot.

 

Full hardware list below:

- Asus TUF GAMING X570-PLUS (WI-FI)

- Ryzen 7 3700x

- 32GB DDR4 3600 Memory

- 3x Seagate Ironwolf 4TB HDD

- 500GB WD M.2 Blue SSD

- MSI Geforce GTX 1660ti

- Gigabyte Geforce 710

- LG BH16NS55 BluRay Optical drive

- Running Unraid 6.8.3

Edited July 28, 20205 yr by Husker28

 

I've now updated the bios and set Power Supply Idle Control to typical current idle. But no change.

I can get it to boot with the 710 a little more reliably now (without the 1660ti) - managed to grab the diagnostics file. 

mainframe-diagnostics-20200728-2008.zip

Edited July 28, 20205 yr by Husker28

*Maybe* try the 6.9 Beta and see if the new kernel fixes this?  I would personally try this either with an earlier beta, before the multiple pools were added, or try it on a spare flash drive to test everything out (or create a backup of your current flash drive and then upgrade to beta)