March 7, 20179 yr Hey everyone, So problems have ensued with my unraid box since this issue. The public IP address issue with my box has been resolved, however, the random locking up has not. Last night the system had its worst crash, which I am looking to the unraid community for help in identifying the root cause. Here is what happened, bear with me, I apologize for the wall of text. The system had locked up yesterday so I was forced to shut it down uncleanly. I brought the system back online, disabled the smb shares and docker (to disable unnecessary network access) and let a parity check run overnight. However, the system crashed some time during the night, during I suspect the parity check. The terminal output from the crash is below. I was unable to pull diagnostics at this moment since the server was unresponsive. After the crash I turned off the box and turned it back on. Fans turned but no video, no LAN activity, nothing. Tried again, same result. I then powered off, unplugged everything, plugged it back in and it turned on. Bios stopped on the message below. I had to reset the BIOS to default settings, then set the BIOS to boot from the unraid USB, and finally it booted into unraid. Everything seemed normal in unraid at this point. All drives were showing available with no errors. Attached is a diagnostics zip file taken from the first successful reboot. fmrlserv-diagnostics-20170307-1208.zip To me it appeared that the server booted normally here, however the syslog does contain a new error I've never seen before: Mar 7 12:07:15 FMRLSERV kernel: blk_update_request: I/O error, dev fd0, sector 0 Mar 7 12:07:15 FMRLSERV kernel: floppy: error -5 while reading block 0 which the server spits out every 3 minutes. Did I re-enable a ghost floppy disk drive when I reset the BIOS and so I'm getting this error? The server is now 3% into a parity check with FCP running in troubleshooting mode. One other strange observation about the reboot process. Since the crash I've since rebooted the system a few times, and each time I am noticing a message regarding NVRAM (from the BIOS?) that I do not believe I have seen before. See the image below. I am not 100% sure on this since the message flashes very quickly (had to take a video in order to capture the still), but I am fairly certain the message is new. That all being said, I believe my problem is hardware related and I am looking to the community to nail this down. Here the system specs and my thoughts on what could be the problem unRAID 6.3.2 Motherboard: ASUS P5K/EPU CPU: Intel Core 2 Quad Q6600 RAM: 4GB DDR2 - when I first built this server these sticks passed memtest running for approx. 24 hrs. However, I am not ruling them out. I have a new set of sticks currently running memtest in another box which I can swap in if they pass. I plan on letting it run for 48 hrs Data drives: hodge-podge of 14 drives (including parity), mostly WD green and reds 2-4 TB in size. Cache: 240 GB SSD Power supply: Corsair CX 500 - my power supply may be under-powering the system. When I built the system I calculated 500 W would be about the bare minimum I could go, however I had this PSU lying around, I knew it was in good condition and from a reputable brand so I went with it. Do my symptoms indicate I have a power problem? I can swap in a larger power supply (600W), albeit from a less reputable brand (thermaltake). The server primarily serves files via smb, however it does also run 3 dockers: Nextcloud, letsencrypt and mariadb. I have these plugins installed: CA, FCP, Nerd tools, preclear, proftpd, and unassigned devices. Is there any test I can perform to identify if my issues lie with my motherboard, cpu or psu? Would running prime 95 on the system be helpful? Thanks Edited March 7, 20179 yr by SRB Fixed uploads
Archived
This topic is now archived and is closed to further replies.