June 5, 20188 yr Had a strange problem with my new unRAID server yesterday. I have email alerts set up and received a chain of email alerts from the server saying "Cache disk is hot (46C)," followed by "Cache disk returned to normal temperature" about 5 times while I was at work. When I got home to investigate, my Dockers were completely unresponsive and my shares and PleX server weren't working at all, and I also noticed all my CPU cores and memory utilization were fluctuating between 95-100% usage. I did a simple reboot (which hurt my feelings because I had just reached 15 days uptime), and everything seems to running completely normally. Nothing has changed in regards to the room temperature where the server I located, or the usage. Any ideas what might have happened and how to possibly avoid this happening again? Specs are a Ryzen 5 2400G, Gigabyte GAB350M Gaming 3, 2x4GB GB DDR4-3000, 250GB Samsung 860 EVO (cache), and 5x4TB Seagate Ironwolf 5400RPM HDD, running unRAID 6.5.2
June 5, 20188 yr Community Expert 46C is normal and perfectly fine for an SSD, it should go even higher after heavy writes, you can change the thresholds just for that device by clicking on it on the main page.
June 5, 20188 yr Author 10 minutes ago, johnnie.black said: 46C is normal and perfectly fine for an SSD, it should go even higher after heavy writes, you can change the thresholds just for that device by clicking on it on the main page. I understand this, and have already changed the thresholds to be more appropriate temperatures. My question is I got these alerts, and when I investigated the server I had maxed out CPU and memory usage and completely unresponsive dockers. Is there some kind of connection here, any ideas on what might have happened and some practices I can implement to avoid this again? Edited June 5, 20188 yr by baconborn
June 5, 20188 yr Community Expert 1 minute ago, baconborn said: My question is I got these alerts, and when I investigated the server I had maxed out CPU and memory usage and completely unresponsive dockers Very likely unrelated.
June 5, 20188 yr Author 3 minutes ago, johnnie.black said: Very likely unrelated. Any ideas on why the CPU and memory usage might have been maxed out with completely unresponsive dockers? It had been running for 2 weeks without a hitch previously with no changes from me directly to any settings or load.
June 5, 20188 yr Community Expert Just now, baconborn said: Any ideas on why the CPU and memory usage might have been maxed out with completely unresponsive dockers? No, diagnostics might help, but likely a docker gone rogue.
June 5, 20188 yr Author 2 minutes ago, johnnie.black said: No, diagnostics might help, but likely a docker gone rogue. Can you explain a rogue docker? I did notice after rebooting the server that 'Fix Common Problems' had an alert about Krusader being depreciated for binhex's version. Could something like that result in the described symptoms?
June 5, 20188 yr Community Expert Just now, baconborn said: Could something like that result in the described symptoms? No idea, if it happens again grab and post the diagnostics, if nothing is visible you'll need to try and find out which one it is, e.g. disabling them one by one.
Archived
This topic is now archived and is closed to further replies.