s_mason16

Did you ever discover a solution. I have the same error

yes, rc1 was stable for me

yes, I didn't have any issues before the update or before updating to rc1

about a week ago I updated to 6.10.0-rc2 and it has been unstable everyday since. I had even rolled back to rc1, but still am forced to do a hard reset nearly daily. after the initial upgrade to rc2, fix common issues warned me of corrupt config on flash drive. Attached is my diag run. any help is greatly appreciated. unraid-m-diagnostics-20220310-1446.zip

I have ran unraid for over a year no major issues on a dual x5650 system. Today I upgraded to the popular Gigabyte GA-7pesh2 and dual e5-2670 v2. Using same ram that never gave me an issue before. I didn't do anything special before the swap and i tested the system on the table first and then installed in a 24bay SM 4u chassis. I then just put my unraid flash drive in and everything seemed fine until about 10 minutes. and the system keeps crashing/timing out and then rebooting. I was on 6.6.0rc4 and upgraded to 6.6.6 just incase after the upgrade. no change in behavior. Watching the temps of the cpus they max out at low 40s. and after the os update i noticed the cpus stayed pegged at 100% for several minutes. did not notice that before the update. Ive ran FCP and it reports back with nothing. attached are the logs. Any ideas would be greatly welcomed. Thanks FCPsyslog_tail.txt unraid-m-diagnostics-20190111-2229.zip

So i did a fresh install and used xfs for my single cashe setup, based on squids reccomendation. Am slowly installing all my dockers and plugins. Get plex started and it has the metadata downloaded. But my cache drive says its only using 2gigs. With xfs does the docker image and appdata no longer live on the cache and moves to the array? Before having issues i was at about 80gigs used after invoking the mover. Im not opposed to having the full docker image move to the array when not in use, but for the sake of performance are my plex clients going to notice the hit loading metadata and normal operations?

thank you for looking through that. That's what fcp was telling me. tbh on nearly the best case im just hoping it was an ssd failure. because then just a replacement ssd and at this point ill probably just start with a fresh install either way. (poor sonarr/radarr/sabnzb settings wahh). but that doesn't exactly explain the smart test 199 crc error on drive 0 and 1. but if its not the ssd then it's a long track down the issue game. or a malfunction with a different hardware piece. or if it was just a software/os error, the fresh install will resolve it, but will my configuration just make it happen again?

my bad, thought that's what logs was, air head moment. unraid-m-diagnostics-20180316-2316.zip

I have this and I have internet access, and the server has two cables running to it. both showing activity. is there a way to fix the drive issue?

Yesterday(march 15), I was remot'ing in because I had a request to download a TV show. Get in no problem everything seems normal go to dockers and launch sonarr webui. it was having a problem load the show's data at first but its not a popular show so i thought maybe it just wasn't uploaded. click to add series and go to download s1, it's still having issues loading show info. won't dl shows. I get out and everything is weird. try to restart the docker and get a execution fail. other stuff is odd, so I run CA fix common problems, it says something is full and that's some error logging should be shut off, gives me the link to do so. and tells me to run the appdata clean up plugin. I run it and without looking (my mistake) delete my plex appdata file but haven't realized what ive done yet. thinking maybe the remote connection is why things are being wonky. and restart the server. and go home and try to do it. Find out I deleted plex, start a new one. downloads all the movie art and meta data fine. CA fix common problems. gives me a warning about having more than one key in the config, which it never has before. the CA apps page is acting weird and says it doesn't have internet connection. and several other little issues. Finally try to start downloading the tv show and sabnzb pauses because it says the disk is full. the cache drive is sitting at 53gb of 118gb and I go into maintenance mode and run a check disk and it is on readonly mode. and the parity drive and disk 1 have a smart issue but say it could be lack of power. Is the ssd in some sort of safe mode or did I just have 3 drives fail at the same time? is the ssd fixable, bad or do i need to just start over with a fresh install of unraid?

ohh so are you not actually selling the case, just the fan and accessories?

when you say you turned it into a hot swap case what do you mean(like added a backplane)? also do you have pictures of it assembled. I know the case, but not with the mods you have done. and if you wouldn't mind just listing your zip code so I can get an idea at shipping costs. Thanks in advance.

yes to windows 10, but i don't think it's the fall update that did it, because I had the update before i started really using my unraid machine and it was working a month ago and I did the 6.4 upgrade about 3 weeks ago. edit, but hey what do you know, smb1 was enabled. so i disabled it and re-enabled it and it's showing it again. Thanks Hoopster

it just kind of happened. I'm guessing after 6.4 update maybe. but the unraid system doesn't show up in network attached computers in windows. I can still access it with it's ip but i don't have a friendly icon to click on. I made sure smb sharing was on and that they are on the same workgroup. also that on the windows pc that network discovery and sharing are turned on. any ideas?

just used a very strong press wash the cm logo came off in a few swipes. then printed on a clear vinyl. big enough to cover the entire silver plate..laid it over and pressed down just enough it wouldn't slide then used a razer blade and cut off the extra and then pressed as many air bubbles as i could out.

I used this logo on a drive cage. Hope that's fine. It's my favorite of the submitted logo ideas so far. Here it is.

Got it! and removed that awful uncentered CM logo and added this and blew out and cleaned it.. Thanks a lot for an awesome sale and transaction! Can't wait to install it. PS.. to any other buyers tdallen is a treat to do business with.

From what i've seen the archive drives are slow af. so on paper and parity checks you would see a nice bump in speed. but besides the parity check (which is dependent of read speed) i bet less than often would you notice the performance improvement yourself in real world tasks. Edit, and if youre using an ssd for cache you should never be waiting for data to be written to your drives, just the cache. which would be 2 to 3 times faster than the ironwolf drives.

tdallen, I'll buy it and at your asking price. this will be perfect for me till i can get my hands on a nice rackmount chassis.

you could do a dual e5-2680 v2 build for under $800 and get a passmark score of 22k. dual 4114 may produce passmark score of 27k and just the cpu's alone will cost $1400. reddit build guide for suggested build

hmm, i tried that and it didn't change anything. I guess i'll just save whats in the folder for plexpy and try reinstalling it and see if i can dump the past logs into it. Thanks for the suggestion. I hope I don't have to do that for all of the dockers. lol.

I installed a new sas controller card, before hand all drives were plugged into the mainboard. I cross-flashed a dell perc h310 to to lsi 9211 it mode. plugged all my drives into it and booted up. Well it didn't like my ssd cache drive and rejected it and put it in unassigned devices. so i plugged that back into the mainboard and it was recognized again. But now some of my dockers are not working properly, specifically plexpy and the tautilla app, could be more those are just the two ive noticed so far. I'm afraid it created a new appdata directory and files when it auto started the array after the first reboot. How or what do i do to fix the dockers? Thanks in advanced.

This is a VERY basic explanation and might be wrong in some parts just to simply describe how it works or what might happen with data that is accessed often it moves a lot, and when it goes through different media,(hard drive to ram and back) it starts moving at different speeds and can get get jittered (jitter is measuring time difference in packet inter-arrival time) this can cause corruption in your data eventually making the data unable to be read, which is the exact opposite of what we want our file server to do. So ecc ram will check and fix the data and make sure it leaves in correct order before it sends it to its destination. in basic terms ecc memory is like insurance. just an extra step to maintain data integrity.

I'm not sure that motherboard even supports ecc, i looked at the product page and its doesn't say it does anywhere.

Hey m4r0c4s, I'm glad you're having great success with unraid, im a user of about the same time as you with similar build and have been loving it. I don't see any issues with what you're trying to accomplish, I would say to test it out first and then if you see fit add more memory, VMs love ram, even though 24 is a good amount. two vm workstations might eat that up depending on what programs you are running on them. Also if data transfers speeds to the network and internet are highly important, you may consider assigning them each their own ethernet plug rather than making it a virtual one on the vm.