GoGa_M

Hi All, Sorry I always forget to post my results... Downgrading to 6.9.1 seems to have fixed all the crashing problems. Nothing more I can tell or have done to fix the problem

Welp, it crashed again today.... So all the things I did, did not help at all It crashed while I was not home, and I don't know the exact time of the crash. Last thing I'm gonna try is to downgrade it. See if that helps. Downgrading to 6.9.1. I know that version ran for 60+ days without problems. Anyone got any other ideas to try?

So I have been quite quiet in this thread for the past month(s), but I have been testing some things. I just forget to post my findings... So here we go: It seemed that the server would only crash, when I was taking a backup (using Duplicati or IperiusBackup). But ONLY (as far as I know) when I was backing up data on my cache drives in raid1. And it only crashed about every 2-3 backups. So some backups were running fine. So I stopped doing backup of the cache for a while, and it ran for 24+ days without crashing. But then it randomly crashed again sometime after... I tried to replace one of the SSD's in the Raid1. It did have a SMART error, but only due to old age. It was a old SSD anyways. After the new SSD was working, I tried running a backup, and it crashed again within 2 days... Now, I only have 6 SATA ports in my motherboard, but I need 7 in total (5 for HDD Array, 2 for SSD Cache). When I setup the raid1 cache, I purchased a "PCIe to 2x SATA port adapter" (StarTech.com 2 Port SATA 6 Gbps PCI Express SATA Controller Card), and connected one of the HDD drives to the PCIe card. both SSD's were connected to the motherboard. Yesterday I removed the PCIe card, and now only using one SSD Cache. To see if the PCIe card caused the crash somehow. I am not sure, but I THINK it all slowly started after I started using Raid1 cache. Ran a full backup of the cache during the night, and the server is still running fine. Ill give ti a couple of days, and run backup every night to see if it crashes again. Lets see what happens in a few days

Today it finally crashed again... Attached is dianostics, and also the syslog that have been running since last time it crashed. Is anyone able to analyze the syslog to see what makes it crash? Dont mind the log spam from 192.168.2.13. That is just a Lansweeper server scanning my Unraid. goga-unraid-diagnostics-20210903-1759.zip syslog

I am definitely not good with numbers.... All dockers combined are using about 6GB ram

Server has not crashed yet, but today I got a notification from Fix Common Problems, telling me "Out Of Memory errors detected on your server" Server has 32GB Ram installed. VM Running with 2GB All running docker containers's memory load added together equals about 61003 mb ram used. According to https://www.linuxatemyram.com, then my system should be fine (I think). root@GOGA-UNRAID:~# free -m total used free shared buff/cache available Mem: 31987 9016 835 900 22135 21905 Swap: 0 0 0 And running dmesg | grep oom-killer shows the following output. root@GOGA-UNRAID:~# dmesg | grep oom-killer [129937.393367] cluster-Cluster invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 [256112.858195] mongod invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0 Attached is a new Diagnostics, also containing the syslog. Hope anyone is able to help solve this problem goga-unraid-diagnostics-20210822-1832.zip

I've now enabled Syslog Mirror to flash. Let's see when it crashes again. If it crashes again...

In the past month or so, my Unraid has crashed about every 4-6 days. It has been in use for about 6ish months, and only had these crashes in the past month or so. WebUI does not respond to anything, and the only way to reboot it is by pulling the plug... It has always happened while I am at work, so I don't know a specific time it happens. I am running 1 VM. Windows Server 2019, with 2GB Ram, 1 CPU core I am also running the following Dockers: Jackett. Krusader. Radarr. Sonarr. CAdvisor. Bazarr. Syncthing. Netdata. Pihole-DoT-DoH. Plex. Unifi Controller (Limited to 2 GB ram). Tubesync Attached is a Diagnostics export. Hope someone has a solution or is able to help. goga-unraid-diagnostics-20210818-1657.zip