UNRA1DUser Posted March 15 Share Posted March 15 (edited) Hi, I've been struggling with freezes again for a few weeks now. I have already tested the following: I bought a new PSU and connected it (FAILED) Removed powertop. Also every setting in the Go file. (FAILED) Removed Tips & Tweaks (FAILED) Removed my undervolting in the BIOS (FAILED) I ran Unraid in safe mode without any Plugins. Just the Dockers are running. VM is completly disabled (FAILED) I changed the USB-Stick from SanDisk Ultra Fit USB 3.2 Flash-Laufwerk 32 GB to Samsung BAR Plus USB-Stick Typ-A, 64 GB (FAILED) I also changed my MZHOU USB 3.1 zu 3.0 Adapter,Interner 19-Poliger Header Auf USB 3.1/Typ C 20-Poliger TYP-E - Frontplatten Adapter Unterstützungssysteme - WINXP/ 7/8 / 10 32 / 64BIT / MAC/Linux Adapter (FAILED) Also tried a different USB 2.0 Port with the Adapter (FAILED) Docker is set to IPVLAN (FAILED) I started a memtest86 and let it run for 21 hours. 5 passes without errors. 4x32GB ECC RAM - 128GB (Successfull) I have started a stress test for the CPU. It ran for just over an hour on all cores. CPU had 70 degrees. -> stress --cpu 20 (Successfull) I ran scrub on my Cache Pool. 2x SK hynix Gold P31 2TB PCIe NVMe Gen3 M.2 2280 in a ZFS-Encrypted - Mirror (Successfull) I ran SMART tests on both 12TB WD Red Plus WD120EFBX 256MB 3.5" (8.9cm) SATA 6Gb/s. (Successfull) I am out of Ideas. Currently I am running another memtest86 on all RAM. I think my server has been crashing ever since I started running the Palworld server. But I don't think it's because of this Docker or does anyone has the same behavior ? In that Topic you can see my BIOS settings (I removed the OC -> "Dynamic Vcore (DVID)" set to "auto" AND "PCH Core" set to "auto"): What else could I do? Thanks! tower-diagnostics-20240314-2102.ziptower-syslog-20240312-1118.ziptower-syslog-previous-20240312-1117.zip Edited March 15 by UNRA1DUser Quote Link to comment
JorgeB Posted March 15 Share Posted March 15 1 hour ago, UNRA1DUser said: I ran Unraid in safe mode without any Plugins. Just the Dockers are running. Did you also test with docker service disabled? That would be a good clue if issue is hardware related or not. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 31 minutes ago, JorgeB said: Did you also test with docker service disabled? That would be a good clue if issue is hardware related or not. So you mean disable docker service and start a Benchmark on CPU and RAM? Would that stress the CPU and RAM correctly? I have 128GB RAM and 20 threads. 128 GB = 128000 MB / 1024 MB = 125 stress --cpu 20 --vm 125 --vm-bytes 1024M Or should I also stress the Storage and IO ? Is that the right Syntax? stress --cpu 20 --vm 125 --vm-bytes 1024M –hdd 20 –io 20 Besides that, my last memtest86 was running 21 hours and passed 5 times. Is that enough? Should I stop my current second memtest86 run? It´s actually running for 15 hours and has 3 passes. Quote Link to comment
JonathanM Posted March 15 Share Posted March 15 24 minutes ago, UNRA1DUser said: So you mean disable docker service and start a Benchmark on CPU and RAM? No, just disable the docker service and run the server normally to see if it still freezes. 2 hours ago, UNRA1DUser said: I think my server has been crashing ever since I started running the Palworld server. Does it freeze if that container is not running? Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 (edited) 24 minutes ago, JonathanM said: No, just disable the docker service and run the server normally to see if it still freezes. I can try that. Is there a problem if I stress test the Server in that mode? Otherwise there wouldn´t be any load to it. And how long should the Server run in that mode? 24 minutes ago, JonathanM said: Does it freeze if that container is not running? As I know than the Server is running and running without problems. But there is also nothing installed that is using a lot of Resources of the Server itself. Last time I started the Unraid Server I stopped the Palworld Server and it runs for 4-5 hours before I restartet it by myself and tried another test (Went to sport in the mean time and forgot to start the Palworld Server again). I don´t know anymore if this was in safe mode or not. As I remember correctly the freezes started after I installed Palworld. But it´s not my first time that the Server is freezing -> After I bought this "Adapter" and connected it to the internal USB2.0 Port those freezes were gone. But I think I also played around and removed some Dockers. So could be possible that also another Docker was the reason for that. So should I abort my current running memtest86 and start the unraid server normally without the Palworld Docker and see if its running? Or should I first boot Unraid into safe mode without Docker, VMs and Plugins and wait for some hours ? Edited March 15 by UNRA1DUser Quote Link to comment
JorgeB Posted March 15 Share Posted March 15 39 minutes ago, UNRA1DUser said: And how long should the Server run in that mode? Long enough to see if it no longer crashed, I would take the typical time it takes to crash and double it, while using the server as a NAS only. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 11 minutes ago, JorgeB said: Long enough to see if it no longer crashed, I would take the typical time it takes to crash and double it, while using the server as a NAS only. Than it should be something like 2-4 Hours. I will stop the memtest86 now and start the Unraid Server in safe mode without Docker, VM and Plugins. I will also start a stress test to get some load on it. Let´s see whats gonna happening. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 I closed the memtest86 now. 4 Passes again without Errors. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 (edited) I started the Unraid Server in normal mode and disabled VMs and Dockers and stopped the Array. I also started a stress test. It´s running since 17:35. So about 1 hour for now. Server is still running. CPU Temp 72 C and the cpu Mhz is @ 4899.996 Mhz stress --cpu 20 --vm 125 --vm-bytes 1024M I can just see a lot of Network drops. Is that Normal? Edited March 15 by UNRA1DUser Quote Link to comment
JorgeB Posted March 15 Share Posted March 15 10 minutes ago, UNRA1DUser said: I can just see a lot of Network drops. Is that Normal? It can be, I wouldn't worry about that for now. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 31 minutes ago, JorgeB said: It can be, I wouldn't worry about that for now. Can I also increase the NIC RX and TC Buffer? Do you have experience with that ? Quote Link to comment
JorgeB Posted March 15 Share Posted March 15 Not really, but i believe the Teaks and Tips plugin can help with that. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 8 minutes ago, JorgeB said: Not really, but i believe the Teaks and Tips plugin can help with that. I will give it a try after we fixed the freezes. can those freezes really come from just the palworld docker ? Is that possible ? Quote Link to comment
JorgeB Posted March 15 Share Posted March 15 Many times that I've seen a container crashing the server. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 41 minutes ago, JorgeB said: Many times that I've seen a container crashing the server. Do you also know the reason for that? I mean I really like to host a palworld server. And it seems to work for a lot of people. I can’t understand why one container is freezing a whole server. Quote Link to comment
UNRA1DUser Posted March 15 Author Share Posted March 15 (edited) Stress test and Server are still running. No Problems. Close to 3 hours run time. (21:25) I will let it run until tomorrow morning. By the way. The Server is using 190 W under Benchmark. stress --cpu 20 --vm 125 --vm-bytes 1024M Edited March 15 by UNRA1DUser Quote Link to comment
UNRA1DUser Posted March 16 Author Share Posted March 16 I stopped the Stress test yesterday 11:00 PM. So it rans about 4 hours without Errors / freezes. After that the Server was still running without VMs and Dockers until now. Without freezes. Uptime in Sum 8 hours. So it´s not a Hardware Issue. I will start the Docker again and Stop the Palworld Server. Quote Link to comment
JorgeB Posted March 16 Share Posted March 16 1 hour ago, tjsyl said: I have been having the same issue for the last 2 weeks. Since this thread is still active, and to avoid confusion, please start your own thread. 1 Quote Link to comment
MassimoMx Posted March 16 Share Posted March 16 On 3/15/2024 at 10:44 AM, UNRA1DUser said: I think my server has been crashing ever since I started running the Palworld server. interesting, that you mentioned the Palworld Server, I had similiar problems since I used that container, but now 6 days after resetting my unraid, and without installing the palworld server my unraid server hasn't crashed once. My post about this similiar issue here if you want more information: Quote Link to comment
UNRA1DUser Posted March 16 Author Share Posted March 16 51 minutes ago, MassimoMx said: interesting, that you mentioned the Palworld Server, I had similiar problems since I used that container, but now 6 days after resetting my unraid, and without installing the palworld server my unraid server hasn't crashed once. My post about this similiar issue here if you want more information: That´s really crazy. My Server didn´t freezed or restarted since I deleted the Palworld Docker. But I still don´t understand why Quote Link to comment
ich777 Posted March 17 Share Posted March 17 15 hours ago, UNRA1DUser said: My Server didn´t freezed or restarted since I deleted the Palworld Docker. But I still don´t understand why Did you let the container un the entire time? I would recommend that you run daily restart from the container, the container is running since day one on my server. Quote Link to comment
UNRA1DUser Posted March 17 Author Share Posted March 17 (edited) 4 hours ago, ich777 said: Did you let the container un the entire time? I would recommend that you run daily restart from the container, the container is running since day one on my server. Yes, All my Containers gets a restart every night via the AppData Backup Plugin. But if I run the Palworld Container my Server gets freezes after 1 - 4 hours. Sometimes also longer times. But I must say, I limited the Palworld Server to 32 GB. (--memory=32G) But we are not more than 4-5 Players at ones online. And Also if no one joins the Server Unraid freezed after 1 hour. Until now my server is running without any freezes or problems. I would wait some more days. If the server keeps running I will install the Palworld Docker again and see whats gonna happening. Edited March 17 by UNRA1DUser Quote Link to comment
ich777 Posted March 17 Share Posted March 17 58 minutes ago, UNRA1DUser said: But if I run the Palworld Container my Server gets freezes after 1 - 4 hours. I can't reproduce this, my container is now running for about 10 hours and it's working fine, I restart it every 24 hours: As you can see in the screenshot I'm running it with a 50GB cap and have no issues whatsoever. Are you sure that you're CPU or RAM is not overheating? I know about OOM errors and as a result crashing servers but it's different from what you are reporting, the power supply and everything in your server is also up to the task? Quote Link to comment
UNRA1DUser Posted March 17 Author Share Posted March 17 I tested also another PSU with 750 W. Did 2 memtest86 for several hours and also run stress for some hours on CPU and RAM. Everything works fine. CPU highest temps with Turbo mode enabled are 72 C. No freezes or crashes. I don´t know the temps from the RAM but I guess it would crash if the memtest86 is running and at least if the stress test for CPU and RAM is running. But everything was working without crashes. Quote Link to comment
UNRA1DUser Posted April 18 Author Share Posted April 18 (edited) Hi again, my server did run for several days without freezing. For 2 days my Server freezed again. Found something like this in my Logs. I also attached both log files. Is maybe one M.2. SSD broken? How could I test this? Or is my Mainboard broken? PCIe Bus Error? Thanks! tower-syslog-20240418-0624.zip tower-syslog-previous-20240418-0624.zip Edited April 18 by UNRA1DUser Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.