August 29, 20178 yr I am not sure of the stability on Ryzen systems is going to be better, but its still freezing completely every 3 to 5 days, of course no log and diagnose info,. sometimes after a hard reset its going to start an parity sync (dirty shutdown). Most settings of the bios is on default, C state is On. i use this machine headless so no monitor and gpu... system: AMD Ryzen 7 1700 ASRock AB350M Pro4 (latest Bios) Corsair Vengeance LPX CMK16GX4M2B3000C15 2x Supermicro AOC-SAS2LP-MV8 Currently the server is online for about 2 days, 5 hours, 31 minutes without visible problems..
August 29, 20178 yr Yeah, I had my 1700 lock up this morning, usually every week or so for me. This is my combo Gigabyte Technology Co., Ltd. - AX370-Gaming K5-CF AMD Ryzen 7 1700 Eight-Core @ 3000 Nvidia 1050 TI passed through to a Windows VM
August 29, 20178 yr The folks with AMD Ryzen problems might want to read this: https://www.extremetech.com/computing/254750-amd-replaces-ryzen-cpus-users-affected-rare-linux-bug
August 29, 20178 yr Those with Ryzen builds, just checking to see if you have read the below thread? Prior to BIOS updates, the trick was to disable C-State. Early last week, I installed a Intel PRO/1000 dual port network adapter and I started experiencing random lock ups. After pulling my hair out, I took a chance and disabled C-State in the BIOS again and have been stable thus far.
August 30, 20178 yr only thing I have to report is all is going well but...... still no temps for my nvme drives.....
August 30, 20178 yr Did lent the gpu and monitor from my neighbour, disable global C states. power use as 78W, now its 120W+!! Hello energy bill! wil let it running for max. 5 day. when i got sometime i will dismantle the cooler to check the UA number... i wish i didnt sold the Intel system, stable as rock. edit: 120W is with al disk on, 78 is with almost every hdd off. the watt use is not onle the server, the cheap UPS use a lot itself. Edit2: did make some time to remove the cooler from it, the cpu is from week 16. so, a pre week 25. i will make contact with the reseller for a RMA (or make direct contact to AMD, but i am not from America) Edited August 30, 20178 yr by sjaak
August 30, 20178 yr 3 minutes ago, Can0nfan said: not sure if its a bug, but Plugins has lost its "check for updates" button By design If you have dynamix set within Notification Settings to check for updates, then whenever you go to the plugins tab it will automatically check for updates at that time. If you have dynamix set to never check for updates (ie: App Auto Update plugin configured), then the check for updates button will be there.
August 30, 20178 yr 48 minutes ago, Squid said: By design If you have dynamix set within Notification Settings to check for updates, then whenever you go to the plugins tab it will automatically check for updates at that time. If you have dynamix set to never check for updates (ie: App Auto Update plugin configured), then the check for updates button will be there. thanks
August 31, 20178 yr On 7/30/2017 at 9:39 AM, Living Legend said: Sorry for the lack of detail on this post, but I recently did an upgrade from 6.3.5 stable to 6.4.0-rc7a. The upgrade process went without a hitch. Here is what I've noticed so far: - Super snappy web GUI. Night and day between this and stable. - After the upgrade, I lost access to the web UI for a few dockers. Zoneminder, Headphones and MythTV are 3 I can remember losing access to. - My server is freezing up a few times a day. I can't access webUI or SSH in when this happens so I've had to do a hard reboot through IPMI. - After reverting back to stable, everything has resolved except I did not regain webUI access to all of the dockers that I had lost them to. - Still having some issues with VM GPU passthrough, specifically audio. Issues persist on both versions. Not sure if this is applicable here, but here is my diagnostics log: unraid-diagnostics-20170730-0935.zip Anyone else with the freezing issues mentioned above?
August 31, 20178 yr so notice my dockers apps really misbaving and vm was freezing a lot with 6.4 example, Plex was slow and sluggish at times sometimes not able to display anything and ended with a Database error. streaming would hiccup a lot probably due to this docker and my unifi one... my unifi controller was flacking out so bad it was causing my firewall to reboot every couple days. SabnzbD was constantly not downloading anything, unless i went to settings and did a connection test, then wait 5min then downloads would start. Deluge would constantly lose connection and need a restart to get back into the gui coping from location to another location on the same share through VM would freeze VM frequently and cause write errors. alas I downgraded back to 6.3.5 everything appears everything is snappy again,. looking very forward to stable 6.4 loved the new gui options, just need a less buggy system for my users right now.
August 31, 20178 yr 10 hours ago, Living Legend said: Anyone else with the freezing issues mentioned above? Have you made any networking changes to docker in the past? I've seen a few people with custom configs get stuck when upgrading to 6.4 as there are more networking features included in this release.
August 31, 20178 yr 5 hours ago, Tuftuf said: Have you made any networking changes to docker in the past? I've seen a few people with custom configs get stuck when upgrading to 6.4 as there are more networking features included in this release. What do you mean by networking changes to docker? Per container, I've made plenty of changes. But in regards to docker as an entity itself, I don't believe I've tinkered with any settings. And when I'm talking about freezing issues, this is an entire OS freeze where I can't even access my server through the command line. I've had to due numerous reboots through IPMI to get back and running. That's why I'm back down to 6.3.5 Edited August 31, 20178 yr by Living Legend
August 31, 20178 yr all my docker settings are default...although some directories and such were made through CLI so think im going to run Docker Safe perms see how that goes. as i am back to 6.3.5 and seeing issues although not as bad as 6.4... Edit*****turns out it could be my my firewall it seems to be rebooting every 10-12 hours and guess when its down and my controller (docker app) sees that it freaks out limiting my network. im going to go through logs with their support. since my wife power cycled the firewall its been pretty stable. we will see for how long Edited September 1, 20178 yr by Can0nfan
September 1, 20178 yr Been running 7a since it was released - no issues that I can tell of yet!! Need to get a second parity disk!
September 3, 20178 yr i decided after doing some maintenance on the hardware on my server that i'd try 6.4 rc7 again....since replacing my ethernet cables and applying new Thermal Paste to the CPU/Heatsink so far everything is running pretty good although the binhex/plexpass docker seems to drop once in a while giving me database errors (its sitting on my Raid 0 2x256GB SSD cache pool)
September 20, 20178 yr On 8/29/2017 at 1:28 PM, sjaak said: I am not sure of the stability on Ryzen systems is going to be better, but its still freezing completely every 3 to 5 days, of course no log and diagnose info,. sometimes after a hard reset its going to start an parity sync (dirty shutdown). Most settings of the bios is on default, C state is On. i use this machine headless so no monitor and gpu... system: AMD Ryzen 7 1700 ASRock AB350M Pro4 (latest Bios) Corsair Vengeance LPX CMK16GX4M2B3000C15 2x Supermicro AOC-SAS2LP-MV8 Currently the server is online for about 2 days, 5 hours, 31 minutes without visible problems.. I've been running an 1800X basically since release. I initially had the same issues that you all are talking about here. I am running an 1800X on an Asus Prime x370 board. I initially was using 2 sticks of 16GB Corsair Dominator Platinum RAM when I did the build. I was rebooting roughly every 2 days usually in the middle of the night. (Waking up to the server locked up) Since the OS runs on the RAM I took the 4 sticks of 8GB Vengeance LPX RAM on my main rig and swapped them with the dominator platinum. 0 issues with reboots since then on 6.3.5 or 6.4.0-rcq8
September 20, 20178 yr On 8/29/2017 at 2:28 PM, sjaak said: I am not sure of the stability on Ryzen systems is going to be better, but its still freezing completely every 3 to 5 days, of course no log and diagnose info,. sometimes after a hard reset its going to start an parity sync (dirty shutdown). Most settings of the bios is on default, C state is On. i use this machine headless so no monitor and gpu... system: AMD Ryzen 7 1700 ASRock AB350M Pro4 (latest Bios) Corsair Vengeance LPX CMK16GX4M2B3000C15 2x Supermicro AOC-SAS2LP-MV8 Currently the server is online for about 2 days, 5 hours, 31 minutes without visible problems.. Did you read through this post: https://forums.lime-technology.com/topic/55150-anybody-planning-a-ryzen-build/ Although a fix was introduced within c7a, I still had issues with the system going into a "sleep" state, so I simply disabled C-State, and it's been stable for 22 days, 1 hour and 2 mins thus far. Asus released another BIOS update last week, so I plan to run the update once a 4TB drive is delivered so I can do both at the same time. Plan is to enable C-State after the BIOS update and see what happens...
November 4, 20178 yr Not sure where to post this, but I just updated a couple of plugins as well as some dockers. I ran the Fix Common Programs plugin and it displayed the following error: Your server has run out of memory, and processes (potentially required) are being killed off. You should post your diagnostics and ask for assistance on the unRaid forums However, I'm seeing that the server shows plenty of free memory... any ideas? Edited November 4, 20178 yr by luisv
November 4, 20178 yr Diagnostics attached. Just upgraded and error is gone. davault-diagnostics-20171103-2053.zip
November 4, 20178 yr 10 hours ago, luisv said: Just upgraded and error is gone. Just rebooting would clear the error. Doesn't mean it won't come back.
November 4, 20178 yr 14 minutes ago, trurl said: Just rebooting would clear the error. Doesn't mean it won't come back. Understood... hopefully the diagnostic file I posted will be helpful.
November 4, 20178 yr The webUI needed memory for something, and linux decided to kill off the Plex Transcoder. Hard to say what was happening though on the 1st @ 3:08 am. Maybe the upgrade will change things. Its a wait and see thing now.
Archived
This topic is now archived and is closed to further replies.