Stroker Posted August 27, 2018 Share Posted August 27, 2018 (edited) Hello Guys & Girls! I build by first Unraid Server a few Days ago and having quite alot of fun fiddeling around. All went fine for the last couple days but yesterday I configured some Times for Parity checks/Auto Plugin and Docker Update as well as Backups though "Backup/Restore Appdata" and "Auto Update Applications". Today I woke up to oddly CPU spikes inside the Stats. Every 10 Seconds there is a Spike of 30-50%. After a lot of fiddeling I found out that the Dynamix Folder Caching was to blame. I couldn't remeber that something like that appeared before but in my opinion I changed nothing that should affect the CPU in this ways. I didn't even change the settings for Folder Caching itself. Now I'm quite desperate. A Screen on how my CPU Usage looks the second I turn on the Folder Caching is attached. In the beginning you can see the initial scan. On the Second Screen you can see my Settings inside the Folder Caching App. Today I excluded some of my Shares but this helped little. only when I include just 1 or 2 the spikes apear smaller. The Update which arrived today added quite a lot of additional Settings but the Problem persists. I found another Post with the same Problem here in the Forum with the Tag Solved But sadly I cannot say why and how it was solved? It mentioned the Setting "Scan User Shares" in the post but this setting vanished in the new Version 2018.08.27a and was disabled from the beginning (as it is default). Can somebody give me a tipp how to troubleshoot this Problem futher? The Spikes dissapear the second I dissable Folder Caching but I quite liked the feature. Diagnostics are also attached. If any other Log-File helps just let me know. My System: Core i3-3240 8GB RAM Asrock B75 Pro3-M with 6 Attached HDDS (5x 3TB 1x 6TB) and 2x Cache SSDs (1x 64 & 1x 120 GB) Unraid 6.5.3 Dynamix Cache Directories 2018.08.27a all Docker Container & VM's are stopped Best Regards Stroker unraid-diagnostics-20180827-1938.zip Edited August 27, 2018 by Stroker adding System Data Quote Link to comment
bonienl Posted August 27, 2018 Share Posted August 27, 2018 An update to Dynamix Folder Caching was done, some time earlier today. This update includes as well enhancements made by @Alex R. Berg and allow better cpu and resource handling. Give it a try and see if it improves your situation. Quote Link to comment
Stroker Posted August 27, 2018 Author Share Posted August 27, 2018 (edited) Hey bonienl, Yes the new Version 2018.08.27a was released today but sadly this changed nothing on the cpu spike side. All the Screens you can see are with Version 2018.08.27a and with the Settings from the 3rd Screenshot Edited August 27, 2018 by Stroker Quote Link to comment
bonienl Posted August 27, 2018 Share Posted August 27, 2018 I can't really explain the spikes you are seeing. The new version allows to set a level depth, in case a folder structure is deeply nested, in this case it might help to limit the scan depth. Do you have any folders with lots of files and/or levels which are included? Try to exclude them, and see if that makes a difference. Quote Link to comment
Stroker Posted August 27, 2018 Author Share Posted August 27, 2018 (edited) Yes its possible that such directorys exist (Project Folders of Java or Android apps come here to mind for me) but nothing deeper than a NTFS Partition could handle. This Files and Directorys also existed before. Every Backup/Docker Folder or VM folder is already excluded. If I try to set the "Use Adaptive depth" to "Fixed" and set a relativly low level like "7" (everything else is empty so default) then I see a constant Load of 30-60% on my CPU for minutes after minutes. Not even the Timeouts seem to work for me. If I read the settings right the longest would be " Scan timeout for a scan during disk-idle period (sec): " but the default says it is only maximal 150 sec. At the Screenshot it ran for 5 Minutes straight. If I just add my Video share which has a max depth of ~8-10 with all settings back to default then the spikes are still there... I don't think that such a depth should be a problem in any way... On the end of this CPU activity the Disks spun down so no activity on these for the whole time right after Caching activation. Edit: Sorry but I just noticed. Should this Thread not better be in the plugin Support part of the Forum? If yes could it get moved? sorry for the mistake Edited August 27, 2018 by Stroker Question Move Quote Link to comment
Stroker Posted September 2, 2018 Author Share Posted September 2, 2018 Does anyone have an idea was is causing this behaviour? For now i have deactivated the Plugin but I quite liked the feature... Best Regards Quote Link to comment
drumstyx Posted October 12, 2018 Share Posted October 12, 2018 Just experimenting, but I excluded appdata and CPU usage seemed to decline a lot, as appdata was the heaviest of the searches (due to plex folder structure, I assume) It's not perfect, but along with the latest version, it seems to keep CPU usage down, and disk reads low. Quote Link to comment
Stroker Posted October 17, 2018 Author Share Posted October 17, 2018 Thanks for the tipp but sadly I already exclude all no data folders. (Backups, tmp, appdata,vms,....) but sadly the problem persists... even with the new unraid versions.... How much files does your log say you have when running the Folder Caching? I'm just wondering if I have just that many files... but in total I have just ~10TB and the bulk of it would be videos... Quote Link to comment
3doubled Posted April 10, 2019 Share Posted April 10, 2019 (edited) I know this is an old thread, but I just wanted to add that excluding the \appdata folder in Dynamix Cache Directories and setting the max level depth to 8 greatly reduced the CPU spiking behavior for me. Edited April 10, 2019 by 3doubled 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.