goldbondx

What is the status with noted error previously: 2026/05/19 10:14:33 [DEBUG] Default SQLite driver initialized 2026/05/19 10:14:33 [ERROR] unable to load config, waiting 5 seconds before exiting... 2026/05/19 10:14:38 [FATAL] error unmarshaling YAML data: [136:3] unknown field "defaultLandingPage" 133 | darkMode: true 134 | dateFormat: false 135 | debugOffice: false > 136 | defaultLandingPage: "" ^ 137 | deleteWithoutConfirming: false 138 | disableOfficePreviewExt: "" 139 | disableOnlyOfficeExt: .md .txt .pdf 140 | I came back to a previously work but upgraded version I hadn't used in a bit and received this. I reinstalled and have the same error. Update: Commented out offending lined relating to defaultLandingPage and appear to be up and running.

So, I see a Booklore back on Github and Grimmory has been giving me some issues on upgrades like this evening again. (3.0) Error: Updated to 3.0.0 tonight on Unraid Docker and all library paths are showing as "inaccessible". Looked at the log files and this error comes up at restart: 2026-04-22T20:40:13.521-04:00 INFO 1 --- [booklore-api] [MessageBroker-1] o.s.w.s.c.WebSocketMessageBrokerStats : WebSocketSession[0 current WS(0)-HttpStream(0)-HttpPoll(0), 0 total, 0 closed abnormally (0 connect failure, 0 send limit, 0 transport error)], stompSubProtocol[processed CONNECT(0)-CONNECTED(0)-DISCONNECT(0)], stompBrokerRelay[null], inboundChannel[pool size = 0, active threads = 0, queued tasks = 0, completed tasks = 0], outboundChannel[pool size = 0, active threads = 0, queued tasks = 0, completed tasks = 0], sockJsScheduler[pool size = 2, active threads = 1, queued tasks = 0, completed tasks = 0] And, this error comes up attempting to access library/books: 2026-04-22T20:44:57.573-04:00 ERROR 1 --- [booklore-api] [ async-14] o.b.s.library.LibraryProcessingService : Library path not accessible: /books/fiction Are we changing back to Booklore again? Update: I don't know what is going on but it looks like I lost or didn't have a share to the books collection.

Would it be simpler to just chown on the image folder to allow writes given everything else seems to work? (at least that is what AI is suggesting) And if so, to what? The issue seems to be writing of book covers. (and other images) Update: I deleted the 2 entries you suggested I remove and now it seems to work..messed up. mind boggling but happy nonetheless..

I changed the repository from Booklore to Grimmory and all seems to work except I cannot write covers it seems to /image folder: 2026-04-02T06:56:07.299-04:00 WARN 1 --- [booklore-api] [mcat-handler-79] o.b.s.metadata.BookMetadataUpdater : Failed to download cover for book 22414: Error reading files from path: Failed to create directory: /app/data/images/22414 2026-04-02T06:56:11.129-04:00 INFO 1 --- [booklore-api] [MessageBroker-1] o.s.w.s.c.WebSocketMessageBrokerStats : WebSocketSession[3 current WS(3)-HttpStream(0)-HttpPoll(0), 3 total, 0 closed abnormally (0 connect failure, 0 send limit, 0 transport error)], stompSubProtocol[processed CONNECT(3)-CONNECTED(3)-DISCONNECT(0)], stompBrokerRelay[null], inboundChannel[thread-per-task], outboundChannel[thread-per-task], sockJsScheduler[pool size = 1, active threads = 1, queued tasks = 0, completed tasks = 0] I have added the group and user ID as instructed and performed the chown as instructed. Any help would be appreciated.

Thanks @bmartino1 . I am very hesitant to jump into scripts. However, this gave me some ideas where to start researching more. 1 comment I have as I continue to troubleshoot. GPU statistics are hammering 1-2 CPU cores (100%) like clockwork every 2-4 seconds. I verified this when I uninstalled gpu statistics plugin. The CPU usage went to <1 and stayed that way until I reinstalled and selected my RTX3070. Then the 1-2 CPU cores started bouncing to 100% again like clockwork. I have deselected my card in GPU statistics and seems to have stabilized. I will start up the array and monitor for several hours. It appears occurrences of [WARNING] [pool www] server reached max_children setting (50), consider raising it are indeed tied to load of Dashboard (specifically Main) in browser. Only occurrences thus far are tied to refresh or tab changes as I monitor real-time.

OK it has been 3 days after removing unassigned drive and unassigned drive plus with no max_children message that I can see. That said, still flooded with avahi_key_new() failed. Moving on.. Update: So, I started tackling the Avahi error and low and behold the max_children error came back. All I did was rename a share with a blank in it, and it seem to come up during some additional log info. This didn't exist before these last updates leading to 7.2.3. I don't know what is going on and not deep enough skills to fix them. What I do know, I am not the only 1 to report these problems.

I suspected that was the case and you indirectly confirmed that removal is only option. Thanks for the quick reply.

Removed my last post. I am not sure anymore on DelugeVPN as it seems this warning (max_children) popped up on every login as you suspected. It is somewhat inconsistent. One critical follow-up question. Plug-ins are remove only. There is no ability to "disable" as mentioned? I tried removing Unassigned Devices as a test and received this: Dec 29 11:40:41 mcunraid nginx: 2025/12/29 11:40:41 [error] 12364#12364: *109459 FastCGI sent in stderr: "Unable to open primary script: /usr/local/emhttp/plugins/unassigned.devices.preclear/include/Preclear.php (No such file or directory)" while reading response header from upstream, client: 1<Desktop Client IP>, server: , request: "POST /plugins/unassigned.devices.preclear/include/Preclear.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php-fpm.sock:", host: "<Unraid IP>", referrer: "https://<Unraid IP>/Plugins" ....... php-fpm[12253]: [WARNING] [pool www] server reached max_children setting (50), consider raising it I quickly re-added the plug-in successfully. I don't use NGINX at all. I access my server ext by VPN only. There is no reverse proxy. Hesitant to go any deeper as do not feel my skills follow your entire recommendations.

Thanks for the quick reply on. The USB issue was dealt with. External drive bay (5 TB) that is failing regularly. It's garbage and throwing it out. Unfortunately, the errors were still happening even after reboot it looks like. I'll keep an eye. The big issue trying to mitigate is item 1 (and 2) and thanks for clarifying root cause. I was unsure if GUI or something deeper. I wanted to confirm nothing that would impact data/writes. Item B seems like a good choice to start and eliminate polling culprits. I do have a bunch of dashboard elements as you mention. I'll be honest. My Linux skills are limited. I have never touched NGINX and hesitant to mess around with the pm.max_requests or slowlog. Will do a bit more research. My only remaining question. I haven't had any issues with this error up until the last 2 updates to OS. The only thing I have done is regular updates to Dockers and added FileBrowser Quantum. What could have introduced this?

Would appreciate any help here as getting similar/same error since 7.2.2. Only thing that has really changed in past couple of months: OS upgrade, FileBrowser Quantum installed, Regular Docker and Unraid Connect upgrades. Rebooted ~a day ago. I do have a few things going on but figured I would start with this error which is becoming more frequent.

Same.

7.2.2 and this just showed up: (4 times in recent log) - RTX 3070 / Nvidia Driver Version: 580.119.02 / GPU statistics is present/enabled. server reached max_children setting (50), consider raising it Seems tied to JF playback. Coincidentally (or not) was dealing with this issue just prior to: NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x22:0x51:884) This seemed to fix itself with a NVIDIA driver update and reboot. Is there some underlying video card/vram/driver issue I am missing that has cropped up recently?

I changed signup: true and reloaded the docker in order to allow new user sign-up. I confirmed showing in yaml. Doesn't seem to do anything. Am I missing something relating to signup? Update can you post the actual settings and full path syntax that are suggested? Do these settings look right? Seems to be a relative path with another config.yaml below data but not a complete list of settings:

I can confirm this is busted as of today's build. It is indicating the password is missing for Proton. (when it has not been changed for over 5 months) I'll try rolling back as well. https://github.com/qdm12/gluetun/issues/2992

I am receiving the same error. I searched and found this: https://github.com/Kareadita/Kavita/issues/2248 However, the problem doesnt' seem resolved on our build? (v0.7.13) Was your issue on PDF's? Did you find a fix?