iarp

Re-reading the quickstart, Remote Access to LAN is just server and LAN. I want server, lan, wan.

Before reading below: I wrote the below information before coming to a realization just now. Re-reading the quickstart, Remote tunneled access does NOT seem to be granting LAN access. I'm wanting my clients access to LAN and internet tunneling, but none of the dropdown selections seem to offer this. If that is the case then I'll drop this post and the issue because then nothing is wrong by unraids standards. ---- After a bit more testing I've come to realize I cannot access the LAN using Remote Tunneled Access. The reason I couldn't access websites is because DNS wasn't passing through to 192.168.2.1. When i updated the client to 8.8.8.8 it worked for domain names. Still unable to access LAN. As per MainFreezer's recommendation, adding the vhost0 to PostUp and PostDown allows LAN access. PostUp=iptables -t nat -A POSTROUTING -s 10.253.2.0/24 -o vhost0 -j MASQUERADE PostDown=iptables -t nat -D POSTROUTING -s 10.253.2.0/24 -o vhost0 -j MASQUERADE This fixed everything previously because I was still using my routers dns which running the above allowed access to the LAN. Aside from those two entries, I have not modified anything else.

You're just modifying the iptables, which I can do in console.

I ran the following, no luck. iptables -t nat -A POSTROUTING -s 10.253.0.0/24 -o eth0 -j MASQUERADE Just tried wg0 instead of eth0 as well.

I've been banging my head against a wall here for days now only to figure this out just now. If I disable docker from starting and restart my machine, wireguard clients set to Remote tunneled access can connect and the connection goes through the server just fine. However once i enable docker, the connection dies. We can still access the internal server itself but no LAN or WAN access. Unraid 6.12.4 eth0 bonding/bridging = No. There is an eth1 but its unused. storage-diagnostics-20230927-1014.zip

Excessive reading and writing of a usb stick isn't an issue?

I upgraded about a week ago, I'm already setup with dual nic and docker is on its own so I wasn't worried about the macvlan issue. I decided to redo my cache drive as zfs and lost my docker in doing so. Reconfigured and reinstalled everything and I'm back in business. That was two days ago now. I've just noticed that the reads on flash are super high, like in the millions and almost a constant 1.5MB/s with jumps to 5MB/s. I went into terminal and started inotifywait --timefmt %c --format '%T %_e %w %f' -mr /boot and the following is the output Sun Aug 20 13:23:01 2023 ACCESS /boot/ bzfirmware Sun Aug 20 13:23:01 2023 ACCESS /boot/ bzfirmware Above repeats 80 more times... Sun Aug 20 13:23:01 2023 ACCESS /boot/ bzfirmware Sun Aug 20 13:23:01 2023 ACCESS /boot/ bzfirmware Sun Aug 20 13:23:01 2023 ACCESS /boot/ bzfirmware Sun Aug 20 13:23:01 2023 ACCESS /boot/config/plugins/dynamix/ dynamix.cfg Sun Aug 20 13:23:01 2023 ACCESS /boot/config/ docker.cfg Sun Aug 20 13:23:01 2023 ACCESS /boot/config/plugins/dynamix/ monitor.ini Sun Aug 20 13:23:02 2023 ACCESS /boot/config/plugins/dynamix.my.servers/ myservers.cfg Sun Aug 20 13:23:02 2023 ACCESS_ISDIR /boot/ config Sun Aug 20 13:23:02 2023 ACCESS_ISDIR /boot/config/ Sun Aug 20 13:23:02 2023 ACCESS /boot/config/ Pro.key Sun Aug 20 13:23:02 2023 ACCESS /boot/config/ Pro1.key Sun Aug 20 13:23:02 2023 ACCESS_ISDIR /boot/ config Sun Aug 20 13:23:02 2023 ACCESS_ISDIR /boot/config/ Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ CommunityApplicationsAppdataBackup.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ FRS.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ appdata.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ archived.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ backup.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ clonezilla.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ development.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ docker.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ documents.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ downloads.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ family.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ games.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ gitea.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ iansdocs.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ music.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ recordings.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ system.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ temp_data.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ tvshows.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ veeam.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ video.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ webserver_media.cfg Sun Aug 20 13:23:02 2023 ACCESS /boot/config/shares/ youtube.cfg ^C That bzfirmware followed by reading of the share config files is over and over and over again every two seconds. I stopped docker, still happening. I uninstalled all my plugins and restarted, still happening. The Open Files plugin shows nothing using any of the files listed above. I stopped docker again and stopped the array, its still reading from flash constantly. So i decided to just start killing processes within htop that could've been related to anything to do with the boot drive. Killing emhttpd stopped the excessive readings. On a fresh boot or restart, the ACCESS /boot/config/shares/* and ACCESS config/pro.key are checked every second. But the bzfirmware is not constantly read. It's usually a few hours after startup that this issue starts.

I'm just curious about the dashboard display of System Memory stats. I have 16 GiB installed, 15.5 GiB Usable. The RAM tends to sit around 80% (12.4 GiB) and ZFS is almost always at 100% (2 GiB). Does the RAM bar include the 2 GIB used by ZFS, or is the RAM bar always going to show 2 GiB free but its actually used by ZFS?

If you're coming from google. I had to re-enable NFS globally, go to each share i wanted exclusive and set NFS export to No, save, change secondary storage from None to Array, save, change secondary storage back to None, save. Re-disable NFS globally. That did it. This system isn't realizing that NFS is disabled globally.

I'm unsure where else to put this post as no other forums seem to fit the bill. https://unraid.net/product/gamers https://unraid.net/product/data-storage-users https://unraid.net/product/multi-os-users https://unraid.net/product/digital-media-mavericks On those 4 pages you have links to lime-technology.com/forums/ which redirects to unraid.net/forums/ which breaks at this point with no redirection to forums.unraid.net. It will work as long as the person knows to replace unraid.net/forums/ with forums.unraid.net.

Parity check happened last night, I'm setup to get email, gotify, and browser based notifications. I noticed today that the email, gotify, and browser messages all state 0 errors. The Main tab and the History modal both states 2 errors. I'm more inclined to believe the Main and History over the notifications. storage-diagnostics-20230404-0908.zip

Setting disable_xconfig=true within /boot/config/plugins/nvidia-driver/settings.cfg solved this.

Unraid 6.11.5 Motherboard: ASUS P8Z68-V PRO GEN3 - Initial graphics adapter is set to iGPU - Legacy boot mode PCI GPU: NVIDIA Quadro P400 - installing driver latest: v525.78.01 with nvidia-driver plugin - intel_gpu_top is not installed. - /boot/config/modprobe.d/i915.conf exists as an empty file to enable intel drivers as per these instructions - I tried adding nomodeset to the boot append options I've tried a number of other settings that I've now forgotten. All I'm getting is a black screen with a solid, not blinking, cursor in the top left corner. I'm wanting unRAID on igpu leaving the p400 for jellyfin. What else can I do? It works if I remove nvidia-plugin and prevent the driver from installing.

It won't. The move script that mover uses is not standard, its custom written and seems to copy to .partial, deletes the original and then renames the partial to the real name. That renaming does not trigger inotifywait.

I've been trying to track down why so many NEW files do not have a hash. I get emails every single time due to either hash mismatch (usually on nextcloud.log despite excluding *.log files) or hash missing altogether. I dug into the source code of the plugin and replicated the inotifywait command so that i could watch for myself what was going on. inotifywait -mr -e close_write --format '%w%f' /mnt/disk1 /mnt/disk2 and after running mover here is what i see /mnt/disk2/documents/ubuntu-18.04.1-server-amd64.iso.partial /mnt/disk1/backup/unraid/STORAGE/flash/config/editor.cfg.partial /mnt/disk1/backup/unraid/STORAGE/flash/config/super.dat.CA_BACKUP.partial /mnt/disk1/backup/SQL/mariadb/vmosa/2022-11-12-08.00.01.sql.tgz.partial /mnt/disk1/backup/SQL/postgres11/airsonic/airsonic_2022-11-12-08.05.02.sql.tgz.partial /mnt/disk1/backup/SQL/postgres11/authelia/authelia_2022-11-12-08.05.03.sql.tgz.partial /mnt/disk1/backup/SQL/postgres11/family_photos_dev/family_photos_dev_2022-11-12-08.05.03.sql.tgz.partial /mnt/disk1/backup/SQL/postgres11/family_photos_prod/family_photos_prod_2022-11-12-08.05.04.sql.tgz.partial /mnt/disk1/backup/SQL/postgres11/film_convert_dev/film_convert_dev_2022-11-12-08.05.04.sql.tgz.partial /mnt/disk1/backup/SQL/postgres11/film_convert_prod/film_convert_prod_2022-11-12-08.05.05.sql.tgz.partial /mnt/disk1/backup/unraid/STORAGE/flash/config/plugins/dynamix.file.manager/dynamix.file.manager.txz.partial /mnt/disk1/backup/unraid/STORAGE/flash/config/plugins/dynamix.file.manager.plg.partial /mnt/disk1/backup/unraid/STORAGE/flash/config/plugins/unassigned.devices.plg.partial /mnt/disk1/backup/unraid/STORAGE/flash/config/plugins/dynamix.file.integrity/disks.ini.partial As you can see, every file moved by mover ends with .partial and if i run getfattr on the actual file (since there is no .partial file) root@storage:~# getfattr -d /mnt/disk1/backup/SQL/postgres11/airsonic/airsonic_2022-11-12-08.05.02.sql.tgz root@storage:~# When I navigate into the airsonic backup folder and run getfattr on any file in there (because backup always saves to cache and then mover moves) getfattr -d airsonic* is blank for all files.

6.11.1 with latest plugin versions installed. I'm attempting to preclear a 3TB drive. It's not going to be an array drive, it'll be more of a monthly backup rotation drive. I'm using the preclear as a test. At this point I've attempted to run preclear three (maybe four) times. If I trigger the clearing with the GUI with all default settings on the post-read I get: Nov 08 21:16:13 preclear_disk_WD-WCC7K6PJ0D5Z_786: Post-Read: post-read verification failed! Nov 08 21:16:17 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Error: Nov 08 21:16:17 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Nov 08 21:16:17 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: ATTRIBUTE INITIAL NOW STATUS Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Reallocated_Sector_Ct 0 0 - Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Power_On_Hours 42487 42498 Up 11 Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Temperature_Celsius 22 28 Up 6 Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Reallocated_Event_Count 0 0 - Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Current_Pending_Sector 0 0 - Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Offline_Uncorrectable 0 0 - Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: UDMA_CRC_Error_Count 0 0 - Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: S.M.A.R.T.: Nov 08 21:16:18 preclear_disk_WD-WCC7K6PJ0D5Z_786: error encountered, exiting ... I'm not see errors in syslog for the drive.

I wanted to added ".Recycle.Bin,backup/SQL" to the excluded folders and i accidentally hit my enter key. Doing so resets all settings to default.

I've been digging into various plugins already created for unraid to understand how the events system works. I would like to make use of stopping_docker but i'm trying to understand how to make sure the plugin event file executes before the dockerman event runs. Internally its using the find command and when I run that manually it runs alphabetically. I tried naming the plugin pre and post D to see and i didn't see much of a difference, output in syslog stayed the same log-position wise. Is there something else controlling the order at which events are triggered or is it actually alphabetical?

I went to restart my thunderbird container today as VNC wouldn't connect (it connected yeserday) and now i'm greeted with the following. Its looping to infinity at this time, these exact same lines over and over again. ---Ensuring UID: 99 matches user--- usermod: no changes ---Ensuring GID: 100 matches user--- usermod: no changes ---Setting umask to 000--- ---Checking for optional scripts--- ---No optional script found, continuing--- ---Checking configuration for noVNC--- Nothing to do, noVNC resizing set to default Nothing to do, noVNC qaulity set to default Nothing to do, noVNC compression set to default ---Taking ownership of data...--- ---Starting...--- ---Version Check--- ---Thunderbird not installed, installing--- ---Sucessfully downloaded Thunderbird--- ---Sucessfully downloaded Thunderbird--- ---Preparing Server--- ---Resolution check--- ---Checking for old logfiles--- ---Checking for old display lock files--- ---Starting TurboVNC server--- ---Starting TurboVNC server--- ---Starting Fluxbox--- ---Starting noVNC server--- ---Starting noVNC server--- WebSocket server settings: - Listen on :8080 - Web server. Web root: /usr/share/novnc - No SSL/TLS support (no cert file) - Backgrounding (daemon) ---Starting Thunderbird--- XPCOMGlueLoad error for file /thunderbird/libxul.so: libasound.so.2: cannot open shared object file: No such file or directory Couldn't load XPCOM. UPDATE: According to release notes, the now latest version 102 is not yet upgrade-able from 91 and prior. I set THUNDERBIRD_V to 91.11.0 and we're working once again.

Ok so this is curious, I was looking up grafana and unraid videos on youtube and while watching the video below I noticed that none of his ports and volume mappings are collapsing. The video is only a little over a year old at this point and I've been on the most recent versions as they released. Now I'm wondering if there is an option buried somewhere that disables the collapsing system.

readmore-js is used on the docker tab to allow collapsing of Port Mapping and Volume Mapping columns. I have 40 docker containers of which 8 have enough port mappings and 15 have enough volume mappings to require the readmore-js-toggle chevron-down to show up. (I believe) If the dom resizes in any way then readmore-js's resizeBoxes function is called which causes an immense amount of lag that progressively gets worse the more times the dom size changes. The first time I resize the dom the lag is only about 3-4 seconds. The second time is 11 seconds, the 3rd time is 21 seconds. Looking at firefoxs' profiling recording tool I'm unsure if the running samples means how many times this function was called, but for a simple resize of the browser (un-maximize and then maximized) it says samples for resizeBoxes is 21,856 having spent 21,906MS traced running time. 66% of the time is spent resizing boxes. It isn't just resizing of the dom either, if i leave the docker tab open and i'm focused elsewhere or I've left my computer for a bit, it'll crash the browser. On android if i try to zoom on my phone or tablet, it'll also lag horribly. My main profiling checking has been on Linux Mint with Firefox but this has been happening for a long time over various different machines including Windows 7 Firefox and Chrome, Windows 10 Firefox and Chrome, Linux Mint 20.3 Firefox, Brave, and Chromium, Android phone and tablet both chrome and firefox. The androids get so bad it usually requires i kill the browser. Having now found the profiling tool and realizing its attempting to resize these boxes, I used the following js to delete the readmore-js boxes and everything on the page runs so much smoother. document.querySelectorAll('.readmore-js-section').forEach(e => e.remove());

Have you checked what is in appdata and growing?

I was playing around with stacks in portainer and ran into an issue where it was recreating a container every 10 minutes (until I realized it was happening). I checked unraid and it had 16 randomly named containers stopped. Rather than go slowly through each one with unraids gui i went through portainer to delete the garbage containers. However ever since doing that, those 16 containers all still appear under the advanced settings section. I can't find them using any `docker` type command on the server itself. I'm not sure where its pulling the list from. There has been no backup of appdata since this occurred. So its not like its pulling the 16 extra containers from a backup file or something. Edit: its a runtime thing. unraid seems to store information about docker its containers in a json file rather than querying directly from docker itself. Anything that then uses the DockerClient php class pulls from this json file which'll return the containers that no longer exist. Once i restarted the server everything reset back to what containers actually exist rather than the missing ones.

This connection is direct using putty from windows. Aug 4 13:03:16 storage sshd[6898]: Close session: user root from 192.168.2.20 port 58866 id 0 Aug 4 13:03:16 storage sshd[6898]: Connection closed by 192.168.2.20 port 58866 Aug 4 13:03:16 storage sshd[6898]: pam_unix(sshd:session): session closed for user root Aug 4 13:03:16 storage sshd[6898]: Transferred: sent 2800, received 2416 bytes Aug 4 13:03:16 storage sshd[6898]: Closing connection to 192.168.2.20 port 58866

I only leave terminals in root's home directory /root (via a plain cd command), it's a habit after running into lock issues many years ago. This issue just requires logging into the server and stopping the array, the connection will be immediately disconnected.