sasbro97

I have started several threads since update 7.2. I really don't know what this minor update did to my system, but for weeks now it's driving me crazy. So lately I got the problem that I can not stop or restart the server without crashing. It is always stuck on "Retry unmounting disk share". I have no idea why. The logs won't tell me. I guess I changed one setting somewhere, and since then, I've this problem. The previous solutions from other threads didn't work. I realized that the Docker daemon is already stopped, so that is not the reason. Let's get back to my Docker problem. Sometimes I restart the server, and Docker container won't find each other via name. They are in the same custom network but cannot be resolved. Like, for example, wordpress-db:3306. When I restart Docker via /etc/rc.d/rc.docker restart there is a good chance that it is fixed out of nowhere. But not always. Sometimes I restart several times, and it does not help. I have Docker installed on my cache drive, formatted with ZFS. So I installed Docker as folder as recommended. I also reinstalled Docker more than once. For a while it works but after one restart I got my problems back. I don't know what to do anymore. If the server is in a state where nearly everything works, I'm so scared to do anything. After one server restart, it was like this for Authentik or Nextcloud: INF | event=PostgreSQL connection failed, retrying... (connection failed: connection to server at "172.19.0.45", port 5432 failed: server closed the connection unexpectedly This probably means the server terminated abnormally before or while processing the request.) logger=authentik.lib.config timestamp=1768564550.3915546 So basically, the URL is correctly resolved because the IP is correct, but still, no connection can be created. My logs are attached. Also, my diagnostics. I already tried Docker with macvlan instead of ipvlan, but I think it made it even worse. Please help me. I mean, for sure, I run 150 containers and more, but before the update, it worked without any problems. syslog-192.168.178.29.log unraid-server-diagnostics-20260117-0246.zip

I want to do VM backups and I would like to monitor that they run with Healthchecks. So I want pre and post-scripts to run.

Hello, does anybody have a config that they could share with me for pre- or post-scripts of the Unraid VM Backup plugin? Unfortunately, there is a bug, and the repo is more or less unmaintained: https://github.com/JTok/unraid.vmbackup/issues/47 But if somebody already has some upload scripts and can show the structure of the folder with permissions, it would be awesome.

Is there any chance that we can fix the autostart selection for the FolderView in Docker? It has always been buggy: https://youtu.be/pRY9AEqXo_Q

Since I use Unraid, the Docker Autostart selection is buggy. As I reinstalled all apps now, I can finally show how buggy it is. It also happens if I just select some in a folder and refresh, and select some more in another folder. The other ones will be deselected. See this video: https://youtu.be/pRY9AEqXo_Q Is it caused by FolderView?

Okay I'm done. Everything went down again Now I got serious problems with this mDNS warning as it seems: Nov 24 20:23:27 UNRAID-Server kernel: e1000e 0000:00:1f.6 eth0: NIC Link is Down Nov 24 20:23:27 UNRAID-Server kernel: bond0: (slave eth0): link status definitely down, disabling slave Nov 24 20:23:27 UNRAID-Server kernel: e1000e 0000:00:1f.6 eth0: left promiscuous mode Nov 24 20:23:27 UNRAID-Server kernel: e1000e 0000:00:1f.6 eth0: left allmulticast mode Nov 24 20:23:27 UNRAID-Server kernel: bond0: now running without any active interface! Nov 24 20:23:27 UNRAID-Server kernel: br0: port 1(bond0) entered disabled state Nov 24 20:23:28 UNRAID-Server dhcpcd[1953]: br0: carrier lost Nov 24 20:23:28 UNRAID-Server avahi-daemon[39819]: Withdrawing address record for 192.168.178.29 on br0. Nov 24 20:23:28 UNRAID-Server avahi-daemon[39819]: Leaving mDNS multicast group on interface br0.IPv4 with address 192.168.178.29. Nov 24 20:23:28 UNRAID-Server dhcpcd[1953]: br0: deleting route to 192.168.178.0/24 Nov 24 20:23:28 UNRAID-Server avahi-daemon[39819]: Interface br0.IPv4 no longer relevant for mDNS. Nov 24 20:23:28 UNRAID-Server dhcpcd[1953]: br0: deleting default route via 192.168.178.1 Nov 24 20:23:28 UNRAID-Server dhcpcd-run-hooks: interface=br0, reason=EXPIRE, protocol=dhcp Nov 24 20:23:28 UNRAID-Server update_services: no queued job present Nov 24 20:23:28 UNRAID-Server update_services: queue new job 20, wait for 20s Nov 24 20:23:30 UNRAID-Server kernel: e1000e 0000:00:1f.6 eth0: Failed to enable ULP Nov 24 20:23:31 UNRAID-Server ntpd[2204]: Deleting 1 br0, [192.168.178.29]:123, stats: received=220, sent=220, dropped=0, active_time=9919 secs Nov 24 20:23:31 UNRAID-Server ntpd[2204]: 216.239.35.12 local addr 192.168.178.29 -> <null> Nov 24 20:23:31 UNRAID-Server ntpd[2204]: 216.239.35.8 local addr 192.168.178.29 -> <null> Nov 24 20:23:31 UNRAID-Server ntpd[2204]: 216.239.35.4 local addr 192.168.178.29 -> <null> Nov 24 20:23:31 UNRAID-Server ntpd[2204]: 216.239.35.0 local addr 192.168.178.29 -> <null> Nov 24 20:23:31 UNRAID-Server ntpd[2204]: Deleting 3 br0, [fd31:e9c1:819e::53]:123, stats: received=0, sent=0, dropped=0, active_time=9919 secs Nov 24 20:23:31 UNRAID-Server ntpd[2204]: Deleting 4 br0, [fe80::ca7f:54ff:fe50:de29%5]:123, stats: received=0, sent=0, dropped=0, active_time=9919 secs Nov 24 20:23:48 UNRAID-Server reload_services: execute queued job 20This killed my access to the server even though it is running. AdGuardHome uses br0 and runs as a DNS server, so it breaks everything. The whole logs are attached. Another horrible bug now is that I have clearly set auto-start for the invoiceninja-web container to false. Somehow, Unraid starts it at every reboot. I looked into cat /var/lib/docker/unraid-autostart and there isn't this container in the UI as well. Where does Unraid take this advice from to start it? I know the autostart has always been buggy, but that is so annoying... I can see from the logs that yesterday the warning started: Nov 23 19:49:37 UNRAID-Server avahi-daemon[58558]: *** WARNING: Detected another IPv4 mDNS stack running on this host. This makes mDNS unreliable and is thus not recommended. *** Nov 23 19:49:37 UNRAID-Server avahi-daemon[58558]: *** WARNING: Detected another IPv6 mDNS stack running on this host. This makes mDNS unreliable and is thus not recommended. *** syslog-192.168.178.29.log

@JorgeB added. Question: Would it be possible that a Docker Socket Proxy running as a Docker container is somehow the problem? I mean, it started after an update and worked before, but (for now) I changed Traefik back to use /var/run/docker.sock, and I don't see outages. It's still slow as hell, but no 500 error. But how could this affect Matomo's ability to connect to its database container? unraid-server-diagnostics-20251124-1940.zip

I cannot believe this. It started again yesterday. Now it's only one container behaving like this. It started like this when I had this issue as well. After rebooting Unraid, it seems to work flawlessly for at least an hour, but in general, something is wrong because I can see that calls to the service need between 2000 and 30000ms, and sometimes even timeout. In the container logs, I see: [Mon Nov 24 19:11:25.604359 2025] [php:notice] [pid 87:tid 87] [client 172.19.0.29:56560] [stats.xyz.de] Error in Matomo: Could not connect to the database: SQLSTATE[HY000] [2002] No route to host This may be a temporary issue, try refreshing the page. Please contact your Matomo administrator if the problem persists. 172.19.0.29 - - [24/Nov/2025:19:11:20 +0100] "GET / HTTP/1.1" 500 2278 "-" "Uptime-Kuma/2.0.2" 172.19.0.29 - - [24/Nov/2025:19:11:52 +0100] "GET /matomo.php HTTP/1.1" 200 502 "-" "Uptime-Kuma/1.23.17" 127.0.0.1 - - [24/Nov/2025:19:11:53 +0100] "GET /matomo.php HTTP/1.1" 200 609 "-" "curl/8.14.1" [Mon Nov 24 19:12:23.334519 2025] [php:notice] [pid 53:tid 53] [client 172.19.0.29:35966] [stats.xyz.de] Error in Matomo: Could not connect to the database: SQLSTATE[HY000] [2002] No route to host This may be a temporary issue, try refreshing the page. Please contact your Matomo administrator if the problem persists. 172.19.0.29 - - [24/Nov/2025:19:12:20 +0100] "GET / HTTP/1.1" 500 2278 "-" "Uptime-Kuma/2.0.2" 172.19.0.29 - - [24/Nov/2025:19:12:27 +0100] "GET /piwik.js HTTP/1.1" 200 22396 "https://xyz.de/" "Mozilla/5.0 (Linux; Android 10; K) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/142.0.0.0 Mobile Safari/537.36" 172.19.0.29 - - [24/Nov/2025:19:12:52 +0100] "GET /matomo.php HTTP/1.1" 200 502 "-" "Uptime-Kuma/1.23.17" 127.0.0.1 - - [24/Nov/2025:19:12:53 +0100] "GET /matomo.php HTTP/1.1" 200 609 "-" "curl/8.14.1" [Mon Nov 24 19:13:23.302342 2025] [php:notice] [pid 87:tid 87] [client 172.19.0.29:55620] [stats.xyz.de] Error in Matomo: Could not connect to the database: SQLSTATE[HY000] [2002] No route to host This may be a temporary issue, try refreshing the page. Please contact your Matomo administrator if the problem persists. 172.19.0.29 - - [24/Nov/2025:19:13:20 +0100] "GET / HTTP/1.1" 500 2278 "-" "Uptime-Kuma/2.0.2" 172.19.0.29 - - [24/Nov/2025:19:13:38 +0100] "GET /piwik.js HTTP/1.1" 200 22396 "https://xyz.de/" "Mozilla/5.0 (Linux; Android 10; K) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/142.0.0.0 Mobile Safari/537.36" 172.19.0.29 - - [24/Nov/2025:19:13:52 +0100] "GET /matomo.php HTTP/1.1" 200 502 "-" "Uptime-Kuma/1.23.17" 127.0.0.1 - - [24/Nov/2025:19:13:53 +0100] "GET /matomo.php HTTP/1.1" 200 609 "-" "curl/8.14.1"Before, you only saw 200er status codes. How is it possible that it starts the connection to the database container? The database container is running, and I can see no issue. This issue is driving me insane. It must have something to do with Docker networking. The more time the server is running, the more containers will become slowly when calling them. How can I debug this or find out what is going on? Syslogs do not really show any reason. Nothing is logged. Still, these logs are concerning me: Nov 24 17:41:30 UNRAID-Server sshd[2156]: Received signal 15; terminating. Nov 24 17:41:30 UNRAID-Server sshd[39256]: Server listening on fd7a:115c:a1e0::3801:2f78 port 1997. Nov 24 17:41:30 UNRAID-Server sshd[39256]: Server listening on 100.76.47.120 port 1997. Nov 24 17:41:30 UNRAID-Server sshd[39256]: Server listening on fd31:e9c1:819e::53 port 1997. Nov 24 17:41:30 UNRAID-Server sshd[39256]: Server listening on 192.168.178.29 port 1997. Nov 24 17:41:32 UNRAID-Server avahi-daemon[39819]: *** WARNING: Detected another IPv4 mDNS stack running on this host. This makes mDNS unreliable and is thus not recommended. *** Nov 24 17:41:32 UNRAID-Server avahi-daemon[39819]: *** WARNING: Detected another IPv6 mDNS stack running on this host. This makes mDNS unreliable and is thus not recommended. ***

Hey there, I'm wondering if I'm the first one asking this question. I couldn't find an answer. I need `logrotate` for my Docker Traefik service. I mean, I created a file a while ago in `/etc/logrotate.d` and was wondering why the logrotate didn't work. Probably because Unraid overrides the logrotate folder with its own defaults every boot. Do I need a user script at the first array start to add the custom logrotate script every time? I just want to be sure that this is the way to go and I'm not missing something. Kind regards

I know @Vr2Io but still I had to select them all manually because there are some old apps in between. But for the record, it seems that the problem has been solved now. Needed massive time for all the container reinstalls and selecting auto restart for most of them was buggy as hell (as always...) Not sure why the minor update caused this.

It was from 7.2.0. @JorgeB

Hey there, some hours after Unraid 7.2.1 came out, I already updated. That's probably where the problems started. First, I thought it was about Traefik, but then I realised I get tons of problems that even healthchecks are failing sometimes, and a Docker container cannot reach the localhost. I guess it would be an option to completely remove the Docker folder and let everything reinstall, but it takes a lot of time to select all the apps again... is there a faster way? There are many networking problems. I run a WordPress instance, and I have many 500 errors. First, it said there is no contact to the DB, then to Redis. But it is not all the time. Sometimes it loads, and sometimes it does not. I had to disable Redis in WordPress because it could never connect, but the container runs and none of the configuration changed. The same Redis container works for other containers that need Redis. You can see these random errors and restarts here. Out of nowhere, there is no connection, and it is in the container also some n8n flows failed or had a timeout. There is a general issue in my system that has nothing to do with exposed services. So it cannot be Traefik or Cloudflare. The system itself runs smoothly without any problems. That is confusing me. I have another friend with a very fresh install of Unraid that also upgraded to 7.2.1, and he has no problems. I also found no entries about similar behaviour in the forum. Let me know if there is any quick way to reinstall Docker with all apps without having to select everyone myself... (164 containers) And thanks for any help! unraid-server-diagnostics-20251121-2204.zip

In the end, I have no idea. Everything works as it used to before...

Hey @JorgeB , It fixed itself. It was very unlucky because the day I posted this, I had to go on a trip for 3 days very early in the morning, and I could only restart the server physically. I created a script that reboots the server every 4,5 hours and let it run until yesterday (so for 3 days). Then I deactivated the script to see if the problem still exists, and until now the server has been running for 1 day and 10 hours. If I remember it correctly, I swapped some hardware of the server a year ago and had the same problem that the server randomly crashed then. I also never found out why. Did any other users report a similar behavior? I mean, it's weird because now everything is just like nothing happened. Even the br0 network came back.

@JorgeB I will try those things and come back if I have a question or if I find a solution. Thanks for the help!

@JorgeB I hope that helps. I know that one 4 TB disk started to have a smart error. First it was just with value 1 and now 4. But this already before the disk replacement and it's just a RAID 1 data disk, not even used for Docker. unraid-server-diagnostics-20251106-1021.zip

Hello, first of all, I want to say that I uploaded a very big syslog file because I think it's better to give more than needed, if that's okay. 2 days ago (04.11.2025), I started to replace a 6 TB disk of my array with an 18 TB disk (parity is 18 TB as well). I also had the old one attached, but as an unassigned device and not mounted. The rebuild needed 1 day, but everything was fine. So I stopped the server again to put everything in place and get rid of the old disk (05.11.2025). I booted again, and everything was fine. Somehow, I think 5 hours after the start the server died out of nowhere. I couldn't see why in the logs. So I had to press the button on the case of my server and shut down the server, and started it again at probably 19:00 on 05.11.2025 (could also be a bit later). I wanted to go to sleep at approximately 00:30 on 06.11.2025, and the server again crashed, so I did the same. The next morning, the server was offline again (after 5 hours already, but I slept). I rebooted, and my pools (named cache and safety) were empty. The array was there and worked, but I found the disks in unassigned devices and had to put them in the empty pools again because I rebooted several times and didn't see it first. Additional information: my br0 network was gone after the disk replacement, and after some reboot, it was randomly back again, and it is gone again now. Has it also something to do with this or is it because of the crashs? Do you have any idea what is going on? syslog-192.168.178.29.log

Altes Thema, aber dasselbe Problem. Ich dachte eigentlich, ich hätte es, aber ich bin nun im Ausland per VPN nach Hause über WireGuard in meiner FRITZ!Box verbunden. Ich nutze ebenfalls br0 für AdGuardHome. Zu Hause hatte ich seit der Aktivierung von IPv6, die ich für ein Thread Netzwerk für mein Smart Home brauchte, erstmal nur Probleme und es dann mit der jetzigen Konfiguration aber geschafft, dass alle Geräte über AdGuardHome laufen. In der FRITZ!Box habe ich unter IPv4 Adressen für Lokaler DNS-Server die IP-Adresse von AdGuardHome, also bei mir 192.168.178.2 angegeben. Bei IPv6 Adressen, ebenfalls unter Heimnetz, so: Meine Unraid Einstellungen: Meine AdGuard DNS Einstellungen: dns: bind_hosts: - 0.0.0.0 - '::' port: 53 anonymize_client_ip: false ratelimit: 0 ratelimit_subnet_len_ipv4: 24 ratelimit_subnet_len_ipv6: 56 ratelimit_whitelist: [] refuse_any: true upstream_dns: - '[/fritz.box/]192.168.178.1' - '[/168.192.in-addr.arpa/]192.168.178.1' - '[/local/168.192.in-addr.arpa/]192.168.178.1' - https://dns10.quad9.net/dns-query upstream_dns_file: "" bootstrap_dns: - 9.9.9.10 - 149.112.112.10 - 2620:fe::10 - 2620:fe::fe:10 fallback_dns: [] upstream_mode: load_balance fastest_timeout: 1s allowed_clients: [] disallowed_clients: [] blocked_hosts: - version.bind - id.server - hostname.bind trusted_proxies: - 127.0.0.0/8 - ::1/128 cache_enabled: true cache_size: 4194304 cache_ttl_min: 0 cache_ttl_max: 0 cache_optimistic: false bogus_nxdomain: [] aaaa_disabled: false enable_dnssec: false edns_client_subnet: custom_ip: "" enabled: false use_custom: false max_goroutines: 300 handle_ddr: true ipset: [] ipset_file: "" bootstrap_prefer_ipv6: false upstream_timeout: 10s private_networks: [] use_private_ptr_resolvers: true local_ptr_upstreams: - 192.168.178.1 use_dns64: false dns64_prefixes: [] serve_http3: false use_http3_upstreams: false serve_plain_dns: true hostsfile_enabled: true pending_requests: enabled: true Scheinbar umgehe ich geschickt IPv6 irgendwie, da in meinen AdGuardHome Logs immer nur IPv4-Adressen auftauchen. Das war nämlich mit einer anderen Konfiguration so, dass Windows Rechner IPv6 hatten und dem entkommen waren. Und nun ist es so, dass die VPN-Verbindung definitiv weiterhin über IPv6 geht und deshalb meine schönen Redirects für meine Domain nicht klappen. Der Windows Rechner mit der VPN Verbindung gibt bei ifconfig nämlich: nslookup int.xyz.de Server: fritz.box Address: fd31:e9c1:819e:0:3e37:12ff:fedb:e7c Name: int.xyz.deDie IPv4-Adresse taucht auch in AdGuardHome auf. Kurioserweise sogar manche Redirects, die aber nicht klappen. Wo liegt der Fehler?

@Zak501 @levnetet I had the same problem. Updated, and the permissions were no longer sufficient. I updated the permission to 777 (rw-rw-rw) and it worked again. I have no idea why and it's not really nice. I also mentioned it here already, but the developer couldn't reproduce it somehow: https://github.com/stephondoestech/unraid-config-guardian/issues/11#issuecomment-3268341808

@itimpi thank you for the quick reply. I had the syslog server set to this: But there are no logs/syslog-previous.txt for me. And honestly, I only got the unraid server in use. There is no other server, so I don't know how to set up a remote syslog server. I know that mirroring to flash is not made to be used all the time.

Hello guys, sorry I'm really in trouble because everything I use daily is on my unraid server as well as my smart home. Randomly, at 8am, while I was sleeping, the server died. I could not open the WebGUI of unraid anymore, so I did an unclean shutdown. I restarted it and thought it was over, or some hardware died (even though the server was still running technically before). Unraid was able to boot again (felt longer than normally) and now everything looks normal but the Docker Service failed to start. And I need to know what the reason of the crash was. This never ever happened. I hope the diagnostics have information about the crash. I could not really find out what the reason was... So thankful for any help here. unraid-server-diagnostics-20250823-1502.zip

Is it possible to only use the flash backup? I don't want to back up my applications, but the flash disk and this would be really helpful.

Sorry please for beginners. Where to add those then? I'm already annoyed because since I activated IPv6 my internal AdGuardHome redirections stopped working properly...

Sorry for coming back to this old topic, but I'm in the same situation. With this post I realized that I even had no IPv6 DNS server in unRAID because my router (FritzBox) had IPv6 not correctly enabled. But I have the same problem. I don't get any errors if I run the matter-server container on br0 but on host it won't work. Where or how to set sysctl -w net.ipv6.conf.all.forwarding=1? Didn't work for the container. My docker settings:

You need it for an individual sender name instead of [email protected]. It's the common syntax and is used for many services. For example, Authentik, Outline, Reactive Resume, and more...