aglyons

The thing is, MACVLAN creates 'virtual' MAC addresses. Routers don't have a problem with that as they appear as separate devices. IPVLAN drops the unique MAC addresses and simply assigns the IP to the host MAC. That's the problem. It may well be working just fine. Everything is responding to pings and is operating as expected. One of the containers is a PiHole for the br0 network. If you feel up to giving me some pointers on setting up a VLAN on the UDM for use with the custom network on Docker, I'm all ears.

So I had issues with Docker causing kernel panic crashes. I was advised to change from MACVLAN to IPVLAN in the Docker config. It seems to have solved the crashing but it has had another effect on the rest of my network. Specifically, my router. I have a Ubiquiti UDMProSE router. I don't think that should really matter which make or model but with it I can see more details of the network. I have a static IP assigned to the Unraid host network as well as any Docker guest that is also using the host br0. In the admin of my router, it shows the MAC address of the host flipping IP addresses at random from one static IP to another. When one is showing, the other is not known to the router. There are no results when you search for the other static IP. I put a community post up on the UI boards. We'll see if anything comes of that. Maybe I am wrong and all of this is entirely cosmetic.

So it appears that this may have caused another issue. I started to notice instability with some containers on br0 networks and now bridge networks as well. One of my containers is trying to connect to an external SFTP server and it's timing out failing to maintain a constant connection. I noticed in my router, Ubiquiti UDMProSE that the MAC address of the server shows the various IP's of the containers flipping from one IP to another. Could this be the result of IPVLAN instead of the MACVLAN? I do have two other NIC's in the server that I could assign one dedicated to the PiHole. I'd have to figure out how to create the custom network in Docker and attach a specific eth to it. Any advice there? UPDATE: So I have confirmed that when Docker is set to IPVLAN, it causes issues, at least with Ubiquiti routers, where outbound connections are timing out and failing. I changed back to MACVLAN and the problems disappeared. Set it to IPVLAN and the problems came back again. I am not experienced enough to troubleshoot further on this but I am willing to try suggestions and to perform testing.

So I realized I had to shutdown the Docker service to change this setting. Once down I could change it but then, Docker won't start. There is a line in the console; Aug 30 11:30:53 KNOXX root: '/mnt/user/system/docker/docker.img' is in-use, cannot mount I'm trying a warm boot to see if that clears it up. UPDATE: So the warm boot solved the problem of Docker not starting. I have it set to IPVLAN. I'll watch and see how this runs. Thanks for your assistance!

My system, lately, has started to get erratic with it's stability. It crashed hard just this morning and checking the logs showed nothing out of the ordinary. This however was displayed on the monitor I have connected. The biggest problem is I have, aside from losing a ton of data I'd rather not lose, is a PiHole Docker container runs on here. When the system goes down that kills the internet for the whole house (= angry wife & kid). Can anyone offer any advice? SyncMaster SA650 </TASK> Modules linked in: maculan xt_nat xt_tcpudp veth xt_conntrack xt_MASQUERADE nf_c onntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat nf_conntrack nf_defrag _ipv6 nf_defrag_ipv4 br_netfilter xfs md_mod iptable_filter ip6_tables iptable_ filter ip_tables x_tables atlantic bmx2 intel_powerclamp coretemp crc32c_intel i ntel_cstate ipmi_ssif i2c_core intel_uncore megaraid_sas input_leds led_class ip mi_si umi acpi_power_meter button [last unloaded: atlantic] --[ end trace 9e2cc8776e9cb2ed ]--- RIP: 0010:nf_nat_setup_info+0x146/0x773 [nf_nat] Code: 15 c0 6a 00 00 89 c0 48 8d 04 c2 48 8b 10 48 85 d2 74 07 48 81 ea 98 00 00 00 45 31 c0 48 85 d2 of 84 6b ff ff ff 8a 44 24 4e <38> 42 46 74 19 48 8b 92 98 00 00 00 48 85 d2 0f 84 52 ff ff ff 48 RSP: 0018:ffffc90000274790 EFLAGS: 00010202 RAX: ffff888177ac0206 RBX: ffff888c85f 183c0 RCX: 1cd53ef870e390c7 RDX: 002900002b26ff68 RSI: 5b01b43358f245ef RDI: ffffc90000274768 RBP: ffffc90000274860 RO8: 0000000000000000 R09: b7ff817916811ca9 R10: 8e54dbffc0bc8b40 R11: cca2307651abe750 R12: 0000000000000000 R13: ffffc90000274874 R14: 0000000000000000 R15: ffffc900002747b8 FS: 000014b46c781840(0000) GS: ffff8897dfc00000(0000) kn1GS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CRO: 0000000080050033 CRZ: 0000557b78fa6298 CR3: 00000007be032000 CR4: 00000000000006e0 Kernel panic not syncing: Fatal exception in interrupt Kernel Offset: disabled ---[ end Kernel panic not syncing: Fatal exception in interrupt ]---

I figured it out. I had assigned an IP on the dockerz network. Another container on the same network was started prior to NPM starting and it was assigned the same IP address. Once I cleared the assigned IP NPM started ok. I made the assumption that defining an IP on a container would reserve the IP from the Docker DHCP. But the Docker network is not like a router. There is no IP reservation. At least defining it in a container is not going to reserve it lol.

Running into a 403 trying to start the container. There is zero content in the log window. I've done a force update on the container. Still no dice. Any thoughts?

Has anyone figured out the permissions issues with the config folder? I have set permissions to RW-RW-RW and set the owner to 'root' for every folder including the parent and it still cannot access the files. Docker APP Config Permissions and ownership on parent folder Permissions and ownership on folder contents ********************** LOG FILE OUTPUT ********************************** [1661374194] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:a8::e port 53 [1661374194] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:a8::e port 53 [1661374194] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:12::d0d port 53 [1661374194] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:2::c port 53 [1661374194] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:2d::d port 53 [1661374194] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:2d::d port 53 [1661374195] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:1::53 port 53 [1661374195] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:503:c27::2:30 port 53 [1661374195] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:2d::d port 53 [1661374195] libunbound[22:0] error: udp connect failed: Cannot assign requested address for 2001:500:a8::e port 53 /opt/unbound/etc/unbound/unbound.conf:293: error: cannot open include file '/opt/unbound/etc/unbound/a-records.conf': No such file or directory /opt/unbound/etc/unbound/unbound.conf:294: error: cannot open include file '/opt/unbound/etc/unbound/srv-records.conf': No such file or directory /opt/unbound/etc/unbound/unbound.conf:300: error: cannot open include file '/opt/unbound/etc/unbound/forward-records.conf': No such file or directory read /opt/unbound/etc/unbound/unbound.conf failed: 3 errors in configuration file [1661374195] unbound[1:0] fatal error: Could not read config file: /opt/unbound/etc/unbound/unbound.conf. Maybe try unbound -dd, it stays on the commandline to see more errors, or unbound-checkconf ** Press ANY KEY to close this window **

So I did a very non-scientific test and timed a section of files. It transferred 30.9GB of data in 8 minutes. If my math is right, that works out to approx 66MB/s average transfer rate. This is a significant performance hit when considering hardware capabilities IMO. I've read people talk about the purpose of UnRaid and running this kind of server in their operations. Many have described their viewpoint as this is an archive server. Somewhere to put files that are not accessed very often. I didn't come across this sentiment until well after I set all of this up. This may be my downfall as I did not expect this level of performance hit. I deal with a lot of large media files as I work with video content. I need not only large secure storage but also fast. I upgraded my network to support 10GBe connections between my workstation and the UnRaid server. With this performance hit, it seems like it was a waste of $$ as the network is not the bottleneck. I foresee the comments from that last statement. Secure, Fast, Affordable - Pick two out of the three options. I think I'll have to research TruNas and see how I can port everything over.

True, just checked the specs on the Ironwolf drives and they are rated at 240MB/s on paper. I guess my frustration is the constant up and down speeds. Especially dropping down so low for so long. I know they are different animals but I don't see this kind of fluctuation from the old Synology box I have which doesn't have any highspeed cache at all.

The transfers speeds are bouncing around topping out at 160MB/s but then dropping down to >30MB/s then back up for a short period but then back down. It never goes much higher than 160MB/s which is still way off the 500MB/s potential of a single drive and more realistically, stays in the lower 30MB/s or less speeds for the majority of the activity.

@JorgeB - All drives are Seagate Ironwolf, 5x6TB + 3x8TB except parity. Both parity drives are Seagate Exos 12TB and 14TB. @wgstarks - I've had a challenge trying to get rsync on UnRaid to appear as a target on my Synology Hyperbackup APP to no avail. I'm not that much of a CLI guy at present so running rsync at the CLI is not something that I readily know how to do.

I have had my issues with network performance and now I am seeing overall internal performance that I can't ignore or accept. I am transferring files from one share to another share, internally using the Dynamix File Manager. I have 8 drives that are all SATA3 600Gbps. So why is it that this internal file transfer is running at 11MB per second?? That is 2.2% of the drives potential throughput! I understand that UnRaid does not stripe data across drives. But going from one SATA3 drive to another SATA3 drive, internally should NOT be going this slow. I've seen transfers start at a high speed but then peter out to almost nothing dropping as low as 6MB/s The target nor the source share DO NOT use the cache. I'm starting to feel like UnRaid was not the right option to go with and I should have gone with Trunas instead.

Most consumer ISP's do not allow servers to be used on residential connections. Some block 80 and 443 by default. While I know that a hacker could do a port scan and attempt entry, I would prefer to not have standard ports exposed. For both reasons. I am not confident enough in ensuring everything is locked down properly. I have a good firewall but opening ports on servers bypass that security and literally opens your network up. I'm just getting started with UnRaid and homelabs. We gotta start somewhere. PS: If the plugin author is reading this, it would be great if the link to the docs was presented in the UI.

Yes, you should but reading the NPM forum post, I should be putting a TXT value as such; But the UI, when I choose my DNS provider, Google. I am presented with this and I have no idea what to do with it. There is no simple TXT value choice in the dropdown.

so the non-standard port issue is a result of letsencrypt. There is a solution but it does not seem to be possible with this container. It requires a TXT DNS entry. I can't seem to figure out how to do this with the current container options. https://community.letsencrypt.org/t/using-encrypt-for-non-standard-ports/20164/3

Has anyone managed to get external non-standard ports working, specifically with Nextcloud? I've managed to get it working with standard 443 https but if I try to use a non-standard external port everything gets borked. I've added this to the advanced nginx config section as I read in a post listen 8585 ssl http2; didn't seem to work out right. From what I've seen in the logs that config is not getting added to the conf file. [6/3/2022] [2:03:52 AM] [Nginx ] › ℹ info Reloading Nginx [6/3/2022] [2:03:57 AM] [SSL ] › ℹ info Requesting Let'sEncrypt certificates for Cert #9: domain.domain.com [6/3/2022] [2:03:57 AM] [SSL ] › ℹ info Command: certbot certonly --config "/etc/letsencrypt.ini" --cert-name "npm-9" --agree-tos --authenticator webroot --email "[email protected]" --preferred-challenges "dns,http" --domains "domain.domain.com" [6/3/2022] [2:04:00 AM] [SSL ] › ✔ success Requesting a certificate for domain.domain.com Successfully received certificate. Certificate is saved at: /etc/letsencrypt/live/npm-9/fullchain.pem Key is saved at: /etc/letsencrypt/live/npm-9/privkey.pem This certificate expires on 2022-09-01. These files will be updated when the certificate renews. NEXT STEPS: - The certificate will need to be renewed before it expires. Certbot can automatically renew the certificate in the background, but you may need to take steps to enable that functionality. See https://certbot.org/renewal-setup for instructions. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - If you like Certbot, please consider supporting our work by: * Donating to ISRG / Let's Encrypt: https://letsencrypt.org/donate * Donating to EFF: https://eff.org/donate-le - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - [6/3/2022] [2:04:01 AM] [Nginx ] › ℹ info Reloading Nginx [6/3/2022] [2:04:01 AM] [Express ] › ⚠ warning Command failed: /usr/sbin/nginx -t -g "error_log off;" nginx: [emerg] no "ssl_certificate" is defined for the "listen ... ssl" directive in /data/nginx/proxy_host/5.conf:6 nginx: configuration file /etc/nginx/nginx.conf test failed [6/3/2022] [2:04:33 AM] [SSL ] › ℹ info Testing http challenge for domain.domain.com [6/3/2022] [2:04:48 AM] [SSL ] › ℹ info HTTP challenge test failed for domain domain.domain.com the host was not found 5.conf # ------------------------------------------------------------ # domain.domain.com # ------------------------------------------------------------ server { set $forward_scheme https; set $server "192.168.200.88"; set $port 444; listen 80; listen [::]:80; server_name domain.domain.com;

I currently don't have any pass through enabled, heck I don't even have any VMs or Docker's installed yet. I also haven't come across any performance or system freezes. Everything seems to be working normal. Hopefully I didn't just jinx myself lol

OK, parity check completed with 0 errors. But, I have UnRaid set to send SNMP messages to my Synology log service. I am finding UnRaid sends a massive number of log messages all set as Log Severity of 'ERR'. I just received 376 log entries regarding.... 2022/05/20 16:07:00 [error] 6727#6727: *904052 limiting requests, excess: 20.132 by zone authlimit, client: 192.168.200.254, server: , request: PROPFIND /login HTTP/1.1, host: knoxx

New V6.10 update and now I see a series of DMAR errors popping up in the monitor I did a search and came across people suggesting doing a parity check. I just started one so we'll see what comes from that. I'll post diags if I keep seeing this pop up or if by request.

All's good now. Connected to mothership. Thx for your help!

Nope, changing that in the Management section of the settings page didn't even change the tld that UnRaid responds to. <-----UNRAID-API-REPORT-----> SERVER_NAME: KNOXX.lyons ENVIRONMENT: production UNRAID_VERSION: 6.9.2 UNRAID_API_VERSION: 2.46.3 (running) NODE_VERSION: v14.15.3 API_KEY: valid MY_SERVERS: authenticated MY_SERVERS_USERNAME: aglyons RELAY: connected MOTHERSHIP: ok ONLINE_SERVERS: KNOXX[owner="aglyons"] OFFLINE_SERVERS: ALLOWED_ORIGINS: http://localhost, http://IPV4ADDRESS, https://IPV4ADDRESS, http://knoxx, https://knoxx, http://knoxx.local, https://knoxx.local HAS_CRASH_LOGS: no </----UNRAID-API-REPORT----->

ah ha, I think I figured it out I am using my own TLD '.lyons' with my PiHole. Unifi does not like .local for some reason. It causes a bunch of problems. I failed to update the TLD in UnRaid. It still says .local

Nope, no RP -yet that is. I do plan on going there once I get the hang of it. Plugin is up to date report contents looked ok to me too <-----UNRAID-API-REPORT-----> SERVER_NAME: KNOXX ENVIRONMENT: production UNRAID_VERSION: 6.9.2 UNRAID_API_VERSION: 2.46.3 (running) NODE_VERSION: v14.15.3 API_KEY: valid MY_SERVERS: authenticated MY_SERVERS_USERNAME: aglyons RELAY: connected MOTHERSHIP: ok ONLINE_SERVERS: KNOXX[owner="aglyons"] OFFLINE_SERVERS: ALLOWED_ORIGINS: http://localhost, http://192.168.XXX.XXX, https://192.168.XXX.XXX, http://knoxx, https://knoxx, http://knoxx.local, https://knoxx.local HAS_CRASH_LOGS: no </----UNRAID-API-REPORT----->

Came across this recently and it won't go away. I searched around the forum and found nothing. knoxx-diagnostics-20220519-1233.zip