[指南] 如何解决自定义网络上容器的 macvlan 和 ipvlan 问题

[sonic6]

1 hour ago, BurntOC said:

The server hangs finally drove me to abandon macvlan for now and I flipped the switch back to ipvlan again.  I've rebooted a couple of times since for various reasons, then in reviewing my FCP scan this morning it is still reporting DOZENS of macvlan call trace warnings - and I'm not running macvlan.  Meanwhile my very similar Unraid server with the same model quad nic and pretty similar network setup (on a 10th gen Intel vs the 11th gen on this one) hums along on macvlan very nicely.  Frustration peaking.

You should open a Bug Report with your Diagnostic File 

[diedrichg]

On 6/21/2023 at 11:30 AM, Kilrah said:

Then why are you using br0/br1?

Good point. I guess I had forgotten the definition of macvlan and just started following IbraCorp and Christian Lempa without really thinking about what I was setting up. I switched back to the standard bridge and the addresses are as I wish them to be. Thanks for the punch in the arm.

[BurntOC]

36 minutes ago, sonic6 said:

You should open a Bug Report with your Diagnostic File 

 

I figured out that removing macvlan via modprobe and clearing out the syslog entries for it stopped the bogus message (credit to another post here for that).  I'll probably do the diag submission, but honestly I'm tempted to wait a month or two to see if kernel updates fix it as there seems to be a number of these oddities with 11th gen systems right now.

[Masterwishx]

changed to br1 docker network without host access ,all was working fine no macvlan traces found .

but today somehow founded again !??!

 

[nraygun]

Uncle.
I've tried to get this to work on my system but I think I'm missing something. While I was not getting macvlan issues, I figured I'd configure it this way on a separate network in anticipation of going to 6.12.x. Plus my R710 has 4 network ports - may as well use 'em.

 

I configured my system as outlined in the original post. When I change my Swag container to use br1, I can't get to the webui. It seems with the old configuration using a custom docker network, it gives the container a different IP on a different network(172.18.0.3 443/80) that maps to the hosts network(192.168.0.100 1443/180).

 

With proxynet custom docker network:
172.18.0.3:443/TCP <-> 192.168.0.100:1443
172.18.0.3:80/TCP <-> 192.168.0.100:180

 

With br1 configured as in this post:

192.168.0.2:443/TCP <-> 192.168.0.2:443
192.168.0.2:80/TCP <-> 192.168.0.2:80

 

Relative noob here. Hope someone can steer me in the right direction.

 

 

 

[Omri]

I have 10th gen Intel with macvlan problem...

It's also (likely) not router related

Happens with asus router but also with OPNsense 

And it's happening with different nics on same PC

One of them is Intel i-225v and the other Aquantia 10G.

Out of ideas...

[aglyons]

I wonder how many others are running an Aquantia chipped NIC?

[Chunks]

My system is a 5950x Ryzen, and my only NIC is an Intel I211 gigabit

 

I have a dual port Intel X550 PCIe NIC coming this week, just to try to offload all macvlan stuff to a new interface.

[danob312]

For those who still have panics... no bonding and no bridging on both eth0 and eth1

[kl0wn]

This resolved the issue for me as well but dedicating a NIC to one network is kind of a bummer.

[nraygun]

On 6/24/2023 at 5:45 PM, nraygun said:

Uncle.
I've tried to get this to work on my system but I think I'm missing something. While I was not getting macvlan issues, I figured I'd configure it this way on a separate network in anticipation of going to 6.12.x. Plus my R710 has 4 network ports - may as well use 'em.

 

I configured my system as outlined in the original post. When I change my Swag container to use br1, I can't get to the webui. It seems with the old configuration using a custom docker network, it gives the container a different IP on a different network(172.18.0.3 443/80) that maps to the hosts network(192.168.0.100 1443/180).

 

With proxynet custom docker network:
172.18.0.3:443/TCP <-> 192.168.0.100:1443
172.18.0.3:80/TCP <-> 192.168.0.100:180

 

With br1 configured as in this post:

192.168.0.2:443/TCP <-> 192.168.0.2:443
192.168.0.2:80/TCP <-> 192.168.0.2:80

 

Relative noob here. Hope someone can steer me in the right direction.

 

 

 

 

Can anyone help me? I think I'm missing something here but can't figure it out.

[nraygun]

On 6/28/2023 at 8:08 AM, nraygun said:

 

Can anyone help me? I think I'm missing something here but can't figure it out.

 

I think I got it!

I had to turn off bridging on eth0 and change a few ports here and there.

Everything seems to work except for qbittorrentvpn - I can't get to the webgui nor can the -arr's get to it.

I can get to webgui of other containers just fine, just not this one. Nextcloud+Mariadb+Swag works fine on the new br1 too.

 

If anyone has any suggestions, I'd appreciate it. I'll go pursue this in the binhex-qbittorrentvpn forum.

[nraygun]

18 hours ago, nraygun said:

 

I think I got it!

I had to turn off bridging on eth0 and change a few ports here and there.

Everything seems to work except for qbittorrentvpn - I can't get to the webgui nor can the -arr's get to it.

I can get to webgui of other containers just fine, just not this one. Nextcloud+Mariadb+Swag works fine on the new br1 too.

 

If anyone has any suggestions, I'd appreciate it. I'll go pursue this in the binhex-qbittorrentvpn forum.

 

Got qbittorrentvpn working after a long night. Had to change the webgui port. I had to turn on the proxy in a browser, then go to the webgui and change it in the app. One odd thing, even though I changed the webgui port to 8585, unRaid's port mapping still shows 8080. But hey, it works.

Edited June 30, 2023 by nraygun

[wgstarks]

8 minutes ago, nraygun said:

then go to the webgui and change it in the app

You probably needed to change it in the docker config-

https://github.com/binhex/documentation/blob/master/docker/faq/qbittorrentvpn.md

Scroll down to Q4.

[enJOyIT]

What is bridging for? I disabled this (because I still get kernel panics) and created a custom network eth1 for my docker containers. 

 

It's working like when I'm using br1.

 

So what is the difference here?!

[Evidenz]

I assume this fix will not allow VLANs on eth1?

[nraygun]

5 hours ago, wgstarks said:

You probably needed to change it in the docker config-

https://github.com/binhex/documentation/blob/master/docker/faq/qbittorrentvpn.md

Scroll down to Q4.

Well, I tried that and there is no container port field when I click add. If I edit one of the existing ports, it shows a container port. I might just leave well enough alone.

 

Edited June 30, 2023 by nraygun

[nraygun]

2 hours ago, nraygun said:

Well, I tried that and there is no container port field when I click add. If I edit one of the existing ports, it shows a container port. I might just leave well enough alone.

 

I guess the deal is there is no container port when on a custom network. Things are weird when using this new br1 I created on eth1.

I noticed when I changed the WEBGUI_PORT to 8585, it does show up this way in the app - I just can't get to it without going through the proxy. The only thing that seems to work is leaving WEBGUI_PORT at 8080, configure Firefox to use the proxy, and then go into the app with 192.168.0.6:8080 and changing to 8585 in the app. If I do it any other way, I can't get to the ui without going through the proxy.

I also see that my Host Port 3 is set to 8585 with the container port set to 8081 but 8081 doesnt show up in the port mappings.

[Evidenz]

Does anyone know if this roll-back in the latest update (2023-06-29) has anything to do with the here discussed mcvlan situation?

 

[sonic6]

1 hour ago, Evidenz said:

Does anyone know if this roll-back in the latest update (2023-06-29) has anything to do with the here discussed mcvlan situation?

 

unfortunately no, this rollback has nothing to do with the macvlan problem

[rdagitz]

Anyone know how to downgrade to a version of Unraid to avoid this issue until someone gets this figured out?

[sonic6]

2 minutes ago, rdagitz said:

Anyone know how to downgrade to a version of Unraid to avoid this issue until someone gets this figured out?

for me it is 6.11.5, for other it is 6.10.x... for the one next 6.9.x is stable. slady to say, you have to test it for your own case.

[nraygun]

Oh FFS!

 

Still trying to understand the whole qbittorrent thing, but I fired up a VM for a routine thing I do on the first of the month and it didn't work. 😞

 

The VMs were using "virbr0" which I guess is not there anymore since br0 is not available. I was able to change one VM to use br1 and it worked, but some of the other VMs don't have the "Network Source:" option. I tried copying the XML from the working VM but got a duplicate MAC address situation, I think.

 

What sort of things do I have to do with the VMs with this dedicated Docker network?

 

I'm starting to doubt the value of using this dedicated network.

[rdagitz]

Went back to 6.11.3. I had changed my cache pool to zfs so BEFORE the downgrade, I moved everything off of cache, formatted the drives to btrfs and then recreated the cache pool AFTER the downgrade. Maybe that was overcautious ... but I had no problems.

 

Here are the instructions I followed for the downgrade:

https://docs.unraid.net/unraid-os/manual/upgrade-instructions/#manual-upgrade-or-downgrade

[kevin182]

I cannot get eth1 to come back after upgrading to 6.12.2. I can see my NIC in the system devices but no new network adapter is appearing. I have tried three different NICs. Attaching diagnostics.

 

UGREEN USB to Ethernet Adapter

ASUS 2.5G Ethernet USB Adapter

Both of these appear in system devices but no eth1 in network settings

 

2.5GBase-T PCIe Network Adapter RTL8125B

This 1 does not appear at all in the system devices and I have added the plugin for the Realtek drivers.

 

lucifer-diagnostics-20230704-2354.zip