Darrell Enns

@MadSpindel It grabs the lowest possible chunk of the available UID range. If you only ever have 1 pod, then you won't run into the issue. If you have multiple...

I've also run into the issue when trying to run logstash in swarm mode (for collecting syslog messages from various hosts). The logstash "host" field always appears as 10.255.0.x, instead...

Perhaps the linux kernel IPVS capabilities would be of some use here. I'm guessing that the IP change is taking place because the connections are being proxied in user space....

@vfarcic I don't think that's possible with the way it works now. All client connections come from the same IP, so you can't translate it back. The only way that...

@mrjana The whole idea of using IPVS (instead of whatever docker currently does in swarm mode) would be to avoid translating the source IP to begin with. Adding an X-Forwarded-For...

@tlvenn Do you know where this is in the source code? I could be wrong, but I don't believe it is using IPVS based on some things I've observed: -...

Definitely agree with @lbriais here. An option to specify the loopback device is the best solution. The workaround of moving other applications away from the first loopback device doesn't work...

I would propose simply adding a new configuration option which allows the user to specify the v4l2loopback device they wish to use. This should allow for all of the more...

@DanielRios549 The loopback devices are created when the v4l2loopback kernel module is loaded. There is no provision for applications to create new ones after the module has already been loaded....

@mikeperalta1 > Would it be possible/practical to scan all existing loopback devices, unload the module, then reload the module with all previous devices plus a new one? If any other...