https://rancher.com/ logo
Title
w

witty-honey-18052

04/08/2023, 5:03 PM
I ended up grabbing the Node IP from the rancher-desktop dashboard, and manually added entries in wsl's etc/hosts file, and those ended up working as a workaround. such as:
172.26.122.158       mysite.localhost
c

calm-sugar-3169

04/11/2023, 5:26 PM
Hi Colin, are you referring to the host file on your windows machine? if so, rancher desktop does nothing with that. The host file in the distro should look like something like below (the
172.24.224.1
is the wsl interface IP address):
# This file was automatically generated by WSL. To stop automatic generation of this file, add the following entry to /etc/wsl.conf:
# [network]
# generateHosts = false
127.0.0.1       localhost
127.0.1.1       DESKTOP-PCNAME.localdomain     DESKTOP-PCNAME

# The following lines are desirable for IPv6 capable hosts
::1     ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
# BEGIN Rancher Desktop configuration.
172.24.224.1 host.rancher-desktop.internal host.docker.internal
# END Rancher Desktop configuration.
w

witty-honey-18052

04/12/2023, 7:17 PM
@calm-sugar-3169 my WSL /etc/hosts is missing the rancher-desktop section
c

calm-sugar-3169

04/12/2023, 7:17 PM
hmm, that doesn’t seem right, you may need to recreate it or do a factory reset if you can
not 100% sure why
/etc/hosts
would be missing 🤷
w

witty-honey-18052

04/12/2023, 7:22 PM
same, trying to work through that and replicate it
here's what mine looks like. It could be related to windows updates screwing something up and/or replacing the file.
# This file was automatically generated by WSL. To stop automatic generation of this file, add the following entry to /etc/wsl.conf:
# [network]
# generateHosts = false
127.0.0.1       localhost
127.0.1.1       computer.localdomain   computer

# The following lines are desirable for IPv6 capable hosts
::1     ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
I'm doing a factory reset + uninstall + reinstall to see if that helps. maybe there was something upgrade related.
👍 1
c

calm-sugar-3169

04/13/2023, 3:43 PM
It's possible because that looks like a default file, and the section we append to it is not there.
w

witty-honey-18052

04/13/2023, 3:45 PM
just out of curiosity, does the section get added on install, versus on startup?
I did manually add the section to /etc/hosts, and it didn't seem to help. but i didn't try to also check the windows hosts file. I tested by curling within WSL, and it couldn't reach my traefik workloads
after reinstall, same thing, missing section:
# This file was automatically generated by WSL. To stop automatic generation of this file, add the following entry to /etc/wsl.conf:
# [network]
# generateHosts = false
127.0.0.1       localhost
127.0.1.1       computer.localdomain   computer

# The following lines are desirable for IPv6 capable hosts
::1     ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
I can access my ingress through chrome on windows, but can't curl from WSL
krumware@computer:~/code/suse$ curl <http://krum-web.localhost>
curl: (6) Could not resolve host: krum-web.localhost
I have a feeling that your WSL may have gone into weird state that I have encountered a few times in the past. For some reason the networking stack gets into this strange state that everything starts to make no sense which could be related to the dirty iptables. Are you able to do factory reset and then followed by
wsl --shutdown
?
w

witty-honey-18052

04/13/2023, 6:45 PM
I'll try that. I had the same issue on another WSL machine that I'm trying again on now
@calm-sugar-3169 wil any of the crew be at kubecon? travelling that way now, could be convenient to debug in person at some point
c

calm-sugar-3169

04/17/2023, 5:55 PM
Hi Colin, yes, there will be some folks at Kubecon for sure, I’m not 100% sure who though. I can find out and let you know.
w

witty-honey-18052

04/17/2023, 6:03 PM
That would be great! Some of this might be related to me deploying the rancher helm chart locally too, but it's difficult to debug
c

calm-sugar-3169

04/18/2023, 5:05 PM
SUSE booth at Hall 1, Booth D1, also, talk to @thankful-hospital-82027 he is most likely will be able to debug this with you.
💯 1
w

witty-honey-18052

04/18/2023, 6:12 PM
Awesome, thank you!
t

thankful-hospital-82027

04/18/2023, 8:21 PM
Hi @witty-honey-18052 , indeed I’ll gladly help you 👍 Let’s talk during the week.
w

witty-honey-18052

04/18/2023, 9:34 PM
Awesome, thanks! I'll be by early if someone's around. This issue is blocking me on Rancher Extensions work and would be great to knock out so I can promote extensions with other partners
t

thankful-hospital-82027

04/19/2023, 5:18 PM
hi @calm-sugar-3169 , I checked it with Colin and after updating WSL to the store version the Windows host files are not populated with the
rancher.localhost
entry. The workflow Colin uses is then impacted as the address is not reachable from either WSL (integrated distro) or powershell
We’re trying a last test with a previous version of RD and see if there’s a potential regression with 1.8.1