traefik: 1.2.2 breaks endpoints, redirect loop
What version of Traefik are you using (traefik version
)?
1.2.2
What is your environment & configuration (arguments, toml…)?
Endpoints are added dynamically via Docker. Probably something is wrong with my configuration. 1.2.1 works, though.
traefikLogsFile = "data/traefik.log"
accessLogsFile = "data/access.log"
defaultEntryPoints = ["http", "https"]
[retry]
[entryPoints]
[entryPoints.http]
address = ":80"
[entryPoints.http.redirect]
entryPoint = "https"
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
[acme]
email = "mail@example.com"
storageFile = "data/acme.json"
onDemand = false
caServer = "https://acme-v01.api.letsencrypt.org/directory"
entryPoint = "https"
acmeLogging = true
[[acme.domains]]
main = "example.com"
[docker]
endpoint = "unix:///var/run/docker.sock"
domain = "example.com"
watch = true
exposedbydefault = false
swarmmode = false
What did you do?
Upgraded from 1.2.1 to 1.2.2.
What did you expect to see?
Normal operation.
What did you see instead?
All URLs result in a 302 loop.
If applicable, please paste the log output in debug mode (--debug
switch)
> 255.255.255.255 - - [13/Apr/2017:07:15:58 +0000] "GET /my/url/ HTTP/2.0" 302 5 "" "User-Agent" 228 "" "" 0ms
> 255.255.255.255 - - [13/Apr/2017:07:15:58 +0000] "GET /my/url/ HTTP/2.0" 302 5 "" "User-Agent" 229 "" "" 0ms
> 255.255.255.255 - - [13/Apr/2017:07:15:58 +0000] "GET /my/url/ HTTP/2.0" 302 5 "" "User-Agent" 230 "" "" 0ms
> 255.255.255.255 - - [13/Apr/2017:07:15:58 +0000] "GET /my/url/ HTTP/2.0" 302 5 "" "User-Agent" 231 "" "" 0ms
About this issue
- Original URL
- State: closed
- Created 7 years ago
- Reactions: 2
- Comments: 21 (9 by maintainers)
From browser log it seems that it redirect https://domain.tld/ to https://domain.tld:443/ and since the browser will clean :443 it is looping. Example : https://try.gitea.io
Fixed by #1433. I will push 1.2.3 ASAP.
@andrewcfitz sorry, this time it should work.
@emilevauge, not sure why it is looping but diving in to see if I can find the cause. Didn’t hit this scenario in my testing.
Same here. Had to rollback to v1.2.1.
@f3l1x @sapk any chance to get some logs ?