Hello guys,
I’m having doubt on why the dsn generated by my Sentry (on-premise) is like this:
://bd38ea654d374917b4d5a70c699c425:5df877f466b54eac84b8398df0eec9a@/221
I think the right form should be:
https://bd38ea654d374917b4d5a70c699c425:5df877f466b54eac84b8398df0eec9a@mysentry.com/221
So, does anybody know how to fix it? Or is there anything with my configuration?
Thanks!
T̶h̶a̶t̶ ̶i̶s̶ ̶p̶e̶r̶f̶e̶c̶t̶l̶y̶ ̶n̶o̶r̶m̶a̶l̶.̶
Read answer below.
This stackoverflow link describes it in some detail and also links to the RFC source. Both are worth a read.
matt
June 11, 2018, 4:28pm
3
This is not normal. It’s not just a protocol relative URL. If you look closer, it’s missing the hostname as well.
This is due to missing the system.url-prefix
setting. We use this to construct the DSN, but when it’s missing, you’ll get a value like you showed.
1 Like
zeeg
June 11, 2018, 10:28pm
4
Heads up this can happen with the Docker install as the default caching backend is in-memory (and per process/thread). We probably should fix the defaults here.