I must have misconfigured something. But I cannot find out what the problem is. Is anone else experiencing lots of "upstream prematurely closed connection while reading response header from upstream" errors in their nginx log?
Tried tweaking buffers sizes etc but I see no difference. Get the feeling that nodebb is terminating polling clients connections.
Hmm I don't think we expose isAdmin and isModerator globally to the templates so that you can them everywhere. Maybe we should, everything in the config object is exposed so if you open your console and type config any variable you see under that object you can use in the templates. You can also take a look at the routes by prepending api to the url like https://community.nodebb.org/api/topic/1917/users-shouldn-t-be-able-to-purge-their-posts/10 you can use stuff from there in the template as well.