Might as well add another cog to my plan im on cent os
-
Okay, had a dr appt with the kid now I'm back with an update.
- Recompile complete, updated Apache installed now.
- Working on web socket proxy.
- Going to reinstall nginx themselves and use that as a fallback.
They are going above and beyond.. Maybe it's a quiet rainy day for them like it is for me I don't know but their level of support is phenomenal. Remember this was all a "DIY issue" fixed on a best efforts basis. Their best efforts are freakin' awesome!
-
The quest continues!! @julian here's an email from LW Support they had a question at the bottom for you guys so I figured best to just cut/paste in case it can help anyone down the road of the dev process for you guys in general.....
Steve,
I am glad to be of service. I also tried using mod_proxy_wstunnel in Apache 2.4 however it was not working, I was only getting a 500 internal server error. Here is a link to the page:
Here is the configuration I used:
ProxyPass / ws://127.0.0.1:4567/
ProxyPassReverse / ws://127.0.0.1:4567/I have changed it back to using just mod_proxy with the following configuration and the 500 errors are gone however the site is still slow to load the footer:
ProxyPass / http://127.0.0.1:4567/
ProxyPassReverse / http://127.0.0.1:4567/Perhaps the developers at nodeBB will have insight in a way to make nodebb work properly utilizing mod_proxy_wstunnel? I'd like to know if they think something like this could work.
-
-
@OutlawedOffroad -- any chance you can get a member of LW Support onto this forum to chat?
Done. Just emailed thread link and let them know everyone wants to work together to figure it out for both the community and them as a host that can support NodeBB users.
-
Productive day yesterday and night both on my end starting to re-color, resize, etc.. the theme but more importantly LW has totally delivered the awesome sauce. I'm waiting on permission to share since it's their code/work but I can post where I'm at today.
1 - nginx version needed is not in their current build. The people responsible for that did tell support it can be done, but they need a few days. The last thing I'm going is putting pressure as I believe they have gone above and beyond as it is. When they get it sorted the last major "issue" of lag will be squashed and I can tell family and friends, etc.. to check it out.
2 - Dan from the late night support kicks major ass. He offered 2 nights ago to help setup scripts for the cpanel service checker. Now every 3-5 minutes it checks for an active http response on port 4567. If it does not get one it fires up the kill script to clear out any hung processes and starts it back up.
I will share the .sh files once they give me the thumbs up. I just believe in asking first when it's someone else's hard work put in.
Less than a week so far and the thread on here basically saying CentOS guys have a lot of work to get it going has become CentOS guys have 1 or 2 small hurdles left to clear and then we can all use NodeBB.. BOOOOOM!!!
-
Their timeline of "a few days" apparently includes the time between telling me and doing other things.. Already working on updating ngix today for the server.. Can't say it enough how great their support is for anyone else needed a self-hosted option they are going above and beyond the call of duty to help make this happen for me.
More details later.. Football and food now..
-
So I moved to Digital Ocean for the sake of not screwing up my existing forums.. the proxy was screwing up the tapatalk and whatever else so it was not worth breaking 2 good sites for 1 test site.
dad-forums.com now is proxy on and resolving to the right stuff, blah blah..
but... i'm having that same data issue.. here's my config file per the wiki..
upstream nodebb {
server 162.243.229.221:4567;
}server {
listen 80;server_name dad-forums.com; root /home/steve/mynodebb/install; location / { proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_set_header X-NginX-Proxy true; proxy_pass http://nodebb/; proxy_redirect off; # Socket.IO Support proxy_http_version 1.1; proxy_set_header Upgrade $http_upgrade; proxy_set_header Connection "upgrade"; }}
anyone see something i'm missing here?
-
Weird cause at dad-forums.com:4567 it's working just fine with the counters and no delays anywhere.. so it's still happening now on DO as well.. so it must be something i'm doing or not doing.
next up is to move the data over i'm sure that's going to be a treat lol
-
PITA I have to keep rebooting the DO server for the site to work more than 5 min at a time. Then it goes to a white page..
I made the update to the IP, powering it down and up now.. restarting nginx, node, forever, whatever else seems to do nothing and it just hangs.
-
Here's how the site looks... when going direct to dad-forums.com where it's going the proxy.. keep in mind if I go to the same link and add the port :4567 it works just fine.
-
Now ~5min later the dad-forums.com ghosts up to a white page, but the :4567 still trucking along with no problems...
Jeeeeeeeez I'm glad I got a lot of hair so if I pull some out there's plenty left.
edit: spoke too soon, both sides a white screen now lol.
-
from nginx error log, I have pages of this error.
2013/12/10 19:34:22 [error] 614#0: *1 connect() failed (111: Connection refused) while connecting to upstream, client: 67.225.149.136, server: dad-forums.com, request: "GET /socket.io/1/?t=1386704047703 HTTP/1.1", upstream: "http://127.0.0.1:4567/socket.io/1/?t=1386704047703", host: "dad-forums.com", referrer: "http://dad-forums.com/"
-
this is the other error I'm seeing a bunch of as well:
2013/12/10 20:14:24 [error] 614#0: *450 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 71.183.91.42, server: dad-forums.com, request: "GET /socket.io/1/xhr-polling/h_jR4ziaFtyNQriIAv2D?t=1386706416376 HTTP/1.1", upstream: "http://127.0.0.1:4567/socket.io/1/xhr-polling/h_jR4ziaFtyNQriIAv2D?t=1386706416376", host: "dad-forums.com", referrer: "http://dad-forums.com/"
