These are chat archives for scytherswings/Plex-Board

12th
Apr 2017
Andrew Newell
@scytherswings
Apr 12 2017 00:57
Yo @seanvree did everything go alright last night?
Andrew Newell
@scytherswings
Apr 12 2017 01:08
if you've got an update for this bug: scytherswings/Plex-Board#84 that would be good to hear
Sean Vreeland
@seanvree
Apr 12 2017 09:06
@scytherswings ...yes, did some research last night, I found what the issue is, but not sure how to fix it just yet. I'll update the issue
also, didn't get a chnage to pull the new update...about to do that now
Sean Vreeland
@seanvree
Apr 12 2017 09:24
@scytherswings posted an update. Still needed testing...running some tests tonight, but I need help on the BASH scripts. so we will need to do a teamviewer
Andrew Newell
@scytherswings
Apr 12 2017 10:21
Ah okay, thanks for the update. I can do a TeamViewer tonight probably around 9pm or 10pm EST if that works for you
Sean Vreeland
@seanvree
Apr 12 2017 13:28
Hey, need to make some changes to the wiki for windows
not the pending service install stuff
but just the root windows wiwki page
I'll email you some suggestions
minor
Andrew Newell
@scytherswings
Apr 12 2017 13:29
Sounds good
Sean Vreeland
@seanvree
Apr 12 2017 13:47
Hey
you around real quick?
quick quesiton @scytherswings
step nine
TYPE serverSetup.sh | MORE /P > serverSetupConverted.sh && MOVE /y serverSetupConverted.sh serverSetup.sh This will convert the \n to \r\n for windows to use.
I'm assuming TYPE should NOT be run as a command?
or wait, nevermind
yeah, that's good
right?
Andrew Newell
@scytherswings
Apr 12 2017 13:49
Yeah that's an actual command heh
Sean Vreeland
@seanvree
Apr 12 2017 13:49
kk
Andrew Newell
@scytherswings
Apr 12 2017 13:49
In normal CMD.exe
Sean Vreeland
@seanvree
Apr 12 2017 13:49
right
I got what your'e doing
i'll send you my suggestions
we good
Andrew Newell
@scytherswings
Apr 12 2017 13:50
Sweeeeeet
Sean Vreeland
@seanvree
Apr 12 2017 14:02
Hey
is this still current:
To run the server: bundle exec rails server puma -e production --won't work currently because windows can't fork the process. need to edit config/puma.rb and change daemonize true to daemonize false
Andrew Newell
@scytherswings
Apr 12 2017 14:03
No, my latest change fixed that behavior
Sean Vreeland
@seanvree
Apr 12 2017 14:04
what branch?
Andrew Newell
@scytherswings
Apr 12 2017 14:04
Master
Sean Vreeland
@seanvree
Apr 12 2017 14:04
k
Andrew Newell
@scytherswings
Apr 12 2017 14:04
AFK
Sean Vreeland
@seanvree
Apr 12 2017 14:04
but the cmd is still
bundle exec rails server puma -e production
?
Andrew Newell
@scytherswings
Apr 12 2017 14:05
No
Sean Vreeland
@seanvree
Apr 12 2017 14:05
derp
update the damn wiki then bro!!!
ha
Andrew Newell
@scytherswings
Apr 12 2017 14:05
It's the one in startserver.sh
Gah
Sean Vreeland
@seanvree
Apr 12 2017 14:05
wait, what "One"?
O
I"m looking at startserver.sh now
last line:
mkdir -p tmp/pids && touch tmp/pids/puma.pid && exec bundle exec puma -C config/puma.rb config.ru
or this:
bundle exec puma -C config/puma.rb config.ru
Sean Vreeland
@seanvree
Apr 12 2017 14:24
@scytherswings
Andrew Newell
@scytherswings
Apr 12 2017 14:35
Yes that bundle exec command is correct
Sean Vreeland
@seanvree
Apr 12 2017 14:36
got it, but the puma.rb file still needs to be edited
just checking
Andrew Newell
@scytherswings
Apr 12 2017 14:42
No it should be fine now if you have the master version I think
Sorry at work right now lol
Ohhhh
Shit ok
Sorry, you're right. I'm crazy
My changes to better support windows haven't made it into master yet. They're still in Dev
Sorry
Sean Vreeland
@seanvree
Apr 12 2017 14:47
nah, I got ya
yeah, had to change the .rb file
fuck
okay
so I need to start over in dev ya dick?
sheesh
Andrew Newell
@scytherswings
Apr 12 2017 14:48
I'm on mobile at work, cut me some slack. Punk!
Lol
Sean Vreeland
@seanvree
Apr 12 2017 14:49
so, the ONLY manual change that needs to happen after pulling the dev branch is the CMD to run ? correct?
or has that been fixed in DEV?
Andrew Newell
@scytherswings
Apr 12 2017 14:50
I think you shouldn't need to change the scripts. But if you do then that would be the first thing to try
Sean Vreeland
@seanvree
Apr 12 2017 14:50
k
Andrew Newell
@scytherswings
Apr 12 2017 14:51
The puma.rb file should handle windows just fine now
Sean Vreeland
@seanvree
Apr 12 2017 14:51
got it
we'll see about that
Andrew Newell
@scytherswings
Apr 12 2017 14:51
Hah. Yeah
Sean Vreeland
@seanvree
Apr 12 2017 14:59
okay
got thru it!
wow
so, yeah, the last line on startserver.sh still will not work on windows
need to be this:
bundle exec puma -e production -C config/puma.rb config.ru
OORRR
Andrew Newell
@scytherswings
Apr 12 2017 15:00
Does windows not support touch ?
Sean Vreeland
@seanvree
Apr 12 2017 15:00
nope
Andrew Newell
@scytherswings
Apr 12 2017 15:00
Damn
Sean Vreeland
@seanvree
Apr 12 2017 15:00
you don't need all that jazz
tho
Andrew Newell
@scytherswings
Apr 12 2017 15:00
That would explain those errors you posted then
Sean Vreeland
@seanvree
Apr 12 2017 15:00
YUP
now
hold up
NOw, if you run SH startserver.sh like vefore
THEN BAM
it works
AND BONUS:
well, hang on, let me chick
Andrew Newell
@scytherswings
Apr 12 2017 15:03
Yo, I just added set -e to the dev serverSetup script. That will make it die if any command fails. Should help debugging I think
Sean Vreeland
@seanvree
Apr 12 2017 15:04
YAAAS
I figured it out
the service thing
god damn I"m good
fixed two major problems:
so somehow we will need to just edit the last line for windows
that's easy
Andrew Newell
@scytherswings
Apr 12 2017 15:05
Yeah that's not too bad
Sean Vreeland
@seanvree
Apr 12 2017 15:05
so, the script will run seperatly for win and linux
ANNNND:
Andrew Newell
@scytherswings
Apr 12 2017 15:05
I can do that
Sean Vreeland
@seanvree
Apr 12 2017 15:05
you ready?
service start:
Andrew Newell
@scytherswings
Apr 12 2017 15:06
Works as a service like you'd expect?
Sean Vreeland
@seanvree
Apr 12 2017 15:06
Nohup sh startserver.sh**
works like a gem
Andrew Newell
@scytherswings
Apr 12 2017 15:07
Daaaayuummm
Sean Vreeland
@seanvree
Apr 12 2017 15:07
so, as long as the startserver.sh works, then that will work
then if that works, then service starts
Andrew Newell
@scytherswings
Apr 12 2017 15:07
I'll add those fixes now
Sean Vreeland
@seanvree
Apr 12 2017 15:07
i'm on the new version
got it up
Made the changes to the windows google file
and I"m emailing you some recommeneded changaes to the wiki base dir
Andrew Newell
@scytherswings
Apr 12 2017 15:08
Cool. I'll have to wait till tonight actually, stuff to do at work
Sean Vreeland
@seanvree
Apr 12 2017 15:10
yup, no biggie buddy
Andrew Newell
@scytherswings
Apr 12 2017 15:12
Awesome, thanks for those instructions, I'll add them tonight
Sean Vreeland
@seanvree
Apr 12 2017 15:12
you'll also have a new name tonight...RIIIIIGHHHHT?
Andrew Newell
@scytherswings
Apr 12 2017 15:12
Ehhhhhhhhhhhhhhhhhhhhh welllllll
I'm not set on a new name yet. Gotta think of migration strategies too
Sean Vreeland
@seanvree
Apr 12 2017 15:53
OKay, yeah
we good! bro
question - what's the diff between asset host and webhost tho?
seems like one of those should be for the local IP
then the webhost SHOULD be whatever their reverse proxy is, right?
why do I have to set BOTH to the FQDN?
because I'm doing some thinking now
and before on the old build, my server config looked like this:
relative_root: /plexstatus
web_host: https://seanvree.com
asset_host: https://seanvree.com
so how did it even know what port to run on?
Andrew Newell
@scytherswings
Apr 12 2017 16:02
So asset host has to point to wherever the assets get served from, if you had, say, a CDN then they could be different
Sean Vreeland
@seanvree
Apr 12 2017 16:03
so, how does it know what port to serv on?
in the above example?
Andrew Newell
@scytherswings
Apr 12 2017 16:03
The web host is what the actual HTML and etc run from
Well
It should use 443 since you specified https
Sean Vreeland
@seanvree
Apr 12 2017 16:03
ahhhhh
I see
did you mess with anything from the previous builds with that?
when I change it to my hostname
exactly how I had it on the perious, it's not service up content
Andrew Newell
@scytherswings
Apr 12 2017 16:05
Somewhat, look at puma.rb and you'll see some logic that figured out what port bind puma to
What does the log message when you start the server say?
Sean Vreeland
@seanvree
Apr 12 2017 16:06
hang tight
Andrew Newell
@scytherswings
Apr 12 2017 16:06
It tells you what it is binding puma to
Sean Vreeland
@seanvree
Apr 12 2017 16:06
debuggin
Andrew Newell
@scytherswings
Apr 12 2017 16:06
Should be in production.stdout.log or production.log
Or both, can't remember
Sean Vreeland
@seanvree
Apr 12 2017 16:06
I, [2017-04-12T09:06:08.246982 #5356]  INFO -- : Setting asset_host to: https://seanvree.com
I, [2017-04-12T09:06:08.246982 #5356]  INFO -- : Setting relative_root to: /plexstatus
so it's running
but I"m getting this in my broswers:
plexstatus Failed to load resource: the server responded with a status of 502 (Bad Gateway)
Andrew Newell
@scytherswings
Apr 12 2017 16:07
The log you should look for is: Binding puma to:
I think it would be in the stdout log
Sean Vreeland
@seanvree
Apr 12 2017 16:08
na,
it's actulaly going to be nohup.log
in the root
hang tight tho
Andrew Newell
@scytherswings
Apr 12 2017 16:09
Ah, I hope it's in both
Not sure how nohup would affect it
Sean Vreeland
@seanvree
Apr 12 2017 16:13
because nohup intercepts the logs that would normall go in rails
that's how it works
yeah, it's not binding to the port
but I don't see any logs saying that
Andrew Newell
@scytherswings
Apr 12 2017 16:14
What port do you expect it to bind to?
Sean Vreeland
@seanvree
Apr 12 2017 16:14
443
because
all the PG logs say are : I, [2017-04-12T09:13:48.394351 #17588] INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T09:13:54.608140 #14896] INFO -- : Setting web_host to: https://seanvree.com
I, [2017-04-12T09:13:54.609121 #14896] INFO -- : Setting asset_host to: https://seanvree.com
I, [2017-04-12T09:13:54.609121 #14896] INFO -- : Setting relative_root to: /plexstatus
over and over
when nomrally it should be making API calls
and I tested it when I devinded the port, and IT works
so theres a bug somewhere
Andrew Newell
@scytherswings
Apr 12 2017 16:15
What does the production.stdout.log have in it? Is it empty?
Sean Vreeland
@seanvree
Apr 12 2017 16:15
hang on, gotta run for a sed
Andrew Newell
@scytherswings
Apr 12 2017 16:15
So it will work if you add :443 m
:443 I meant
Sean Vreeland
@seanvree
Apr 12 2017 16:16
no, I didn't put 443 in the config file
but when I put 3000 it works
Andrew Newell
@scytherswings
Apr 12 2017 16:18
There might be a permissions issue with 443. At least on Linux you have to be root to bind to ports less than 1000
Also there may be some SSL things. I haven't run. It on 443 myself yet
Possibly 443 is taken on your host I'd bet
There would be error logs in stderr
Hmm
Sean Vreeland
@seanvree
Apr 12 2017 16:33
yeah, trying with port 80
Sean Vreeland
@seanvree
Apr 12 2017 17:00
Yeah, okay
so it's not SSL
it's something weird about it binding to the port when using reverse proxy
we'll have to look at what changed from the last build I was on
it's weird cuz it has to listen on a port in order for reverse proxy to work
cuz that's the main setting you have to know when you put that into your websrver
Andrew Newell
@scytherswings
Apr 12 2017 17:13
Right but that's not surprising, it need to bind to something. In general 3000 should be good enough
Sean Vreeland
@seanvree
Apr 12 2017 17:17
right, but if you set the serverconfig to just your FQDN and the base URI
how would it know what port to bind to?
Andrew Newell
@scytherswings
Apr 12 2017 17:28
Well, ideally it falls back to 80 or 443 based on http or https
But being explicit is the best way to make things work as you expect
The example config files do show the port being set
Sean Vreeland
@seanvree
Apr 12 2017 17:31
yes they do
at least on dev they do
Andrew Newell
@scytherswings
Apr 12 2017 17:32
So just to make sure, when you set the port explicitly it works, right?
Sean Vreeland
@seanvree
Apr 12 2017 17:34
Yes
so, I have this right now:
relative_root: /plexstatus
web_host: http://192.168.1.20:3000
asset_host: http://192.168.1.20:3000
so that's NORMALL how it should look
Andrew Newell
@scytherswings
Apr 12 2017 17:35
Cool, that looks right to me
Sean Vreeland
@seanvree
Apr 12 2017 17:35
then you set your rev proxy to look at that addres
However
PB isn't serving it up because if you go in thru a rev proxy - PB sees the URI and regects it
"seanvree.com/ Failed to load resource: the server responded with a status of 502 (Bad Gateway)"
Andrew Newell
@scytherswings
Apr 12 2017 17:37
I need to do more reverse proxy testing. I suppose the issue you're seeing would be caused by the server generating links that always have 192.168.1.20 which won't work outside your local network
Stuff gets confusing
So you set it to http://seanvree.com:3000 what happens?
Sean Vreeland
@seanvree
Apr 12 2017 17:38
yeahhh, I"m not sure
umm, let me check
with no relative root?
Andrew Newell
@scytherswings
Apr 12 2017 17:38
Set that too
Since that's how you configured your proxy
/plexstatus right?
Sean Vreeland
@seanvree
Apr 12 2017 17:40
yeah, so seanvree.com:3000
then /plexstatus
Andrew Newell
@scytherswings
Apr 12 2017 17:42
What's your config look like now
Sean Vreeland
@seanvree
Apr 12 2017 17:43
relative_root: /plexstatus
web_host: http://seanvree.com:3000
asset_host: http://seanvree.com:3000
yeah, so it doesn't bind
Andrew Newell
@scytherswings
Apr 12 2017 17:43
Sweet
Does that work correctly?
Sean Vreeland
@seanvree
Apr 12 2017 17:43
nope
Andrew Newell
@scytherswings
Apr 12 2017 17:43
Whaaaaaa
Sean Vreeland
@seanvree
Apr 12 2017 17:44
it just keeps saying
I, [2017-04-12T10:43:28.701153 #22044]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:43:28.701153 #22044]  INFO -- : Setting relative_root to: /plexstatus
over and over
Andrew Newell
@scytherswings
Apr 12 2017 17:44
Every time you start the process it will log that
Your reverse proxy is configured to point at 3000, right?
Sean Vreeland
@seanvree
Apr 12 2017 17:47
yup
no, the application will continuously say that
it never loads the app
it just keeps saying that over and over
till you sotp it
just posted #86
Andrew Newell
@scytherswings
Apr 12 2017 17:48
Hmmm, ok I'll dig into that tonight. I'm not sure how the process would get restarted over and over
Sean Vreeland
@seanvree
Apr 12 2017 17:49
it's failin gto bind on that port, so it's just keeps trying
is my guess
I, [2017-04-12T10:45:51.790542 #23248]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:45:51.790542 #23248]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:45:51.791497 #23248]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:45:57.952294 #18560]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:45:57.952294 #18560]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:45:57.952294 #18560]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:03.531266 #17672]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:03.532248 #17672]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:03.532248 #17672]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:09.373729 #17352]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:09.373729 #17352]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:09.373729 #17352]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:15.186522 #22376]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:15.186522 #22376]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:15.186522 #22376]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:20.867713 #25260]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:20.867713 #25260]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:20.867713 #25260]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:26.595711 #12724]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:26.595711 #12724]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:26.595711 #12724]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:32.275659 #6536]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:32.276638 #6536]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:32.276638 #6536]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:37.700616 #22688]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:37.700616 #22688]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:37.700616 #22688]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:43.378599 #24364]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:43.378599 #24364]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:43.378599 #24364]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:48.822643 #2684]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:48.822643 #2684]  INFO -- : Setting asset_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:48.822643 #2684]  INFO -- : Setting relative_root to: /plexstatus
I, [2017-04-12T10:46:54.564414 #16664]  INFO -- : Setting web_host to: http://seanvree.com:3000
I, [2017-04-12T10:46:54.564414 #16664]  INFO -- : Setting asset_host to: http://seanvree.com:3000
so every 5 secons exactly
Andrew Newell
@scytherswings
Apr 12 2017 17:50
What's the production.stderr.log have in it?
Sean Vreeland
@seanvree
Apr 12 2017 17:50
where's that file?
Andrew Newell
@scytherswings
Apr 12 2017 17:50
Under the log folder
Should be there with the other logs
Sean Vreeland
@seanvree
Apr 12 2017 17:50
don't have that
only have production.log
Andrew Newell
@scytherswings
Apr 12 2017 17:51
Haha
Sean Vreeland
@seanvree
Apr 12 2017 17:51
lol
Andrew Newell
@scytherswings
Apr 12 2017 17:51
I bet it's a file permission error
Sean Vreeland
@seanvree
Apr 12 2017 17:51
possible
Andrew Newell
@scytherswings
Apr 12 2017 17:51
Try creating a blank text file with that name
Sean Vreeland
@seanvree
Apr 12 2017 17:51
well, wait
Andrew Newell
@scytherswings
Apr 12 2017 17:51
Same with production.stdout.log
Sean Vreeland
@seanvree
Apr 12 2017 17:51
no, it's not a permission issue if it can wrtie to the pro.log file
those errors I posted above are from the pro.log file
but I"ll humor you
Andrew Newell
@scytherswings
Apr 12 2017 17:52
Kk
Sean Vreeland
@seanvree
Apr 12 2017 17:52
production.stderr.log ?
Andrew Newell
@scytherswings
Apr 12 2017 17:52
Yep
Sean Vreeland
@seanvree
Apr 12 2017 17:52
C:\Sites\Plex-Board\log ?
Andrew Newell
@scytherswings
Apr 12 2017 17:52
Yep
Make the stdout one too
Please
Sean Vreeland
@seanvree
Apr 12 2017 17:53
man, I love this service man!
so easy to restart the app!
Andrew Newell
@scytherswings
Apr 12 2017 17:53
Also, what's the command you're using to start the server again? It should have -e production
Sean Vreeland
@seanvree
Apr 12 2017 17:53
no more typing!
yaya
no commands buddy
just click "restart service"
hah ah aha h
DOOOOPE
Andrew Newell
@scytherswings
Apr 12 2017 17:54
Oh shiiit I think something might affect the environment variables on Windows with nohup
Sean Vreeland
@seanvree
Apr 12 2017 17:54
possible
but, just so you know
if I set the server_config to default
everything works fine on the local box
Andrew Newell
@scytherswings
Apr 12 2017 17:55
RAILS_ENV should be set to production
Sweeeeeet
So at least the default config will work for non-proxy users
Sean Vreeland
@seanvree
Apr 12 2017 17:55
but it doesn't work when set for rev proxy
right
it will
Andrew Newell
@scytherswings
Apr 12 2017 17:56
We'll fix that tonight for sure
Sean Vreeland
@seanvree
Apr 12 2017 17:56
coo
Andrew Newell
@scytherswings
Apr 12 2017 17:56
Didn't think I had a regression but there's a lot of plates spinning
Sean Vreeland
@seanvree
Apr 12 2017 17:56
Yeah, so I created those file, still nottaa
Andrew Newell
@scytherswings
Apr 12 2017 17:56
Can you set an environment variable
Sean Vreeland
@seanvree
Apr 12 2017 17:56
at work?
Andrew Newell
@scytherswings
Apr 12 2017 17:56
I am lol
On my phone
Sean Vreeland
@seanvree
Apr 12 2017 17:57
doing what?
Andrew Newell
@scytherswings
Apr 12 2017 17:57
Well... Waiting for a test to run
Sean Vreeland
@seanvree
Apr 12 2017 17:57
na, like what do you do
Andrew Newell
@scytherswings
Apr 12 2017 17:57
Automated testing
Lol
Sean Vreeland
@seanvree
Apr 12 2017 17:57
seems....automated
Andrew Newell
@scytherswings
Apr 12 2017 17:57
I write automated tests in Java
Tests to validate the functionality of our cloud platform
Sean Vreeland
@seanvree
Apr 12 2017 17:59
gotcha
okay, one question, then I gotta get working on some other shit
if I set web host and localhost to https://192.168.1.20:3000
wouldn't that generate an error on the local box because how does it validate certs?
Andrew Newell
@scytherswings
Apr 12 2017 18:00
It could
Because in order for rails to do SSL
Sean Vreeland
@seanvree
Apr 12 2017 18:00
cause the app doesn't have SSL caps, so
Andrew Newell
@scytherswings
Apr 12 2017 18:00
It needs to know what cert to use
Sean Vreeland
@seanvree
Apr 12 2017 18:00
exactly
Andrew Newell
@scytherswings
Apr 12 2017 18:00
Yeah
So terminating SSL at the proxy is the easiest thing
Sean Vreeland
@seanvree
Apr 12 2017 18:01
of course
that's how my whole rig is set up
Andrew Newell
@scytherswings
Apr 12 2017 18:01
Aye
Sean Vreeland
@seanvree
Apr 12 2017 18:01
but in all those applciations...I can always set the base URI
that's where this is a bit confusing
because rev proxy takes SSL 443, then forwards it to HTTP 3000, the reverses it and sends it back to the requestor
so, honestly, I don't know how it was working before on the old version
lol
Andrew Newell
@scytherswings
Apr 12 2017 18:03
Well, this is where I need to spend some time. Every deployment is different which makes the whole thing more confusing
I think before I was handling http:// or https:// as simple strings
Instead of looking at them and assuming port numbers like 443 or 80
So that's probably why it broke
Sean Vreeland
@seanvree
Apr 12 2017 18:05
yup
we need a way to set the port
I knwo that for a fact
Andrew Newell
@scytherswings
Apr 12 2017 18:05
I'm not satisfied with the way it handles reverse proxies and I know it shouldn't be this hard
I just need to do some more research
Sean Vreeland
@seanvree
Apr 12 2017 18:06
correction
so it's working like this:
relative_root: /plexstatus
web_host: http://192.168.1.20:3000
asset_host: http://192.168.1.20:3000
but I remember when you first put that in...you said something about that there's a limiation or something
Andrew Newell
@scytherswings
Apr 12 2017 18:07
Outside your network I think it won't work
Sean Vreeland
@seanvree
Apr 12 2017 18:07
someting about how it had to match the FQDN or it owuldn't serve up to external address?
yup
Andrew Newell
@scytherswings
Apr 12 2017 18:07
Yeah
Sean Vreeland
@seanvree
Apr 12 2017 18:07
that's exactly what's happening
which is wierd because how does it know if the rev proxy is in place?
Andrew Newell
@scytherswings
Apr 12 2017 18:07
Try setting the asset host to https://seanvree.com
It doesn't
That's also part of our problem
Sean Vreeland
@seanvree
Apr 12 2017 18:08
right, but doesn't it just searve it to the webserver?
i'm confused as to why it wouldn't wokr, like how does it know an external client is requesting?
that's the whole point of a rev proxy
Andrew Newell
@scytherswings
Apr 12 2017 18:09
The server generates links inside the HTML and JavaScript
Sean Vreeland
@seanvree
Apr 12 2017 18:09
ahhhh
yeah
I see
so when the go back, they don't work
got it
Andrew Newell
@scytherswings
Apr 12 2017 18:09
Those links have to be reachable from wherever you view the pages from
Yeah
Sean Vreeland
@seanvree
Apr 12 2017 18:09
so, the config we have isn't really set up for reverse proxy...it's just a hack kinda deal?
that's sound crazy ...is it ruby?
Andrew Newell
@scytherswings
Apr 12 2017 18:10
Lol yeah. That's why I want to fix it
Sean Vreeland
@seanvree
Apr 12 2017 18:10
or what appliation is shitting?
Andrew Newell
@scytherswings
Apr 12 2017 18:10
Not it's me
Sean Vreeland
@seanvree
Apr 12 2017 18:10
ahhh
gotcha
well get home ya bitch
ha
JK
coo, immma take off for a bit
Andrew Newell
@scytherswings
Apr 12 2017 18:10
I just don't have it configured properly
Sean Vreeland
@seanvree
Apr 12 2017 18:10
LMK when you around later
Andrew Newell
@scytherswings
Apr 12 2017 18:10
Lol
Will do
Sean Vreeland
@seanvree
Apr 12 2017 18:10
But, I posted the issue with all the info
but yeah, that's our issue
Andrew Newell
@scytherswings
Apr 12 2017 18:11
Thanka for all the help debugging
Good stuff
Sean Vreeland
@seanvree
Apr 12 2017 18:11
that limitation with external FQDN requesting assets
for sho
SERVICE IS DOPE!
esp when doing this
so easy
don't have to keep tying that horrible linux code
why I LOVE windows baby, point and click
fuck CLI
Andrew Newell
@scytherswings
Apr 12 2017 18:11
Hahaha
Sean Vreeland
@seanvree
Apr 12 2017 18:13
got two big bugs figured out today
dope
Andrew Newell
@scytherswings
Apr 12 2017 18:14
Hell yeah
Andrew Newell
@scytherswings
Apr 12 2017 21:54
I feel stupid for not searching this before
facepalm
this basically goes over all the stuff we struggled to figure out
goddammit
Andrew Newell
@scytherswings
Apr 12 2017 22:14
oooooh this could be relevant as well:
... you might need to up your startupTimeLimit if you are seeing IIS stop your Ruby processes that take too long to startup.
@seanvree i'm going to install IIS on my windows machine tonight and try out this installation article
Still won't be home for a few hours unfortunately
Sean Vreeland
@seanvree
Apr 12 2017 22:56
holy shit
wow
don't feel bad man
I'm an IIS geek and I didn't know about this
This looks dope tho
I"ll dig into it too
however, I DON"T think this will apply to us
it will to me since I run a shit ton of ruby/python apps thru my IIS webserver...