@psychobunny Thanks!

https://docs.nodebb.org/en/latest/widgets/create.html

I found ./nodebb setup resets Wiegets configuration from ACP GUI.

How do we preserve the Widgets configuration including backup/migration scenario??
Possibly config files?

One more thing.
I found email template(registration/password-change/digests) is controlled by nodebb-theme-* node_modules.
Do we need to create own ````nodebb-theme-*``` for email template?

@julian Just works as intended.
The approach here totally works.
As you kindly suggested, it's my mistake not to have set subdomain by DNS.
I somehow misunderstood Ngynx takes care of subdomainsDNS IP routing, my bad.
Thanks a lot.

@julian said:

Make sure the subdomain is routed properly via your DNS server (so that sub.mydomain.net goes to the same ip as mydomain.net),

Oh my... I totally miss it!

You have true insights always. Appreciated Julian.

@julian No, the browser says - Oops! Google Chrome could not find sub.mydomain.net

Currently, I'm trying to figure out how to observe nginex behavior with log.

So far, simply,
mydomain.net shows the frontpage of nginex.
mydomain.net/4567 shows the nodeBB properly.

Thanks

@psychobunny @julian

Thanks for this thread, and I think this kind of information can be placed in docs or wiki.

I use nginx now, and the default page of mydomain.net works, and mydomain.net:4567 works for nodeBB.

Now, I try to implement sub.mydomain.net proxy to 127.0.0.0:4567 (nodeBB), so I tried:

/etc/nginx/sites-available/subdomain

server {
      listen   myipaddress:80;
     server_name sub.mydomain.net;
     error_log ./log error;

     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://127.0.0.1:4567/;
        proxy_redirect off;

        # Sockect.IO Support
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "upgrade";
    }
}

But seems not working.

I properly set the symbolic link in /etc/nginx/sites-enabled/

ln -s /etc/nginx/sites-available/subdomain

So here is no problem, I think.

Please advice ,
Thanks.

PS. reading

https://media.readthedocs.org/pdf/nodebb/latest/nodebb.pdf
I've changed

server {
      listen   myipaddress:80;
//........

and
sudo /etc/init.d/nginx restart

The result is same...
sub.mydomain.net does not lead to mydomain.net:4567...

Ubuntu 14.04
nginx version: nginx/1.6.0

Thanks @julian , I understand that.
/etc/redis/redis.conf is configured as│dir /var/lib/redis
Actually, I sort of knew this, but if so, it does not make any sense dump.rdb appears under the home/username/ directory...

I'm new to nodeBB and Redis.
Having said that, I am so confused by Redis behavior along with nodeBB.

Reading the docs-Upgrading NodeBB
https://docs.nodebb.org/en/latest/upgrading/index.html

Backing up Redis
As with all upgrades, the first step is to back up your data! Nobody likes database corruption/misplacement.
All of the textual data stored in NodeBB is found in a .rdb file. On typical installs of Redis, the main database is found at /var/lib/redis/dump.rdb.

However, during heavy tweak/installation of nodeBB, sometimes, dump.rdb appears under my user's home root /home/ken/dump.rdb.
I try to figure out which dump.rdb is actually connected to nodeBB and updated, and observed the time-stamp with redis-cli save/bgsave.
The dump.rdb under the home is updated, and /var/lib/redis/dump.rdb is not updated.
Sometimes, redis-cli save reflect redis-server log, sometimes does not.

Does someone have same kind of trouble? If I miss something, please let me know.

Thanks.

@julian said:

No problem, gh#1592
So we'd add another level of users... those who are "confirmed".. or "verified"... the terminology itself is flexible.

Thanks for your prompt reaction.

As you mentioned at GitHub-issues, some counter system against spamming is must-have, and actually the most common SNS or BB employes the email verification to register.

I cloned the master edge and trying now, but Category Permissions is confusing I would suggest.

I observe that
checked = allow
unchecked = disallow

but, obviously, it does not work like this so far. In other words, the checking status does not reflect the permission status.
As a default nothing is checked, but in fact guest is restricted in some way.
When I start checking the Administrator, with logging out, the category disappear since guest has lost permission to view the contents even if their checking status is untouched.

I know this is under heavy development, but just let you know, the current checking status does not reflect the permission at all.

I wish I can help to code this section, but I'm new here, and wonder I can do right way.

Thanks.

@julian

Thanks for the detailed reply.
I will try v0.5.0 or the pre-released master.

I observe the most common BB systems around employ the level of users as below:

1. administrator (and modifiers with some limited administration privileges )
2. registered users (the registration is not complete until the user verifies their emails)
3. guests (just can view contents)

The concept of registered user(2) or registration with email verification is widely shared among SNS such as Twitter/Facebook/Google+.

The problem of the current nodeBB system as the default I think is the forum is way too open without any email verification, and irregular/uncommon.

Say, I have malicious intention, I can gain extra privileges with offering fake or random email addresses that actually does not exist.

So, I suggest "verified users" as you say should be registered users, and "unverified users" should simply remain as guests.
I see no points having registered-verified-guests users, and this idea would definitely confuse both administrators/developers (us) and users.

What do you think?

Regards.

Allow me for this basic question, but I really cannot find the solution so far.

In this nodeBB forum, I see Announcements category prohibit users to create New Topic, and I would like to manage like this for my own BB.

Looking at ACP, there's Category section, and I tweak Read/Write for Registered users, but the permission seems no change after all.
I see Thread-Tools and I completely understood the feature, but category control is total mystery to me so far. What do I miss?

Another question.

I've implemented emailer, and email confirmation process is successfully done so far.
However, I'm very uncomfortable with that Any users who register (or I would say just input info) without email-confirmation can create New Topic and post replies.
How come the BB is designed like this??

Is there any way to allow only email confirmed users can create Topic or post replies?

Please let me know.
Thanks.

Ok, the problem is SOLVED.

For nodeBB plugin, the current directory is not plugin(nodebb-plugin-foo/) but nodebb root.

So, I placed emailer-amazonses-config.json in nodebb root dir.

Then modified the pointer in the code of the plugin.

Thanks.

PS.

I hardcoded Credentials, then it works at least for the plugin initialisation and nodeBB ready.

Hard-Coding Credentials
http://docs.aws.amazon.com/AWSJavaScriptSDK/guide/node-configuring.html

So, probably
AWS.config.loadFromPath('./config.json');
This line somehow fails to import the credential info form the ./config.json = ( ~/nodebb/node_modules/nodebb-plugin-emailer-amazonses/config.json )

Any idea how to fix this?

@julian said:

Where does ./config.json point to, a config file in your plugin's root folder?

Yes.
./config.json to be pointed by AWS.config.loadFromPath('./config.json'); in library.js
is located on
~/nodebb/node_modules/nodebb-plugin-emailer-amazonses/config.json

{ "accessKeyId": "YOUR_AWS_API_ID",
  "secretAccessKey": "YOUR_SECRET",
  "region": "us-west-2" }

Hm, any chance you're not handling the err value when you make the call to aws using the aws-sdk module?

The current library.js is

var winston = module.parent.require('winston'),
    Meta = module.parent.require('./meta'),
    Emailer = {},
    ses;

Emailer.init = function (app, middleware, controllers)
{
  console.log('===========emailer-amazonses init============');
  var AWS = require('aws-sdk');
  AWS.config.loadFromPath('./config.json');
  ses = new AWS.SES({
      apiVersion: '2010-12-01'
  });
};
//.............
//...............

Thanks.

I'm trying to develop nodebb-plugin-emailer-amazonses
https://github.com/kenokabe/nodebb-plugin-emailer-amazonses

Without nodeBB, I managed to success to send email along with amazonSES.
The test code is
https://github.com/kenokabe/nodebb-plugin-emailer-amazonses/blob/master/testapp.js
with proper 'nodebb/node_modules/nodebb-plugin-emailer-amazonses/config.json'

However, once implemented as nodebb plugin,
error: Connection timed out after 1000ms
occurs.

In fact, commenting out

var AWS = require('aws-sdk');
    AWS.config.loadFromPath('./config.json');
    var ses = new AWS.SES({apiVersion: '2010-12-01'});

helps to succeed to install this plugin and launch nodeBB.
So, somehow AWS.config process online fails.

How does this happen? and how can I fix this?
Please advice. Thanks!

**Here's the log of ./nodebb dev
**
Launching NodeBB in "development" mode.
....
....
info: [plugins] Loaded plugin: nodebb-plugin-emailer-amazonses
....
===========emailer-amazonses init============
info: [sounds] Sounds OK
info: [meta/js] Compilation complete
info: [themes] Compiling templates
info: [meta/css] Done.
error: Connection timed out after 1000ms
TimeoutError: Connection timed out after 1000ms
at ClientRequest.<anonymous> (/home/ken/nodebb/node_modules/nodebb-plugin-emailer-amazonses/node_modules/aws-sdk/lib/http/node.js:51:34)
at ClientRequest.g (events.js:180:16)
at ClientRequest.EventEmitter.emit (events.js:92:17)
at Socket.emitTimeout (http.js:1797:10)
at Socket.g (events.js:180:16)
at Socket.EventEmitter.emit (events.js:92:17)
at Socket._onTimeout (net.js:326:8)
at Timer.unrefTimeout [as ontimeout] (timers.js:418:13)
info: [app] Shutdown (SIGTERM/SIGINT) Initialised.
info: [app] Database connection closed.
info: [app] Shutdown complete.

@julian Thank you. I understood. I will try another way.

According to the docs,
https://docs.nodebb.org/en/latest/upgrading/index.html

On typical installs of Redis, the main database is found at
/var/lib/redis/dump.rdb.

and the file actually is there.

So, what is the dump.rdb in the nodebb root dir?

Thanks.

I'm trying to establish my own nodeBB development cycle.

Here is my ideal plan.

1.Create my own custom-nodeBB repo in GitHub.
2.Git clone the repo to my local machine.
3.Copy everything from the official-nodeBB repo.
4.Develop/Configure my own custom-nodeBB by customizing the official-nodeBB at the local machine such as installing needed plugins (often install my own plugins), and configure everything at ACP, then test it completely.
6.Once the development and test are done, (in ideal scenario) Git push to master of custom-nodeBB in Github, also upload redis file for everything including the plugin configuration, and excluding the forum data.
7.Git clone custom-nodeBB to a remote server machine.
8. Download the redis file for everything including the plugin configuration, and excluding the forum data and place it properly.
9. Start production. and periodically back up the redis file for forum data.

This is my ideal scenario.

However, I am not sure how the nodebb-plugin-settings and data other than forum data are stored.
I suppose that the nodebb-plugin-settings and data other than forum data are stored in redis file, am I correct?
I would like to save and load separately for forum data and everything else, in other words, isolate forum data only.
Is it possible?

If there's a good documentation for detailed data back up, please let me know.

Thanks.

@julian said:

we're using 40mb

For this scale and period, the memory size seems very reasonable. Thanks for this great project @julian

How much RAM do we need for Redis of nodeBB, say, for instance, a very active forum such as this nodeBB community forum?