I am using multer to upload upload and saves images in client/images/post folder.
Here is a code snippet:
const storage = multer.diskStorage({
destination: './client/images/post',
filename: function(req, file, cb){
cb(null,file.fieldname + '-' + Date.now() + path.extname(file.originalname));
}
});
This works perfect in development.
However, when I deploy this node app to heroku,
images that are newly uploaded during production can't be found by Heroku.
I get this error:
https://mywebsite/images/post/photo-1673474603560.png 404 (Not Found)
However, if I look at source of other image tags that already existed when deployed - they are able to get into the folder and load their images.
Any ideas?
I believe Heroku uses an ephemeral filesystem. Please refer https://devcenter.heroku.com/articles/dynos#ephemeral-filesystem
Because of this any file/files that is created on the Heroku Dyno is not permanent. You would most likely have to go with other storage solutions if you want proper persistent file storage. While the current setup you have may work for local development, you would have to go for something like S3 in the deployed environment.
Related
How do I fetch a file from local machine with its file name and dowloand the same in Node.js without using third-party libraries?
I have seen this answer similar to this - How to download a file with Node.js (without using third-party libraries)?
but it shows to download using URL, but I need to fetch and download from my local, is it possible ?
node version I'm using is - 6.10.0
Not sure why you need "download" from your "local" - you're there already no?
If you need to read from your local in order to process or move it somewhere look at the File System module
https://nodejs.org/api/fs.html#fs_file_system
var fileStream = fs.createReadStream(filePath);
then you can do something with that file like upload it elsewhere
s3.upload({ Bucket: bucketName, Key: keyName, Body: fileStream});
hth
I am trying to upload my site. it's a node.js repo on a linux instance on amazon light sail services.
I uploaded my source code repository to my Virtual Server Instance with static IP using FileZilla through a SSh pem key(all fine i guess..), also I modified the file for the apache server as mentioned in the walk-though tutorial but, I am not able to make it show when I write http://54.87.121.120/spanishwithalex/server/
Before I upload my repo I tried with a simple app that worked perfectly,but now i don't know why it doesn't work. I tried various ports, different locations and various forms but I think I am walking in circles.
I attached some pictures to show where I located my files and the error messages.
I hope it's some how clear.
Thanks
location in my instance
structure repo
error message
apache conf
I'm not great on apache config, but I walked through this enough to get you going:
SFTP or SCP your code to a new folder called /opt/bitnami/apache2/spanishwithalex
Edit /opt/bitnami/apache2/conf/bitnami/bitnami.conf using VIM or whatever and replace every line that has this /opt/bitnami/apache2/htdocs
with this: /opt/bitnami/apache2/spanishwithalex
Restart Apache:
sudo /opt/bitnami/ctlscript.sh restart apache
Let me know if that gets it working for you.
I want to know how to download the files to aws-s3 through heroku, and dosen't use the web page.
I'm using wget in Nodejs to download audio files, and wanting to re-encoding them by ffmpeg. Though I have to store the file first, but heroku doesn't provide the space to store.
Then I find the way that heroku can connect with aws-s3, however the sample code is the page for user to upload, and I'm just want to using codes.
This is the code that haven't connected with s3 yet.
const wget = spawn('wget',['-O', 'upload_audio/'+ sender_psid +'.aac', audioUrl]);
//spit stdout to screen
wget.stdout.on('data', function (data) { process.stdout.write(data.toString()); });
// //spit stderr to screen
wget.stderr.on('data', function (data) { process.stdout.write(data.toString()); });
wget.on('close', function (code) {
console.log(process.env.PATH);
const ffmpeg = spawn('ffmpeg',['-i', 'upload_audio/'+ sender_psid +'.aac', '-ar', '16000', 'upload_audio/'+ sender_psid +'.wav', '-y']);
...
});
And it comes the error:
upload_audio/1847432728691358.aac: No such file or directory
...
Error: ENOENT: no such file or directory, open 'upload_audio/1847432728691358.wav'
How can I solve the problem?
Could anyone give some suggest plz?
Maybe won't use s3 can get it?
I don't recommend using S3 until you've tried the NodeJS Static Files API. On the server, that looks like this:
app.use(express.static(path.join(__dirname, 'my_audio_files_folder')));
More on NodeJS's Static Files API available here.
AWS S3 seems to be adding unneeded complexity, but if you have your heart set on it, have a look at this thread.
I am working on an addition to a project to create dynamic HTML reports from data that is stored in a SQLite database. Initially, I tried to do everything client-side using things like browserify and sql.js, but I ran into a lot of issues trying to read from the .db file locally.
For that reason, I have now decided to spin up a very basic web server that will be locally hosted. Essentially, I want the user to be able to navigate to http://localhost:3000 and hit a landing page which is the home page of the report.
I have set up a very basic HTTP server using express with the following code running in node:
var express = require('express');
var app = express();
app.use(express.static(__dirname + '/public'));
app.listen(process.env.PORT || 3000);
This works fine, and exposes the /public directory on port 3000, which has a placeholder index.html as of right now. My problem is, that when I try to start adding my code that reads from the SQLite database, none of the necessary require() functions work (specifically, require('fs'), due to it not being defined.
At a basic level, my question boils down to this:
How can i have the ability to read from the SQLite database file in the HTML/Javascript pages that live on the webserver? Whenever I try to use the necessary functions, it tells me that require() is not defined, or other similar errors.
Any help would be appreciated.
So I'm trying to get chat working on my website, and when I was testing locally it worked great, because port 8080 on my localhost was available and all that good stuff. But now I pushed my code to my Heroku app, and when I try and load my chat page, I get the error stating that it can't get localhost:8080/socket.io/socket.io.js.
I've seen node.js /socket.io/socket.io.js not found
and tried the suggestions, but none worked, even moving the socket.io.js file into a resource file did not work. I'm guessing this is because I'm using express 4.0?
Any help would be appreciated
Thanks
Edit:
So to add more details, since my question could seem a little vague, here is my relevant app.js code:
var client = require('socket.io').listen(8080).sockets;
In my jade file for the chat page, I have:
script (src = `'http://localhost:8080/socket.io/socket.io.js`')
and later on
var socket = io.connect(`'http://localhost:8080`');
and all this works on localhost (I load up on port 5000, socket.io is connected to port 8080). I do this using 'foreman start' with the heroku toolbelt.
When I try and change these to work on heroku, it breaks and I'm not sure how to fix it. I hope this clarifies the question a bit.
Edit 2:
I'm running:
express 4.0.0
socket.io 0.9.16
node 0.10.x
Thanks
Do you have an explicit route in express which catches all other routes? Something like this perhaps:
app.get("/", handlers.home);
app.get("/..." ...);
...
app.get("*", handlers.error);
This might keep socket.io from being able to host it's own js file for the client. There is an easy way to fix this, since you probably already have a public or static folder setup in express. Something like:
app.use(express.static("public"));
Make a new folder called socket.io and copy over the appropriate socket.io.js file into said folder, and all should be well. However note that there are two files named socket.io.js!! So, if you see something like "Uncaught ReferenceError: require is not defined" it means you copied the "node-ey" server side file. Here is the correct client file to copy:
app_dir/node_modules/socket.io/node_modules/socket.io-client/dist/socket.io.min.js
Note #BHendricks: I would have just posted as a reply to your comment, but I currently lack the required reputation.
Edit:
The OPs question probably has more to do with the "localhost" issue. When connecting from a client (say your home IP), as far as your browser knows - localhost implies a connection with the machine which is locally hosting stuff. Since your home machine (or phone) does not host socket.io, this is failing.
What you need to do is have your server embed the socket connection information (either a fully qualified hostname, ip etc). This can be done when the server "renders" the page with the client connection.
What happens when you go to http://localhost:8080/socket.io/socket.io.js?
Does it 404? If it does you need to make sure you have it in a directory that Express is set to serve statically.
app.use(express.static(__dirname + '/public'));
Then put your socket.io.js file in public/socket.io/socket.io.js (relative to your app.js file)
Restart your server and see if that fixes it.
Basically, Express doesn't serve files statically from the file system unless you explicitly tell it where to map from.