Setup ended by handing somebody a login form and an empty dashboard. Everything this application can do was one menu away, and which menu was theirs to discover. The first time the administrator signs in to a new installation they now land on a short ordered list of what is worth doing first — add a client, upload a file, group the people who get the same things, choose how the file lists and the email look, point it at a mail server, add the team, check the scheduler — each a link straight to the screen that does it. The list is filtered twice, and both filters matter. By permission, because a link that answers 403 is worse than no link. And by edition: a managed installation is not sent off to configure a mail server somebody else runs, to create staff accounts that are not its to create, or to check a scheduler it does not host. Those three drop out on Cloud and the other five remain. Two steps tick themselves, because the database can answer them: a client exists, a file exists. Nothing else is checkable without guessing — a theme that was never changed looks exactly like one chosen deliberately — and a tick meaning "we assume so" is worse than no tick. The invitation to the Discord is at the very bottom, after the list. Somebody who has just installed this came with a job in mind, and opening with a social invitation is the fastest way to lose them. The marker is raised where a first administrator comes into existence — the setup screen and `projectsend:admin`, so a container provisioned from environment variables is welcomed too — and it is false by default, so an installation that updates into this feature is not congratulated on an install it finished a year ago. RedirectToWhatsNew becomes RedirectToGreeting and answers for both: they are the same interruption, and a second middleware on the same route would have to know about the first to avoid arguing with it. Installing wins; release notes for a version you never ran are the wrong greeting. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
22 KiB
Installing ProjectSend
This guide is for installing ProjectSend manually on your own server, from the .zip file
published with each release.
If you can run Docker, use Docker instead — it is one command, and everything on this page (PHP extensions, the web server, the background worker, the scheduled tasks) is already wired up for you. See the Docker instructions, and DOCKER.md for keeping your database and uploads outside the containers. Come back here if Docker is not an option on your hosting.
The whole thing takes about ten minutes. You will need shell access to the server and the ability to create a database — this is not an install you can do over FTP alone.
What you need
| PHP | 8.4 or newer, both the command-line PHP and PHP-FPM |
| PHP extensions | bcmath ctype curl dom fileinfo filter gd iconv intl json ldap mbstring openssl pcntl pdo_mysql session simplexml tokenizer zip |
| Database | MySQL 8.0 or newer (we test on 8.4 LTS) |
| Web server | nginx, with PHP-FPM — see the note below |
| Disk space | The app itself is small; plan for whatever your users will upload |
A few notes on that list:
ldapis required even if you never use LDAP. One of the libraries ProjectSend depends on declares it, so PHP will refuse to start the app without it. On Debian/Ubuntu it isphp8.4-ldap; on RHEL-family systems,php-ldap.- nginx is not a preference, it is a requirement. See Why nginx — it is worth two minutes of reading before you commit to a server, because Apache cannot be made to work by configuring it differently.
- Redis is optional. The Docker setup uses it, but a manual install works fine with the database for sessions, cache and queues. If you already have Redis, see Optional extras below.
Why nginx
Your uploaded files do not live under public/. They sit in storage/app/files/, outside the web
root, where no URL can reach them — which is the whole point: a file is only yours to download if
ProjectSend says so, and a file sitting in a guessable public folder has already lost that
argument.
So every download has to pass through a permission check. The obvious way to do that is to let PHP read the file and echo it back to the browser, and that is what most PHP applications do. It works, and it is a bad idea at any real size: a single 5 GB download occupies a PHP process for its entire duration, so a handful of people downloading at once can exhaust every worker your server has while the CPU sits idle. Resumable downloads, byte ranges and progress bars all have to be reimplemented by hand, usually incorrectly.
ProjectSend does the other thing. PHP checks permissions, logs the download, and then answers with
an empty response carrying a header that says "nginx, please send this file." nginx streams the
bytes with the same code it uses for any static file — sendfile, byte ranges, resume support, no
PHP process held open — and the visitor never sees the real path. The header is
X-Accel-Redirect, and the matching location /protected-files/ block in
step 6 is marked internal, which is what stops anyone
from requesting that path directly.
Apache has no equivalent that ProjectSend can use. Apache's closest feature, mod_xsendfile,
reads a differently-named header (X-Sendfile) that ProjectSend does not send, and it is not
installed by default anyway. LiteSpeed has its own third spelling. On any of them the application
installs fine and every page works — you can log in, upload, manage clients, browse the library —
but every download returns an empty response or a 404, because nothing is listening for the
instruction PHP just gave. There is no setting to change; the header names simply do not match.
Two ways out, if nginx really is impossible on your hosting:
- Put nginx in front of Apache as a reverse proxy, serving
/protected-files/itself. This works but is more moving parts than just using nginx. - Store your files in S3-compatible object storage instead (see Storing files somewhere other than this server). Files kept there are never on your server's disk, so downloads become a signed, expiring redirect to the storage provider and the web server is not involved at all. This is a genuine, supported path — just decide it before people start uploading, not after.
Step 1 — Put the files on the server
Download projectsend-x.y.z.zip from the
releases page, upload it to your server, and
unpack it where you want the site to live:
cd /var/www
unzip projectsend-2.0.0.zip -d projectsend
cd projectsend
The release zip is ready to run — you do not need Composer, Node, or npm. Everything the app needs is already inside it.
Important: point your web server at the
public/folder inside this directory, never at the directory itself. Everything abovepublic/— your configuration, your database credentials, your users' uploaded files — is meant to be unreachable from the web. Step 6 covers this.
Step 2 — Create the database
Create an empty database and a user that owns it. From the MySQL shell:
CREATE DATABASE projectsend CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
CREATE USER 'projectsend'@'localhost' IDENTIFIED BY 'a-long-random-password';
GRANT ALL PRIVILEGES ON projectsend.* TO 'projectsend'@'localhost';
FLUSH PRIVILEGES;
Leave the database empty. ProjectSend creates its own tables in step 5.
Step 3 — Tell ProjectSend about your server
Copy the example configuration and open it in an editor:
cp .env.example .env
.env is a plain list of NAME=value lines. The example file is written for the Docker setup, so
there is a fair amount to change. Here is a complete, working configuration for a manual install —
paste it over the top of the file, then change the addresses and passwords to yours:
APP_NAME="ProjectSend"
APP_ENV=production
APP_KEY=
APP_DEBUG=false
APP_URL=https://files.example.com
APP_TIMEZONE=UTC
PROJECTSEND_EDITION=community
DB_CONNECTION=mysql
DB_HOST=127.0.0.1
DB_PORT=3306
DB_DATABASE=projectsend
DB_USERNAME=projectsend
DB_PASSWORD=a-long-random-password
SESSION_DRIVER=database
CACHE_STORE=database
QUEUE_CONNECTION=database
FILESYSTEM_DISK=local
The four that actually matter:
APP_URL— the address people will type to reach your site, includinghttps://. Links in emails are built from this, so getting it wrong means broken links in every notification.APP_DEBUG=false— leave it off. With debug on, an error page will show visitors parts of your configuration.DB_*— what you created in step 2.APP_KEY— leave it empty for now; step 5 fills it in.
The three database lines mean sessions, cache and background jobs all live in MySQL, so there is
nothing else to install. If you have Redis, see Redis — but do the install first.
If your site is served over HTTPS, also add:
SESSION_SECURE_COOKIE=true
And if there is a proxy, load balancer or CDN (Cloudflare, an nginx in front of another nginx)
between your visitors and this server, add TRUSTED_PROXIES too — the .env.example file explains
the format. Without it every visitor appears to come from the proxy, which breaks per-visitor rate
limiting and makes the download log useless.
You can ignore the mail settings for now — email is configured from inside the app once you are logged in. See Sending email.
Step 4 — Set the file permissions
ProjectSend writes to two folders: storage/ (uploaded files, logs, sessions, cache) and
bootstrap/cache/. Both need to be writable by the user your web server runs as — usually
www-data, sometimes nginx.
sudo chown -R www-data:www-data /var/www/projectsend
sudo chmod -R 775 /var/www/projectsend/storage /var/www/projectsend/bootstrap/cache
Step 5 — Prepare the application
Three commands. Run them from the install directory, as the web server's user, so that everything they create ends up with the right owner:
sudo -u www-data php artisan key:generate
sudo -u www-data php artisan migrate --force
sudo -u www-data php artisan storage:link
What they do: the first generates the secret key used to encrypt sessions and cookies (it writes
itself into your .env); the second creates all the database tables; the third makes public assets
like your logo reachable from the web.
Keep a copy of
APP_KEYwith your backups. Anything encrypted with it — including saved credentials for your mail server — cannot be read back without it.
Step 6 — Point your web server at it
A complete nginx server block. Change server_name, and change /var/www/projectsend to wherever
you unpacked the files (there are three places, including one inside /protected-files/):
server {
listen 80;
server_name files.example.com;
root /var/www/projectsend/public;
index index.php;
client_max_body_size 100m;
add_header X-Content-Type-Options "nosniff" always;
add_header X-Frame-Options "SAMEORIGIN" always;
add_header Referrer-Policy "strict-origin-when-cross-origin" always;
location / {
try_files $uri $uri/ /index.php?$query_string;
}
# How downloads are served: ProjectSend checks permissions, then asks
# nginx to send the file. This block must NOT be reachable directly —
# "internal" is what guarantees that, so do not remove it.
location /protected-files/ {
internal;
alias /var/www/projectsend/storage/app/files/;
add_header X-Content-Type-Options "nosniff" always;
add_header X-Frame-Options "SAMEORIGIN" always;
add_header Referrer-Policy "strict-origin-when-cross-origin" always;
add_header Content-Security-Policy "sandbox; default-src 'none'" always;
}
location ~ \.php$ {
try_files $uri =404;
fastcgi_pass unix:/run/php/php8.4-fpm.sock;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME $realpath_root$fastcgi_script_name;
include fastcgi_params;
fastcgi_buffer_size 32k;
fastcgi_buffers 8 32k;
}
location ~ /\.(?!well-known) {
deny all;
}
}
Then check your PHP settings. Large uploads are sent in 20 MB pieces, so PHP never has to handle a
whole 5 GB file at once — but the pieces still need room. In your php.ini:
upload_max_filesize = 100M
post_max_size = 100M
memory_limit = 256M
Reload both services:
sudo nginx -t && sudo systemctl reload nginx
sudo systemctl restart php8.4-fpm
Set up HTTPS while you are here — certbot issues a free certificate and edits the nginx config for you.
Step 7 — Create your administrator
Open your site in a browser. Because no account exists yet, every address takes you to the setup screen, which asks for a site name and the name, email and password of the first administrator. Fill it in, and you are done. The first time you sign in, ProjectSend opens on a short list of the things worth doing first — adding a client, uploading a file, choosing how your file lists and your email look — each one linking straight to the screen that does it. It appears once; afterwards it lives at About → Getting started.
If you would rather not do it in the browser (or you are scripting the install), the same thing from the command line:
sudo -u www-data php artisan projectsend:admin
Two things to finish
The app runs without these, but parts of it will quietly not work. Both take a minute.
The background worker
Sending email and building zip archives of multiple files happen in the background, so nobody sits
watching a spinner. Something has to actually run that work. Create
/etc/systemd/system/projectsend-worker.service:
[Unit]
Description=ProjectSend queue worker
After=network.target
[Service]
User=www-data
Group=www-data
Restart=always
WorkingDirectory=/var/www/projectsend
ExecStart=/usr/bin/php artisan queue:work --tries=3 --backoff=3
[Install]
WantedBy=multi-user.target
Then:
sudo systemctl enable --now projectsend-worker
Without this, no email is ever sent and zip downloads never finish. Restart=always matters
too: saving your email settings restarts the worker so it picks up the new values, and it needs to
come back on its own.
The scheduled tasks
A handful of daily housekeeping jobs — deleting expired files, cleaning up abandoned uploads,
checking for new ProjectSend versions. Add one line to the web user's crontab
(sudo crontab -u www-data -e):
* * * * * cd /var/www/projectsend && php artisan schedule:run >> /dev/null 2>&1
Yes, every minute. ProjectSend decides internally what is actually due; the cron entry just gives it a heartbeat.
To check that both of these are working, log in and open System → Settings → Scheduler. It lists every task, when it last ran and whether it succeeded — along with any background job that failed. If the page says nothing has ever run, your cron line is not firing.
Optional extras
Sending email
Log in, go to System → Settings → Email, and enter your SMTP server's details there. That is
the place to configure it — the settings screen also has a "send test email" button, which will
save you a lot of guessing. The MAIL_* values in .env are only used until you fill that screen
in.
Redis
If you have Redis available, it is faster than the database for sessions, cache and queues. Install
the redis PHP extension and change three lines in .env:
SESSION_DRIVER=redis
CACHE_STORE=redis
QUEUE_CONNECTION=redis
Restart PHP-FPM and the background worker afterwards.
Add REDIS_HOST, REDIS_PORT and REDIS_PASSWORD if it is not a default local install. Restart
the worker afterwards.
Storing files somewhere other than this server
Out of the box, uploads live in storage/app/files/ on this machine. You can point ProjectSend at
S3-compatible object storage instead from System → Settings → Storage — useful when the files
outgrow the server's disk.
Making it faster
For a busy install, let PHP pre-compile the app's routes and views:
sudo -u www-data php artisan route:cache
sudo -u www-data php artisan view:cache
sudo -u www-data php artisan event:cache
You only run these once: projectsend:update notices they are in place and rebuilds them for you
after every update. If you change your mind, php artisan optimize:clear undoes all three.
One command to skip: config:cache
Every Laravel deployment guide on the internet lists php artisan config:cache alongside those
three, and php artisan optimize runs it for you. Don't — not on this application.
Here is why. Caching the configuration writes every resolved setting into one PHP file, and from
then on the framework stops reading your .env at all, on the entirely reasonable grounds that
everything in it has already been baked in. That holds for settings read the normal way, through
config(). ProjectSend reads one value earlier than that — TRUSTED_PROXIES, which has to be
known before the middleware stack is assembled, so it is read straight from the environment. Cache
the config and that read returns nothing.
Nothing breaks loudly. The site comes up, you log in, everything looks fine. But if there is a proxy or CDN in front of the server, ProjectSend goes back to believing every visitor is the proxy: the login rate limiter now counts all of your users as one attacker and locks the whole site out after five wrong passwords, and every row in the download log records the proxy's address instead of the person who actually downloaded the file. Both are the kind of thing you discover weeks later, from a complaint.
If you have already run it — or ran php artisan optimize, which includes it — php artisan config:clear puts things back immediately, and every update clears it too, saying why. The three commands above are safe and give you nearly
all of the speed anyway; config:cache was always the smallest win of the four.
Updating to a new version
cd /var/www/projectsend
sudo ./update.sh
The script ships with every release, in this directory. It asks whether to check GitHub for a newer
version, whether to download it (checking the published checksum), and whether you have a backup —
then takes the site down, unpacks the release, migrates the database, rebuilds whichever caches you
were using, reloads PHP-FPM, restarts the worker and brings the site back up. Your .env, your
uploads and your public/storage link are never touched.
sudo ./update.sh --zip ~/projectsend-2.1.0.zip applies a zip you downloaded yourself, and
./update.sh --check just reports what is available.
UPDATE.md is the full reference: what it does in order, every option, the same steps done by hand, how to check it worked, and how to go back. Read it before your first update — particularly the part about reloading PHP-FPM, which is the step that decides whether an update takes effect at all.
Back up first, every time. The script can dump the database for you (--backup), but your uploaded
files in storage/app/files/ are yours to look after.
When something goes wrong
A page that says "ProjectSend is not configured yet."
This is not an error — it is ProjectSend telling you which setup step is still missing. You reached
the site before creating your .env (step 3) or before generating APP_KEY (step 5). The page
names the exact command to run; do that, then reload.
Every page is blank, or shows a 500 error.
Look in storage/logs/ — open the newest file, the real error is at the bottom. Nine times out of ten it is
folder permissions (step 4) or a wrong database password (step 3).
"Please provide a valid cache path" or "failed to open stream".
storage/ or bootstrap/cache/ is not writable by the web server user. Step 4.
php artisan says "Table 'sessions' doesn't exist".
Something reached the database before migrate created its tables. Finish step 5 in order —
key:generate, then migrate, then storage:link.
Every address redirects me to the setup screen.
That is correct behaviour until the first administrator exists. Finish step 7. If you have already
created one and it still happens, ProjectSend cannot reach your database — check storage/logs/.
Pages load but downloads give a 404, or download a 0-byte file.
The /protected-files/ block is missing from your nginx config, or its alias path does not match
where you installed ProjectSend. It must point at storage/app/files/ and end with a slash. If you
are on Apache or LiteSpeed, no configuration will fix this — see Why nginx.
Uploads fail partway through.
client_max_body_size in nginx, or upload_max_filesize / post_max_size in php.ini, is
smaller than a 20 MB upload piece. Step 6.
No email arrives, and the test email button says it worked. "It worked" means it was queued, not delivered. The background worker is not running — see The background worker.
The CAPTCHA is stopping people signing in, and I cannot get in to switch it off.
It should not be able to: a wrong secret key or an unreachable provider both let people through and
report the problem on the settings screen instead. If you are stuck anyway, add
PROJECTSEND_CAPTCHA_DISABLED=true to .env and run php artisan optimize:clear, or run
php artisan projectsend:captcha-off. Your keys are kept either way. To check a key without
locking anything, php artisan projectsend:captcha-test asks the provider directly.
Links and redirects drop the port number (you are served on :8080, and the site sends you to
port 80).
Recent Debian and Ubuntu nginx packages set HTTP_HOST to $host in /etc/nginx/fastcgi_params,
deliberately — it stops a client-supplied Host header reaching the application — and $host
carries no port. On 80 or 443 that changes nothing. On any other port, every absolute URL the
application builds loses it. Add this to the location ~ \.php$ block, after include fastcgi_params;:
fastcgi_param HTTP_HOST $http_host;
On nginx 1.30 and later the safer form is $host$is_request_port$request_port. Either way this only
applies to a non-standard port; behind a TLS proxy on 443 you do not need it.
Emails and links point at localhost or the wrong domain.
APP_URL in .env. Fix it and run php artisan optimize:clear.
A change I made in .env has no effect.
Run php artisan optimize:clear, then restart PHP-FPM and the worker. Both hold the old values
until they are restarted. If it still has no effect, someone has run php artisan config:cache
(or optimize) on this install — see One command to skip.
Everyone is locked out of the login form at once, or the download log shows the same IP for
every download.
ProjectSend is seeing your proxy or CDN instead of your visitors. Set TRUSTED_PROXIES in .env
(step 3) — and make sure config:cache has not been run, which stops that value from being read
at all. Same section as above.
Still stuck? Ask in the community forum or open an issue on
GitHub, and include the last few lines of
the newest file in storage/logs/ — it is almost always the fastest way to an answer.