ApacheBenchmark ab issue with Rails3 authenticity_token - ruby-on-rails-3

I was trying to play around with AB for performance recording/tracking in a new Rails 3 app.
Since the app always require to be logged in, I had to do POST request to login.
I was able to put the credentials in some text file as follows but the AuthenticityToken is giving the problem.
#login_data.txt
user_account%5Busername%5D=admin&user_account%5Bpassword%5D=adminhr
#AB command
ab -v4 -n100 -t5 -T 'application/x-www-form-urlencoded' -p login_data.txt http://nhc.lvh.me:3000/
#The log
Started POST "/" for 127.0.0.1 at 2011-02-15 11:13:37 +0545
Processing by CompaniesController#index as */*
Parameters: {"user_account"=>{"username"=>"admin", "password"=>"[FILTERED]"}}
Completed in 1ms
ActionController::InvalidAuthenticityToken (ActionController::InvalidAuthenticityToken):
Rendered /Users/millisami/.rvm/gems/ruby-1.9.2-p136/gems/actionpack-3.0.0/lib/action_dispatch/middleware/templates/rescues/_trace.erb (1.0ms)
Rendered /Users/millisami/.rvm/gems/ruby-1.9.2-p136/gems/actionpack-3.0.0/lib/action_dispatch/middleware/templates/rescues/_request_and_response.erb (1194.0ms)
Rendered /Users/millisami/.rvm/gems/ruby-1.9.2-p136/gems/actionpack-3.0.0/lib/action_dispatch/middleware/templates/rescues/diagnostics.erb within rescues/layout (1291.3ms)
The problem is how to pass or ignore that AuthenticityToken when using ab?
Since it gets generated dynamically, can I store somewhere or is there other better solution?

What you could do is set up an environment aside from test and development. You could put a copy of development.rb environment and name it benchmark.rb.
Also put an entry on database.yml for the benchmark environment. You could copy the contents of the development entry for this.
And then in your code you could skip validation of authenticity token if the environment is benchmark. Now fire up rails s -e benchmark and perform your benchmark against that server.

Related

How to solve flutter web api cors error only with dart code?

It seems like CORS error is well-known issue in the web field. But I tried flutter web for the first time ever and I faced critical error.
The code below worked well in app version when it was running on iOS device, but when i tested the same code on Chrome with web debugging from beta channel, it encountered CORS error.
Other stackoverflow answers explained how to solve the CORS issue with serverside files of their projects. But I have totally no idea what is server thing and how to deal with their answers. The error message from Chrome console was below
[ Access to XMLHttpRequest at 'https://kapi.kakao.com/v1/payment/ready' from origin 'http://localhost:52700' has been blocked by CORS policy: Response to preflight request doesn't pass access control check: No 'Access-Control-Allow-Origin' header is present on the requested resource. ]
So, what i want to do is to solve above 'Access-Control-Allow-Origin header' issue ONLY WITH DART CODE! Code below is what i've tried to solve these issues only with my main.dart.
onPressed: () async {
var res =
await http.post('https://kapi.kakao.com/v1/payment/ready', encoding: Encoding.getByName('utf8'), headers: {
'Authorization': 'KakaoAK $_ADMIN_KEY',
HttpHeaders.authorizationHeader: 'KakaoAK $_ADMIN_KEY',
"Access-Control-Allow-Origin": "*",
"Access-Control-Allow-Methods": "POST, GET, OPTIONS, PUT, DELETE, HEAD",
}, body: {
'cid': 'TC0ONETIME',
'partner_order_id': 'partner_order_id',
'partner_user_id': 'partner_user_id',
'item_name': 'cool_beer',
'quantity': '1',
'total_amount': '22222',
'vat_amount': '2222',
'tax_free_amount': '0',
'approval_url': '$_URL/kakaopayment',
'fail_url': '$_URL/kakaopayment',
'cancel_url': '$_URL/kakaopayment'
});
Map<String, dynamic> result = json.decode(res.body);
print(result);
},
Even though i actually had the header "Access-Control-Allow-Origin": "*" which most other answers recommended, the Chrome console printed same error message. Weird thing is that the same code made successful request in mobileApp version. So I think this is only problem with flutter WEB VERSION.
Hope somebody can figure it out and suggest only-dart code to resolve the issue in my main.dart!! Thank you for reading [:
1- Go to flutter\bin\cache and remove a file named: flutter_tools.stamp
2- Go to flutter\packages\flutter_tools\lib\src\web and open the file chrome.dart.
3- Find '--disable-extensions'
4- Add '--disable-web-security'
Since Flutter 3.3.0
I implemented the option to add any browser flag to the flutter command.
flutter run -d chrome --web-browser-flag "--disable-web-security"
Or for drive command:
flutter drive --driver=test_driver/integration_test.dart --target=integration_test/app_test.dart -d web-server --web-browser-flag="--disable-web-security"
Note: This is just for development and testing. Flutter is executed explicitly on the client's browser! You should NOT and you can NOT disable it in production (as stated by #Tommy), as it is a security feature of the browser, and not meant to be changed in dart code. You have to enable CORS on your web server, which is providing the resources of your Flutter app, to ensure it works for everyone.
If you use dart language without Flutter on the server side with shelf, then see this response.
I think disabling web security as suggested will make you jump over the current error for now but when you go for production or testing on other devices the problem will persist because it is just a workaround, the correct solution is from the server side to allow CORS from the requesting domain and allow the needed methods, and credentials if needed.
run/compile your Flutter web project using web-renderer. This should solve the issue both locally and remotely:
flutter run -d chrome --web-renderer html
flutter build web --web-renderer html
This is a CORS (cross-origin resource sharing) issue and you do not have to delete/modify anything. You just have to enable the CORS request from your server-side and it will work fine.
In my case, I have created a server with node.js and express.js, so I just added this middleware function that will run for every request.
app.use(function(req, res, next) {
res.header("Access-Control-Allow-Origin", "*");
res.header("Access-Control-Allow-Methods", "GET,PUT,PATCH,POST,DELETE");
res.header("Access-Control-Allow-Headers", "Origin, X-Requested-With, Content-Type, Accept");
next();
});
And BOOOOM! I received the data.
You just have to look at the settings to enable CORS for your server.
Server side engine like node js or django is really needed to work with flutter web with bunch of external apis. Actually there's high possibility of same CORS error when we try to use internal api because of the CORS mechanism related to port number difference.
There are bunch of steps and answers from SO contributors that recommend to use chrome extensions to avoid CORS errors, but that is actually not cool for users. All the users should download the browser extensions to use the single website from us, which wouldn't be there if we used true server engines.
CORS is from browser as far as i know, so our flutter ios and android apps with same api code don't give those CORS errors. First time i encountered this error with flutter web, i believed i can deal with CORS in my app code lines. But that is actually not healthy way for users and long term dev plans.
Hope all flutter web newbies understand that web is quite a wild field for us. Even though i'm also newbie here, i highly recommend all the flutter web devs from 1.22.n stable to learn server side engines like node js. It is worth try.
And if u came so far down to this line of my self-answer, here's a simple guide for flutter web with node js. Flutter web is on stable channel but all those necessary infra are not fully ready for newbies like me. So be careful when you first dive into web field, and hope you re-check all the conditions and requirements to find out if you really need web version of your flutter app, and also if you really need to do this work with flutter. And my answer was yes lol
https://blog.logrocket.com/flutter-web-app-node-js/
If you run a Spring Boot server, add "#CrossOrigin" to your Controller or to your service method.
#CrossOrigin
#PostMapping(path="/upload")
public #ResponseBody ResponseEntity<Void> upload(#RequestBody Object object) {
// ...
}
I know the question explicitly asked for a solution "with dart code" only, but I was not able to fix the exception with dart code (for example by changing the header).
The disabling web security approaches work well in development, but probably not so well in production. An approach that worked for me in production dart code involves avoiding the pre-flight CORS check entirely by keeping the web request simple. In my case this meant changing the request header to contain:
'Content-Type': 'text/plain'
Even though I'm actually sending json, setting it to text/plain avoids the pre-flight CORS check. The lambda function I'm calling didn't support pre-flight OPTIONS requests.
Here's some info on other ways to keep a request simple and avoid a pre-flight request
https://docs.flutter.dev/development/platform-integration/web-images
flutter run -d chrome --web-renderer html
flutter build web --web-renderer html
This official solution worked for me on Chrome only (Source). But I had to run it first every time.
flutter run -d chrome --web-renderer html
And disabling web security also worked (Source). But the browsers will show a warning banner.
But In case you are running on a different browser than Chrome (e.g. Edge) and you want to keep 'web security' enabled.
You can change the default web renderer in settings in VS Code
File ==> Preferences ==> Settings ==> Enter 'Flutter Web' in the Search Bar ==> Set the default web renderer to html
After hours of testing, the following works perfectly for me.
Add the following to the PHP file:
header('Access-Control-Allow-Origin: *');
header('Access-Control-Allow-Methods: GET, POST');
header("Access-Control-Allow-Headers: X-Requested-With");
This allow the correct connection with the HTTP GET POST with no issue from flutter for me.
I discovered this in the following discussion:
XMLHttpRequest error Flutter
I am getting the same error with php api so i add the php code these lines ;
header('Access-Control-Allow-Origin: *');
header('Access-Control-Allow-Methods: GET, POST');
header("Access-Control-Allow-Headers: X-Requested-With");
I think you may not doing this in right way.
The cors headers should be added in HTTP response header while you added them in you reuqest header obviously.
for more information check out the documentation https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS#what_requests_use_cors
The below solution is great if you are only communicating with a local NodeJS server.
Install NodeJS
Create a basic NodeJS express project
Create a folder to put you NodeJS project in
ex: C:\node_project\
in PowerShell run: npm init in the folder
fill in your desired values
entry point: must be app.js for this example to work
in PowerShell run: npm install express in the folder
create a app.js file in the folder
// init express
const express = require("express");
const app = express();
// set the path to the web build folder
app.use(express.static("C:/Users/your_username/path_to_flutter_app/build/web"));
const PORT = process.env.PORT || 8080;
app.listen(PORT, () => {
console.log(`Server listening on port ${PORT}...`);
});
The value "C:/Users/your_username/path_to_flutter_app/build/web" must be changed to the web build folder in your flutter app.
The app can be accessed through your browser once the app is built, the node server is running, and the browser is at the correct address
Build the app
open PowerShell and navigate to the flutter project's root ex: C:/Users/your_username/path_to_flutter_app/
run flutter build web
turn on the node server
open PowerShell and navigate to the NodeJS server folder ex: C:\node_project\
run: node app.js
Open in your browser
Enter http://localhost:8080/ into the browser
Note that everytime you change your flutter app's dart code you will need to re-run flutter build web
Wrong Server on Target Port 🤦
I feel silly for even admitting this, but I had some other local server running on the targeted port. I've no clue why the server seemed to boot on the same port, or why the iOS app seemed to work, but now that I'm hitting the actual server it's working fine.
I was also getting some 404's mixed in, but originally thought that was due to the CORs error.
Maybe someone else has this same issue and this helps them.
In my case, The problem was in laravel backend code which did not support CORS, So I added the CORS into backend project then it worked successfully in test and live.
The 5th step of Osmans answer should be to add the option
'--disable-site-isolation-trials',
Only this works for me.
Chrome version 106.0.5249.119
Update:
I recommend to use User Rebo's answer. It is now possible to pass --disable-web-security as a browser flag to run & drive commands!
Original outdated answer:
Alternative solution for MacOS & Android Studio (without modifying Flutter source)
We use a similar approach as Osman Tuzcu. Instead of modifying the Flutter source code, we add the --disable-web-security argument in a shell script and just forward all other arguments that were set by Flutter. It might look overly complicated but it takes just a minute and there is no need to repeat it for every Flutter version.
1. Run this in your terminal
echo '#!/bin/zsh
# See also https://stackoverflow.com/a/31150244/410996
trap "trap - SIGTERM && kill -- -$$" SIGINT SIGTERM EXIT
set -e ; /Applications/Google\ Chrome.app/Contents/MacOS/Google\ Chrome --test-type --disable-web-security "$#" ; set +e &
PID=$!
wait $PID
trap - SIGINT SIGTERM EXIT
wait $PID
' > ~/chrome_launcher
chmod 755 ~/chrome_launcher
This adds a chrome_launcher script to your user folder and marks it executable.
2. Add this line to your .zshrc (or .bashrc etc.):
export CHROME_EXECUTABLE=~/chrome_launcher
3. Restart Android Studio
If a simple restart does not work, use Invalidate Caches / Restart in Android Studio to force loading of changes.
Notes
The script also adds the --test-type flag to suppress the ugly warning about the disabled security features. Be aware that this option might also suppress other error messages!
The CHROME_EXECUTABLE takes only the path to an executable file it is not possible to set arguments there.
Without trapping exit signals and killing the process group, the Google Chrome instance was not killed when you hit the Stop Button in Android Studio.
If you are using FVM, I suggest to use flutter_cors package
dart pub global activate flutter_cors
fluttercors --disable
If you face
zsh: command not found: fluttercors
You need to add it to PATH. In my case, I'm using zsh, I add it to .zshrc by
vim ~/.zshrc
Press I to start editing and paste export PATH="$PATH":"$HOME/.pub-cache/bin" to the top of the file
Then press ESC and type :wq to save the .zshrc file.
Now you're good to go
Now, just need to run your flutter web normally. It will trigger Chrome without CORS.
For me none of the solutions above worked on production as it was expected. Altough there is one solution I can suggest which uses CORS proxy to avoid CORS issues on flutter web on production. You can find CORS proxies on this website.
Basically you bypass all the unnecessary headers which your browser appends to your requests, so you may not encounter the same CORS issues when making request to another API. Hope it helps!
It Worked With Me By The Following Code :
in conn.php file put like this :
<?php
header('Access-Control-Allow-Origin: *');
header('Access-Control-Allow-Methods: GET, POST');
header("Access-Control-Allow-Headers: X-Requested-With");
$connect = new mysqli("localhost","db_user","db_password","db_name");
if($connect){
}else{
echo "Connection Failed";
exit();
}
This is a CORS (cross-origin resource sharing) issue and you just need to enable the CORS request from your server-side.
In my case it is Asp.Net MVC Web API and adding below code to Application_BeginRequest at Global.asax worked for me:
HttpContext.Current.Response.AddHeader("Access-Control-Allow-Origin", "http://localhost:7777/");
if (HttpContext.Current.Request.HttpMethod == "OPTIONS")
{
//These headers are handling the "pre-flight" OPTIONS call sent by the browser
HttpContext.Current.Response.AddHeader("Access-Control-Allow-Methods", "POST, GET");
HttpContext.Current.Response.AddHeader("Access-Control-Allow-Headers", "content-type");
HttpContext.Current.Response.End();
}
Use desired urls , Methods and Headers
Also There is no need to change anything in Web.config
If anyone looking for an equivalent of the accepted answer (Osman's) when working with dart web (webdev), here's what worked for me on Dart 2.17.6 (a bit more complex but in case you needed a quick fix, it might be handful).
Find webdev executable (this helps) then you see something like this:
The snapshot file (generated if not exist, as you see) is executed when you want to run app in browser. It contains the code that dart runs when launching chrome (using browser_launcher dart package).
Backup and remove the snapshot file (location in the screenshot above) so it can be regenerated in next run.
Locate browser_launcher package in your pub cache (also you might find location of browser_launcher by searching in the snapshot file) and edit lib\src\chrome.dart, find '--disable-extensions' and add '--disable-web-security'.
Run your app and remove the backup created in step 2.
If you are working with django in the side of the service, you can configure CORS with 'corsheaders', in this link you can find whole the documentation to setup your back end and recognice your requests.
https://pypi.org/project/django-cors-headers/
Go to flutter\bin\cache and remove a file named: flutter_tools.stamp
Go to flutter\packages\flutter_tools\lib\src\web and open the file chrome.dart.
Find '--disable-extensions'
Add '--disable-web-security'

How to locally run my cloudflare worker serverless function, during development?

I managed to deploy my first cloudflare worker using serverless framework according to
https://serverless.com/framework/docs/providers/cloudflare/guide/
and it is working when I hit the cloud.
During development, would like to be able to test on http://localhost:8080/*
What is the simplest way to bring up a local http server and handle my requests using function specified in serverless.yml?
I looked into https://github.com/serverless/examples/tree/master/google-node-simple-http-endpoint
but there is no "start" script.
There seem to be no examples for cloudflare on https://github.com/serverless/
At present, there is no way to run the real Cloudflare Workers runtime locally. The Workers team knows that developers need this, but it will take some work to separate the core Workers runtime from the rest of Cloudflare's software stack, which is otherwise too complex to run locally.
In the meantime, there are a couple options you can try instead:
Third-party emulator
Cloudworker is an emulator for Cloudflare Workers that runs locally on top of node.js. It was built by engineers at Dollar Shave Club, a company that uses Workers, not by Cloudflare. Since it's an entire independent implementation of the Workers environment, there are likely to be small differences between how it behaves vs. the "real thing". However, it's good enough to get some work done.
Preview Service API
The preview seen on cloudflareworkers.com can be accessed via API. With some curl commands, you can upload your code to cloudflareworkers.com and run tests on it. This isn't really "local", but if you're always connected to the internet anyway, it's almost the same. You don't need any special credentials to use this API, so you can write some scripts that use it to run unit tests, etc.
Upload a script called worker.js by POSTing it to https://cloudflareworkers.com/script:
SCRIPT_ID=$(curl -sX POST https://cloudflareworkers.com/script \
-H "Content-Type: text/javascript" --data-binary #worker.js | \
jq -r .id)
Now $SCRIPT_ID will be a 32-digit hex number identifying your script. Note that the ID is based on a hash, so if you upload the exact same script twice, you get the same ID.
Next, generate a random session ID (32 hex digits):
SESSION_ID=$(head -c 16 /dev/urandom | xxd -p)
It's important that this session ID be cryptographically random, because anyone with the ID will be able to connect devtools to your preview and debug it.
Let's also define two pieces of configuration:
PREVIEW_HOST=example.com
HTTPS=1
These specify that when your worker runs, the preview should act like it is running on https://example.com. The URL and Host header of incoming requests will be rewritten to this protocol and hostname. Set HTTPS=1 if the URLs should be HTTPS, or HTTPS=0 if not.
Now you can send a request to your worker like:
curl https://00000000000000000000000000000000.cloudflareworkers.com \
-H "Cookie: __ew_fiddle_preview=$SCRIPT_ID$SESSION_ID$HTTPS$PREVIEW_HOST"
(The 32 zeros can be any hex digits. When using the preview in the browser, these are randomly-generated to prevent cookies and cached content from interfering across sessions. When using curl, though, this doesn't matter, so all-zero is fine.)
You can change this curl line to include a path in the URL, use a different method (like -X POST), add headers, etc. As long as the hostname and cookie are as shown, it will go to your preview worker.
Finally, you can connect the devtools console for debugging in Chrome (currently only works in Chrome unfortunately):
google-chrome https://cloudflareworkers.com/devtools/inspector.html?wss=cloudflareworkers.com/inspect/$SESSION_ID&v8only=true
Note that the above API is not officially documented at present and could change in the future, but changes should be relatively easy to figure out by opening cloudflareworkers.com in a browser and looking at the requests it makes.
You may also be able to test locally by loading the Cloudflare worker as a service worker.
Note:
Use a local web server with https:. Workers won't load using file: or http: protocols.
Your browser will need to support workers, so you can't use IE.
Mock any Cloudflare-specific features, such as KV.
<!doctype html>
<html>
<head>
<meta charset="utf-8">
</head>
<body>
<!-- Service worker registration -->
<script>
if ('serviceWorker' in navigator) {
// Register the ServiceWorker
navigator.serviceWorker.register('/service-worker.js')
.then(
function(reg) {
// Registration succeeded
console.log('[registerServiceWorker] Registration succeeded. Scope is ' + reg.scope)
window.location.reload(true)
})
.catch(
function(error) {
// Registration failed
console.log('[registerServiceWorker] Registration failed with ' + error)
})
} else {
console.log('[registerServiceWorker] Service workers aren\'t supported')
}
</script>
</body>
</html>
Dollar Share Club created Cloudworker. It is not actively maintained, but it is a way to run Cloudflare Workers locally.
You can read about it on the Cloudflare blog in guest post by the original maintainer of Cloudworker.

Rails - 401 Unauthorized when I access action in Production only

I'm using Ruby On Rails 3.0.9 and everything works fine on Development env. When I switch to Production env, or I upload it to our server, after sign in I'm taken back to the same Login page. When I check the log, I can see the following:
Started POST "/users/login" for 127.0.0.1 at Thu Oct 03 16:48:13 -0300 2013
Processing by UserSessionsController#create as HTML
Parameters: {"user"=>{"password"=>"[FILTERED]", "login"=>"demo_admin"}, "utf8"=>"✓", "authenticity_token"=>"+7AEoVXZ9XiagEymVUnOhFHnck4rgDu883E/ySMlCxQ="}
Redirected to http://localhost:3000/admin
Completed 302 Found in 111ms
Started GET "/admin" for 127.0.0.1 at Thu Oct 03 16:48:13 -0300 2013
Processing by Admin::DashboardController#index as HTML
Completed 401 Unauthorized in 1ms
I'm using authorization_rules file in order to manage access, but I've got no problem on Dev env, as I said before.
If I place a breakpoint at the admin/dashboard#index action, it won't be executed, as it's not reached. It breaks at httpserver file (I debugged it step by step), but I cannot understand why it works on Dev and not on Prod env.
Please, help.
Thanks,
Brian
UPDATE
I forgor to mention that, in my ApplicationController, there's a before_filter called check_plan_features and the first thing it asks is unless current_user.blank? #redirects to Admin section.
I've noticed that after signing in, using Devise, current_user has the user's value, but when after redirecting to the admin section, it comes back to the same filter, and this time, the current_user is null. So, I assume that, somehow, the user's session is destroyed after trying to access Admin section. But, as this only happens on production environment, I'm still wondering what could be.
I just had this exact problem (using rails 4 and devise 3).
The solution was to add a domain declaration to the config/initializers/session_store.rb file like this:
YourAppName::Application.config.session_store :cookie_store, key: '_your_session', domain: {
production: 'production_domain',
development: 'development_domain'
}.fetch(Rails.env.to_sym, :all)
Though it is important that this was changed initially due to adding sub-domains to the application.

gitlab :import existing repository .not working correctly

Here is the details of my setup:
Gitlab version: 5.2
Operating System: Centos 6.3
I am Importing an existing repository while creating a new project(/projects/new) .
A new EMPTY project is created however no repository is imported and no error message comes up. While I was digging around I found this in my puma.stderr.log.
=== puma startup: 2013-06-13 15:37:53 +0530 ===
error: Couldn't resolve host 'github.com' while accessing https://github.com/<username>/tester.git/info/refs
This GitLab installation is behind a HTTP proxy server and it seems like the proxy settings aren’t configured. But, my /etc/profile.d/ has a script that sets the proxy system wide with variables http_proxy and https_proxy.
On further investigation I checked whether the problem could be with gitlab-shell unable to reach out to the URL via the proxy and I tried the following.
$./bin/gitlab-projects import-project xxx/tester_test_test.git https://github.com/<username>/tester.git
This seems to work perfectly.
It seems like Puma does not use this gitlab-shell and is trying to reach out to the URL. Which leads me to the question, how do I tell PUMA about my proxy server?
Following is my production.log
Started GET "/favicon.ico" for 170.95.35.204 at 2013-06-13 16:20:00 +0530
Processing by ProjectsController#show as HTML
Parameters: {"id"=>"favicon.ico"}
Rendered public/404.html (0.0ms)
Filter chain halted as :project rendered or redirected
Completed 404 Not Found in 5ms (Views: 0.7ms | ActiveRecord: 0.7ms)
Started GET "/projects/new" for 170.95.35.204 at 2013-06-13 16:20:04 +0530
Processing by ProjectsController#new as HTML
Rendered projects/_errors.html.haml (0.1ms)
Rendered projects/new.html.haml within layouts/application (4.1ms)
Rendered layouts/_head.html.haml (0.7ms)
Rendered layouts/_search.html.haml (57.5ms)
Rendered layouts/_head_panel.html.haml (60.4ms)
Rendered layouts/_flash.html.haml (0.1ms)
Rendered layouts/nav/_dashboard.html.haml (3.0ms)
Completed 200 OK in 72ms (Views: 65.8ms | ActiveRecord: 4.0ms)``
When encountering gitlab import problems, this is a general workaround.
Create a bare project in gitlab
Add it as a remote in an existing cloned [and up to date] repo. You can git fetch from your current origin to bring it up to date.
git remote add mygitlab [bare project URL from #1]
Push the existing repo state to the new remote
git push mygitlab --all
You may want to check some of the various recipes here if you have a problem getting everything to push: Push local Git repo to new remote including all branches and tags
I myself just edit .git/config by hand. Take the mygitlab URL and make it the origin url. You can just delete the mygitlab block, or via command line:
git remote remove mygitlab
This is easier than banging rocks together trying to get gitlab to import something it doesn't want to [for me, it's bitbucket right now].

404 error on heroku db:push

I am getting a 404 error when trying to push my database to Heroku via Taps
(1.9.2#[app_name]_db) heroku db:push --app [app_name]
Loaded Taps v0.3.24
Auto-detected local database: sqlite://db/development.sqlite3
Warning: Data in the app '[app-name]' will be overwritten and will not be recoverable.
! WARNING: Destructive Action
! This command will affect the app: [app-name]
! To proceed, type "[app-name]" or re-run this command with --confirm [app-name]
> [app-name]
Sending schema
Schema: 0% | | ETA: --:--:--
Saving session to push_201209251425.dat..
!!! Caught Server Exception
HTTP CODE: 404
The db:push command used to work fine, then I made some changes to my database by rolling back the migrations, editing them, and then re-migrating. Now I can deploy the app just fine, but the database will not push -- I don't know if this is related to editing the migrations or not.
The app works fine on my machine, and I wanted to eliminate any discrepancies between Heroku's copy and my own, so I created a new app and pushed to that. Same thing: the Heroku app works but will not receive db:push; it errors out with the same 404 above.
Is this a Heroku service temporarily down, or has changing my app caused the 404?
Edit: heroku logs do not show any error message
Heroku support was taking too long to respond, so I found a workaround that communicates with my EC2 instance directly by using the Taps gem.
Go to Heroku dashboard for your database. For me this was at
https://postgres.heroku.com/databases/[my-database-name]
though I navigated by going through Addons.
Click on 'URL' in 'Connection Settings', should give you something like
postgres://[username]:[password]#ec2-[ip_address_numbers].compute-1.amazonaws.com:[port]/[database_name]
Copy this value down, I'll reference it here as [EC2_URL]
Get Taps installed on 1.9.2 gemset if you don't already have it (not sure if 1.9.3 will work, didn't test it)
Set up localhost taps server to facilitate transaction by running in terminal:
taps server postgres://[local_machine_username]#localhost/[name_from_database.yml] [some_username] [some_password]
(note the spaces before username and password)
Then you can process the transaction yourself through another terminal window:
taps pull [EC2_URL] http://[some_username]:[some_password]#localhost:5000
It should run and pull all your data from the local development db to the Amazon instance. You can also do vice versa, or choose a different database, etc. Or not, I'm not a cop.
There are some problems with heroku db commands and ruby 1.9.2 (I have this version).
db:pull ends with "Unable to fetch tables information from"
db:push ends with "!!! Caught Server Exception HTTP CODE: 404"
There is a work-around for this problem. Switch to ruby 1.8.7 (I am using rvm for this) for a moment just to do db operations on heroku and after finish switch ruby back.
I do the same process (have Heroku convert my sqlite database to Postgres), and I was getting this problem yesterday as well. It seems to be working now, so I'm believe it was an issue with Heroku.