Missing config info for Windows

[maicanvlad7]

Hello!

I managed to run the makefile and it created a bin folder containing a vroom file, no extension, no nothing.

I've tried:

1. configuring the path variable and adding the path to the bin folder in windows
2. adding the path to the vroom file directly in the config.yml file

With both methods mentioned above, after I run npm-start on vroom express I still get 'vroom' is not recognized as an internal or external command, operable program or batch file.

What would be the solution to this ? I am currently running windows server but I don't have trouble configuring another system variables except this one.

[nilsnolde]

That’d be more of vroom build issue probably, not so much the HTTP API.. See here for a solution with cmake: https://github.com/VROOM-Project/vroom/issues/454#issuecomment-774672056. it’s been a few months, but should still work out of the box. Another alternative is the docker repo: https://github.com/VROOM-Project/vroom-docker. If you have the right Windows Server version with WSL2.

Feel free to open another issue in either the main vroom repo or our fork.

[jcoupey]

That’d be more of vroom build issue probably

Not sure as building from the makefile looks successful. Sounds more to me like a problem with having a local executable being recognized system-wide? In that case it would still be outside the scope of this repo.

@maicanvlad7 are you able to run vroom -h using the executable you built?

[maicanvlad7]

@jcoupey It doesn't build an executable, the files has no extension as if it has been built on ubuntu It is not .exe or .bat

[jcoupey]

Not sure if you need an extension for the file to be runable, we're already way beyond my windows knowledge...

[nilsnolde]

Since you built a Makefile on Win, I assume it's nmake? No experience with that really, but I if you put the FULL path in the config.yml (or alternatively its directory in the env var $PATH) and it doesn't recognize the command, I'd try to rename it vroom.exe (not kidding..) and if that doesn't work, try CMake with Visual Studio, that is guaranteed to work.

[maicanvlad7]

I finally managed to compile it on a virtual machine xD Long road but worth it

Why do I have to specify a port in config.yml tho' ?

The demo server of OSRM looks something like this: 'http://router.project-osrm.org/route/v1/driving', so no port

My Demo server is hosted on a server with static IP address but I don;t use a port when I send a curl request to it

What is a workaround to this? How can I not use a port?

[jcoupey]

The port value is mostly useful when running your own routing stack. For example, a local OSRM instance listens to port 5000 on localhost.

[maicanvlad7]

Ok, understandable, but what do I do when I have OSRM running on a VPS with static IP? How can I connect to it ? If i delete the port from config I get an error

[nilsnolde]

You'd typically get a domain name you registered, deploy a server app like Apache/IIS and in the server app you forward proxy the domain to vroom-express server. There's a ton of tutorials out there for each and every server environment. If you simply query a static IP it'll likely default to querying port 80 on your server.

[maicanvlad7]

Wait. There actually is a tutorial on how to set up vroom on IIS ? I am paying for that link! No joke. Thanks in advance, you've made a kid happy if I manage to make VROOM work on my current setup

[nilsnolde]

Sorry mate, there’s no tutorial for vroom on IIS, but setting up IIS for some node app.

[maicanvlad7]

@nilsnolde @jcoupey Do you have any idea why when I call vroom with 7 cars starting from the same spot and 400 addresses it just assigns all the addresses to one car ?

[maicanvlad7]

When I call the vroom-docker on my server this happens, if I input a JSON file with 100 addresses and 7 cars on vroom demo server everything works ok. What am I missing? Thanks in advance!

[jcoupey]

@maicanvlad7 check https://github.com/VROOM-Project/vroom/issues/434