I have set up a Salt Master and have got everything to work well but there has been one thing I have not been able to figure out and trying to search for the answer is proving to be difficult so I thought I would see if anyone else has come across this. I have a profile setup in digital ocean that is setup up like the following:
digitalocean-ubuntu:
provider: my-digitalocean-config
image: Ubuntu 14.04 x32
size: 512MB
location: New York 1
backups_enabled: True
When I run the profile it creates the proper instance with all the options correct but the backups_enabled doesn't seem to work. I have tried going on both digital ocean's website as well as looking through the salt documents but I can't figure out if the command changed or isn't supported anymore? If someone knows the answer I would be very grateful to learn what I'm doing wrong.
Salt has two versions of their digital ocean API. One for V1 and one for V2. I wasn't using the newer version so there were issues trying to enable new features.
Related
I wrote a Shiny app, and now I need to turn it into a Stand-Alone Program. The reasoning behind this is that I need to share the app but can't do this with shinyapps.io or a server as I need the app to be able to access user's folders.
So far, I found these 2 tutorials: deploying-desktop-apps and packaging-your-shiny-app. Both of them (supposedly) work on Windows, but I have a Mac, and I want to app to be available for users of all systems, or at least Mac and Linux. Any thoughts and suggestions would be appreciated!
I actually tried to follow the tutorial mentioned above, and can't even install R-portable for my Mac. So I'm looking for something different.
Running a Virtual Machine to follow Windows tutorial is an option, but in this case, the app will be Windows-specific, and I don't want this.
This thread is really old I know, but I'm also trying to find answers on creating a standalone version of R for Mac.
This would support for
https://github.com/chasemc/electricShine
which supports Windows
I'm looking into using the new Realm Mobile Platform for a project of mine. I've gone through the guides and was able to get it up and running locally no problem. My question is, what's the best way to deploy the Realm Object Server so it can be run remotely? I read through the guide found here but didn't really understand it. I only have minimal experience deploying a rails app to heroku. How can I get it deployed to Heroku or a similar service? Any help is appreciated. Thanks!
It's hard to tell you what the "best" way is. There are always drawbacks and benefits to any setup, and everyone has different goals and objectives, so I don't think there is an objective "best way to run it," as you say.
The Realm Object Server doesn't support Heroku for the time being (or at least, no easy one-click-install integration). We know that this is something that people want, so it's on our radar, but I can't give you a definite answer as to when or even if we will do this one day.
The way most people run the Object Server is by running a virtual machine, and running the service inside of that. There are multiple ways to achieve this: start a virtual machine with your favourite cloud provider, and then install the Realm Object Server on top of that. Alternatively, Realm also provides an AMI image, which is Amazon lingo for "a pre-configured virtual machine image," that contains the Object Server pre-installed, and allows you to run your Object Server at the click of a button.
Please bear in mind that Realm Object Server is currently packaged for RHEL/CentOS 6 & 7, and Ubuntu 16.04.
Here are some links that should help you get started:
A basic tutorial on how to setup Ubuntu 16.04 on Digital Ocean
AWS' documentation on launching an EC2 instance from an AMI
Try this image to run realm-objserct server on openshift online.
https://hub.docker.com/r/viksgyl/realm-object-server/
I've researched this feature extensively and need help. PAM is a difficult authentication program for me to thoroughly understand although I'm learning.
We have the option of running either ProFTPD or pure-ftpd on a VPS hosted by hostgator. The OS is CentOS release 5.11. I have complete access to the server. Our goal is to send an email upon successful FTP login by a user. I've successfully used a script which is run by PAM for notification emails dispatched upon successful SSH logins and need to modify it for successful FTP logins. That script which is amazing works fine with ssh (but we need ftp as I've mentioned) and can be viewed here:
http://blog.stalkr.net/2010/11/login-notifications-pamexec-scripting.html
Before I even try to wrap my head around why this script does not work using FTP, despite PAM working on our VPS (for at least ssh), can someone chime in with some theory as to why this is not working at all. I.E., maybe I'm way off base with this idea and it can't work with FTP and PAM in which case I'll just drop this approach. However, when I read configuration settings, etc., on using PAM with these two FTP programs, they seem to indicate this should work.
Additionally, running the following code which is supposed to test for PAM-enabled applications does not return FTP.
ldd /{,usr/}{bin,sbin}/* | grep -B 5 libpam | grep '^/'
Any suggestions or help from gurus would be greatly appreciated.
Found solutions. The answer was to include session_support=YES in the /etc/pam.d/vsftd configuration file. So that's the answer to anyone who may have a similar issue.
I need to run R with Node.js, using Rio (https://github.com/albertosantini/node-rio) as the node binding to Rserve.
I like Heroku but this seems like it is pushing the Heroku envelope beyond what it or I am competent with:
I've looked briefly into installing a custom buildpack
https://github.com/virtualstaticvoid/heroku-buildpack-r
to run simultaneously with node.js:
https://github.com/ddollar/heroku-buildpack-multi
This all seems pretty scary. Anyone got any good advice for how best to host this? My app works just fine locally.
http://prgmr.com/xen/
I currently use this solution to run my Node.js server and it's currently great.
They have wonderful support and they're uptime is 100%. I cannot recommend this any higher, but you will need to know how to set up a simple OS and run it from the ground up.
For example, if you want to run a server without having it stop when you close the SSH connection, you would use screen node script.js and press [control] + [A] + [D] keys.
You might already know this, so simply take my advice and view the website.
After some research and recommendations from Heroku, I believe the Heroku solution would be
Use https://github.com/virtualstaticvoid/heroku-buildpack-r
in combination with
https://github.com/ddollar/heroku-buildpack-multi#readme
to build a multi build pack.
I have recently installed eucalyptus on a Ubuntu 10.04. Everything works fine until I get to the user interface. Whenever I try to get to the store tab I get the following error:
Error 60: server certificate verification failed. CAfile: /etc/ssl/certs/ca-certificates.crt CRLfile: none
I have been doing some research and tried to update the certificate, make changes to the /usr/lib/python2.7/dist-packages/imagestore/lib/fetch.py but none of these solutions worked.
I am really stuck and do not know what to do.
I cannot help resolve the issue on Ubuntu and I suspect the Euca version involved is way out date. If you are willing to work on CentOS then you've got two great options
FastStart - this is an awesome bundle of Eucalyptus and CentOS. I use this to spin up a new cloud after each release, including the pre-release of 3.3. Total time to new cloud is roughly 35 minutes.
http://www.eucalyptus.com/download/faststart
Package downloads - This is your standard install method and the only option to pick if you're trying to build a large cloud.
http://www.eucalyptus.com/download/eucalyptus
BTW - if you are looking to build a large cloud I also recommend you check out the Reference Architectures. These are Eucalyptus deployment blueprints based on lots of real-world, customer environments. http://eucalyptus.com/eucalyptus-cloud/reference-architectures
If you have other questions, I hope we see you on IRC. It's the #eucalyptus channel on freenode: http://webchat.freenode.net/?channels=eucalyptus
Hope this helps