Data science project for Dice job application. Here's a video overview of the app.
Usually I spin up a spot instance because it's cheaper. The m4.xlarge seems to be a decent choice right now. Once the instance is up, you should ssh into it, and the repo should be cloned into it.
Then you need an elastic IP, and have to assign it to the instance. Create the elastic IP (Amazon pool), then choose 'actions' and 'associate address'. The address needs to be associated with the server that was just spun up. Once the IP is associated, you should use that IP to ssh into the instance.
Then go to route 53, and set up the rest:
Follow this guide: http://techgenix.com/namecheap-aws-ec2-linux/
It can take a while (up to 24 hours) for the nameservers to update, and for your web address to actually work. You do not need the period after the DNS server names
Next, you need to start mongo: sudo mongod --dbpath=/var/lib/mongodb --smallfiles
This can be run in a tmux shell, or sent to the background with ctrl+z, then typing bg.
TODO: these next 4 commands should be added to the startup routine of the server, since they need to be added every time.
To be able to access the site, we have to make available port 80: Enter the commands
sudo iptables -A INPUT -i eth0 -p tcp --dport 80 -j ACCEPT
sudo iptables -A INPUT -i eth0 -p tcp --dport 10001 -j ACCEPT
sudo iptables -A PREROUTING -t nat -i eth0 -p tcp --dport 80 -j REDIRECT --to-port 10001
-- this one seemed to the the final key to making it work, after running the above 3
sudo iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-ports 10001
sudo iptables -t nat -D PREROUTING 1
sudo iptables -t nat -A OUTPUT -o lo -p tcp --dport 80 -j REDIRECT --to-port 10001
(doesn't seem to work in the /etc/rc.local file, maybe add to bashrc or something)
Next it's best to run this in a tmux session (tmux new -s server -- creates a tmux session called server), which will mean you can access the running server after logging out and the ssh-ing back in. Navigate to the home dir of the repo and do python app/app.py.
You should be able to access at your ip with port 10001, like so: 54.70.49.112:10001
Yay!
TODO: create function in the scraping file so it runs periodically and can be run from within a tmux shell.
to setup cronjob in ubuntu linux:
sudo crontab -e
and enter the following at the end:
(min hr day month weekday file)
0 10 * * * /usr/bin/python /home/ubuntu/dice_codingskills_project/dice_code/daily_scrape.py >> /home/ubuntu/dice_codingskills_project/scrape_log.log
Do which python to make sure the python path is correct.
Cron will run from the home directory (it used to be the stdout redirect is going to /scrape_log.txt)
Install postfix: sudo apt-get install postfix and choose local configuration, so that any errors from crontab are sent to /var/mail/ubuntu. Check the output with tail -f /var/mail/ubuntu
To see what python scripts are currently running, do ps aux | grep python.
You should also be able to do it like:
0 10 * * * /home/ubuntu/dice_codingskills_project/run_cron.sh
You will also have to do sudo chmod a+x run_cron.sh to make the file executable.
I couldn't get that working, though. To redirect errors, see this
Instead of 0 10 * * *, you should also be able to do
@daily /home/...
although I'm not 100% sure this works.
To check your system time do date "+%H:%M:%S %d/%m/%y". Usually it should be UTC, so 10am UTC is 4am MST.
To check the log of cron jobs, do: grep CRON /var/log/syslog.
http://stackoverflow.com/questions/11255630/how-to-export-all-collection-in-mongodb
(from within bash shell, not mongo)
Export:
mongodump -d <database_name> -o <directory_backup>
i.e.
mongodump -d dice_jobs -o dice_jobs.db
Import:
mongorestore -d <database_name> <directory_backup>
i.e.
mongorestore -d dice_jobs dice_jobs.db/dice_jobs
Then from local machine, you can get the data with scp:
scp -r -i ~/.ssh/ubuntu16lap.pem [email protected]:/home/ubuntu/dice_jobs_jan_2018.bkup /media/nate/data_lake/dice_jobs_backups
Todo: automate backup of db on S3 after scraping is finished (in daily_scrape.py)
You will have to redirect the port to port 80 (standard for browsers): sudo iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-ports 10001 (https://gist.github.com/kentbrew/776580), may also need to do sudo iptables -t nat -D PREROUTING 1
sudo iptables -t nat -A OUTPUT -o lo -p tcp --dport 80 -j REDIRECT --to-port 10001
I think put those things in here:?
sudo nano /etc/rc.local
Run the install_script.sh script
Some problems with MongoDB, to get it started upon startup I had to do sudo mongod --dbpath=/var/lib/mongodb --smallfiles
http://stackoverflow.com/questions/14584393/why-getting-error-mongod-dead-but-subsys-locked-and-insufficient-free-space-for
http://stackoverflow.com/questions/24599119/mongodb-not-working-error-dbpath-data-db-does-not-exist
http://askubuntu.com/questions/61503/how-to-start-mongodb-server-on-system-start
Upon restart, you may need to run:
sudo iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-ports 10001
sudo mongod --dbpath=/var/lib/mongodb --smallfiles (this takes a long time)
If the scraping isn't finishing, likely memory is running out. I was trying to run this on a 8GB memory machine at first, and the scraping crashed for a while without me noticing. I was able to find it was crashing from checking:
nano /var/log/kern.log
which I found out about from here: https://unix.stackexchange.com/questions/27461/how-can-i-know-when-a-cron-job-was-killed-or-it-crashed
Use the requirements.txt to install dependencies.
pip install -r requirements.txt
I had to install matplotlib with the package manager on Ubuntu 16 to get it working properly:
sudo apt-get install python-matplotlib
When fixing the scraping code, I had some issues and had to delete the newest scrapes. This was done with some mongo commands:
mongo
show dbs
use dice_jobs
show collections
# find newest entries and look at location column
db['data science'].find({}, {location: 1}).sort({scraped_time: -1}).limit(10);
# count number of entries and number of newest entries
db['data science'].count();
db['data science'].find({scraped_time: { $gte: ISODate('2018-12-05')}}).count()
db['data science'].remove({scraped_time: { $gte: ISODate('2018-12-05')}});