#100DaysOfWeb in Python Transcripts
Chapter: Days 89-92: Deploying Python web apps to Linux
Lecture: A final check-in on the server

Login or purchase this course to watch this video and the rest of the course contents.
0:00 Let's do one final quick check in on the server. I'll SSH over there. Have a look. So, no security updates
0:08 again it's only been a few minutes, but still good. Now we'll do a quick check in with Glances here. Our web app is up and running.
0:14 We can even sort by memory usage by pressing M or my CPU usage by pressing C there's a whole bunch of commands here.
0:20 So here you can see all of our servers running they're each taking about 100 megabytes, 120 megabytes but if you add all those up you might think
0:29 we're using way more than that a lot of the processors can share memory or read the same shared files off disc and stuff like that.
0:36 So we actually have a ton of memory free we only used 374 megs, we have 1.5, 1.6 gigs left. Now in another browser window, on a separate screen
0:46 I'll go click around and you can see the CPU here. I'll sort my CPU. I'll go click around and just refresh these.
0:54 You can see some of the microSD worker processes not the same one over and over but it's kind of being fanned out across them
1:01 are doing a little bit of work. But even as these requests come in it's only 1% CPU. They're easily handling this
1:07 talking to the SQLite back database and things like that. Serving it up, it's great. So, looks like our server are...
1:14 What do we pay, $10 for this server per month? It is absolutely handling it. I mean here we're getting two requests per second
1:20 something like that, and the overall CPU is only 8.4. And remember a lot of this, right now 1.8 1.6% of the CPU is actually just Glances.
1:32 So Glances actually adds a lot of overhead talking about five versus 3% type of thing. So our server, I think this thing is good.
1:39 We could easily run on the $5 server and have plenty of headroom for this site to run. Super responsive, lots of worker processes. We're golden.

Talk Python's Mastodon Michael Kennedy's Mastodon