Kubedex.com is over 1 month old! Well, barely, 1 month and 2 days to be precise.
People have been leaving updates on my Ingress and Service Mesh comparison spreadsheets and I’ve been merging them in and updating the blogs iteratively. Although, none of the conclusions have changed so if you’ve read them I wouldn’t bother to read them again.
Since the last update I got a bit sidetracked with #DevSecOps stuff and in particular testing container vulnerability scanners. You’ve probably already seen these posts as I’ve spammed Reddit a lot, but in case you haven’t, here they are.
For those who don’t have time to read blogs the summary is that Alpine based containers shouldn’t be scanned with open source security tools as the results are mostly guesses.
I’ve decided to investigate alternative base images and will be looking at Debian, Ubuntu and Redhat minimalised image in a future comparison blog.
The other topic that I got obsessed with for a whole ten minutes was Kubernetes auto-scalers. This is a short blog about options for scaling both cluster nodes and pods on clusters. It has some interesting inputs used for the scaling of both.
Unfortunately, I failed to properly compare serverless options last week. This is a hobby website with zero income and I’m currently bored of FaaS and serverless.
Once I realised that the true power of Lambda was in tying together AWS services that have defined interfaces, and that to do similar on Kubernetes would be a massive headache, I lost interest.
Perhaps I’ll change my mind or get really bored and finish off the spreadsheet and blog in future.
I’m quite interested in comparing Docker Linux base images, clouds that provide Kubernetes and applications that help with the developer experience. Perhaps I’ll do some development work on my Prometheus Helm Chart exporter and try to do some remote debugging and simplified deploys.
Since adding all all 264 charts from the official Helm chart repository I’ve been mostly adding non-official ones.
I was interested to see what progress had been made in the last 30 days so wrote a script to compare Kubedex.com with the Github repo.
[u'dex', u'datadog', u'prisma', u'hl-composer', u'gce-ingress', u'kube-hunter', u'metricbeat', u'unifi', u'karma', u'mattermost-team-edition', u'mysql', u'prometheus-couchdb-exporter']
No incubating charts have been added in the last month. 11 stable charts have been added.
You can see all of the latest Helm charts by clicking this link.
This site now links to 309 different charts. Each of them manually categorised so you can search and browse to compare applications within a specific category.
Quite a few people have submitted new charts so if you know of any that aren’t on the site please submit via the button on the site or leave me a message via the contact form.
The other interesting observation is the clear magnitude of Helm charts in the monitoring category. I’ll need to split this out into time series, logging and event data and tracing.
There are also a lot of business category applications under misc for things like CRM and HR and other line of business apps. I wasn’t really expecting these to be run on Kubernetes. Maybe they are for development and testing.
I’m guessing nobody really cares about this but I’ll add it anyway for posterity. From scratch, and over the period of one month, this site has gone from zero users to over 10,000.
This is quite difficult to achieve as Google doesn’t rank new sites for search results. Almost all of the traffic has come from Reddit and Hacker News. Plus some people have been kind enough to like my posts on LinkedIn and Twitter. Thank you to whoever clicks my links in these places.
As ever, if anyone has an interesting topic they would like me to look into please do get in touch.
Tell us about a new Kubernetes application
Never miss a thing! Sign up for our newsletter to stay updated.
Discover and learn about everything Kubernetes