Why would it make sense to have a "Download from Github" button instead of a "View me on GitHub" button or something?
It of cause makes sense for closed source software, but being open source it invites in people who at the very least want to read the `README.md`, star the repo and most likely want to have a look at the source code.
I kind of panicked when the download button actually downloaded - I was like "hey stop! wait a sec! I want to check out the code first" - maybe it's just me ;)
This one isn't so easy. What should the footer github link to? Assuming they might want to keep the same footer across the whole web site, https://github.com/cachethq/Cachet does not seem right either.
It does look good, but how well does it hold up under load? When the main service goes down the status page will get a huge load spike, and it's probably not hosted on your best hardware.
The demo page doesn't look cachable: a "Cache-Control: no-cache" header, no ETag or Last-Modified headers. It's obvious that they don't want stale pages in a cache, but why not take advantage of a simple Last-Modified to remove most traffic from people refreshing every few seconds?
I know first hand that Cachet has been under heavy development of late and James was having some issues with caching giving false positives so I wouldn't be surprised if he just needs to revisit caching in general now thy the code base is stable. I've been running it in docker containers and I can tell you it's very quick.
It's great to see Stoke-on-Trent represented in the startup arena.
As I'm really keen to see more local jobs and affluence, my comment would be, there's nothing about how the project will be sustained financially, no payment plans or premium support.
Look forward to seeing more about this at Staffs Web Meetup.
i have been testing this tool for a few weeks now and we are deploying it in production now. the biggest problem that remains for us is the lack of integration with Twitter (on which we still need to do manual updates) and lack of caching which makes the application not as available as it could.
that said it's pretty much the only free alternative out there, so it's a welcome and refreshing change. props up to the devs too that have been very fast and responsive on github, fixing issues, bugs and suggestions on the same day sometimes. a rare breed.
Here is a list of things that Cachet is not or does not do:
1. It does not monitor your services. It works only as a way to display the status of your services. However, Cachet is able to receive updates from third-party services via its API.
2. It does not work on a plugin system. There are no monitoring services to extend.
So basically we need both Cachet for displaying the status and another tool to check it (it wouldn't be very good if the tool was on the app itself otherwise it might not be effective when it's truly needed).
I was hoping to have both in one place (the monitoring and the status), is there a "cachet-ready" system?
I'm curious about how large companies manage their status pages. Are they running dedicated servers, internet access lines and DNS etc... for only indicating service status? What happens if status pages go down. Is there a status page of status page?
They usually just host it on a third-party host. It's extremely unlikely that both their systems and the third party will be down at the same time. If only the status page is down but the services are working fine... well, why would you be looking at the status page then?
Sure about that? My experience is that large companies host their dashboard in their own environment and that the dashboard will be down when their own infrastructure is. If by large companies, we are talking about >10 000 employees or so.
Well, it depends on the company. The companies I've worked with do it the way I've described, but I can definitely see enterprises going all politics and deciding to host the status page themselves, rather than do it correctly.
We also use StatusPage.io at my company, but it's hard not to be offended at how expensive it is when you start adding certain features. I've been tempted to make a spin-off for some time now.
You mention in that article that $29 is too high. But would you really need a status page for a service unless it was generating enough revenue to justify $29? A status page seems like a "oh now I have a critical mass of customers and need a way of sharing uptime" vs being something you need pre revenue or early revenue.
I believe so. At my workplace we use Cachet for our internal tools and applications - we wouldn't want to pay $29 a month for something internal like that.
There are a lot of people using Cachet in the same way as we do and even for services or business that are making money.
Does StatusPage do anything that this software doesn't? Just curious, because I was thinking about setting up some form of simple monitoring, but I haven't as of yet investigated any of the solutions in this space. (Forgive me for not checking their homepage; I'd honestly prefer some anecdotal opinion.)
If the two are on par feature-wise, I suppose it comes down to the cost savings of maintaining your own versus the time savings of outsourcing the problem to someone else. It might be nice just to forget about monitoring maintenance.
Considering statuspage.io is set up in redundant data centers and has all the bells and whistles that you need, building and maintaining your own seems less and less desirable to me.
Looks really nice. Curious about the license terms and the change from MIT to "Easier licensing terms"[0] ... What does the change mean in practice and why was it changed from MIT?
One remark to the demo. When you click a circle, a popup shows but its position is outside browser window in my android's chrome. Otherwise, pretty neat!
Hopefully it's fairly straightforward. If not, then I do apologise. Documentation has been focused on the API for now because that's what most of our current user base is looking for.
It of cause makes sense for closed source software, but being open source it invites in people who at the very least want to read the `README.md`, star the repo and most likely want to have a look at the source code.
I kind of panicked when the download button actually downloaded - I was like "hey stop! wait a sec! I want to check out the code first" - maybe it's just me ;)