Given how long it is taking for the status reports to be updated on the server status I suspect that during off-hours no-one at Glowforge is being warned that the servers are down or slow.
Sounds like they need some monitors which can test the servers response time and sound a loud bell somewhere.
If everyone having an issue opens a ticket, that will get their attention at any time.
The service is working. No issue here. So it’s not going to appear to be broken from their point of view.
So far, only a handful of people have opened tickets here over the weekend. Out of 30,000 owners, that’s not significant. People griping on failbook will not get any attention.
Network monitoring is an automated process; it doesn’t require human intervention to report outages. When something does go down, the system administrators get immediate notifications, but since the status page is reporting no issues for today, that won’t have happened. Also, it’s the weekend, and although they do sometimes pop in outside of business hours, it’s not an expectation. Best to eliminate any possible issues on your end while you’re waiting, rather than just assuming it’s a server-side issue.
I hope that is true, that the process is automated - but it seems to me there is a long lag between issues arising the status page being updated.
I’m speculating wildly here, but I think they might have multiple cloud servers and not all of them work correctly. But, could be DNS issues, could be internet issues or indeed could be local to my machines and/or browser.
All I know is that if any network issues arise on any of the websites I administer, I get an immediate email and text notification. And I’m just a hobbyist; I’m sure multimillion dollar companies don’t rely on human monitoring to update their status pages.
There was a software issue. That’s not the same as a server or network issue. You’re not reading all my words. Software issues may or may not be automatically detectible, so the reporting may lag. Network and Server issues, however, generally are, and reporting is almost instant.
The are called compute engines, they come in different flavors for different types of workloads (such as compute vs. general purpose), and they do not exist as independant physical servers. They are spun up on-demand in their globally-distributed virtual environment.
THey could have an elastic cloud server that is seen as one single server - all users get affected at the time. they could have multiple servers - with multiple engines - in which case some users could get affected but not all.
And there are about a dozen different ways all this could be configured.
I think we’re all speculating wildly again, so suggest we knock this thread on the head.
I specifically asked you if the machine need to be turned on or off and you said it makes no difference :
This says the update was not applied because the unit was powered off.
This clearly states why both my machines were not working : First it was the cloud service being down and today it was the updates.
UGH … having two machines gives me a clue as to what is happening and why but you need understand that both updates and the sever going down results in the machine not working correctly until its finished. It can only finish when the Unit is Turned On.
To further clarify: The software is browser-based. It runs in your browser. The bug that caused the “stuck” problems the other day was in the software.
The firmware updates for the machine are applied when you cycle the power. For those you need to turn your machine off and on again.