… but they have a point in that the size of the file is most likely not the issue here. I can only imagine that a “full queue” wouldn’t even start processing long enough to time out on a too-large file.
That said, I have not seen this queue error.
3 Likes
That’s interesting info.
And to reply to your question- I have no idea what they can do to fix it. I have only heard talk about making it non-cloud based. Just not having to rely on server, from my perspective, would be ideal. Apparently, according to FB admin, I’m only “griping” and not asking for a solution or even an idea of how long this is expected to be an issue. I can’t deal with the social media admin power trip junk, so I guess I will start searching the forum and seeing what other people are experiencing. I’ve just noticed lots of other people dealing with this this week and I’m sure it’s due to new updates. So that gives me an inkling of hope that it’ll get better in the coming weeks. Hopefully.
Yeah, it’s not gonna be non-cloud based and (I’m guessing) part of why the admin was saying you were just griping is that there isn’t a solution currently available. So saying you want to leverage the “thousands” of people who are having issues, yet what do you want to do with this leverage?
Have you submitted a ticket to the staff? Folks on here who were hitting that seemed to only be getting it off and on, and there was a Google outage during that time. If you’ve been getting it for a week solid that seems like an issue you might want to submit to the staff (which technically you’ve done here by posting in P&S, though your original post doesn’t actually ask a question). You may be having a different issue that’s individual to your machine, part of the joy of the cloud based is it’s easy for them to check the logs on your machine and see exactly what it thought was happening, they might be able to address your individual problem.
Worth a shot eh?
4 Likes
The whole point of cloud-based is that you can arbitrarily scale your resources up or down at any time. The systems used by Google, AWS, Microsoft and many others are all designed to behave this way. It’s a primary selling point - buy the processing and storage you need now and when you need more you just have to pay for it and it’s yours.
These cloud based systems are already geographically distributed, both for fault tolerance and performance. You don’t want all the cloud requests generated on the East Coast to have to go all the way to Washington or California to be serviced. The way the workloads are distributed is invisible to the end user, so you have no easy way of determining where the work is actually getting done. But trust me, these cloud providers are laser focused on performance and you just can’t perform well in this use case with a single centralized processing and storage resource.
So if GF’s systems are indeed being overwhelmed leading to Queue Full errors, it’s trivially easy to fix.
1 Like
Absolutely worth a shot. I think FB is likely providing a smaller microcosm of people experiencing the issue. I appreciate your help! And I had no idea of a google outage- I also didn’t know that is where the server is based, so that’s also helpful. Thanks a lot!
2 Likes
your “trivially” easy fix have taken Google (not Glowforge) engineers 8 days so far and they’re still not at 100%…
3 Likes
Not necessarily.
Just because the cloud provider has resources, doesn’t mean every customer has unlimited access.
It could very well be the case that GF customers have used up all allocated (and subscribed) resources, and that is what drives the error.
That’s what I think is happening. The processing of our jobs is not overly complex, similar tasks for 3D printers and CNC machines (for example) take just a few seconds on a reasonably modern computer.
Moving away from cloud to local computer isn’t about complexity, it’s about having to support software on a gazillion different operating system versions and applications.
4 Likes
She has, since she posted here in Problems and Support.
1 Like
I meant prior to her call to arms 
2 Likes
How? You just said Google’s servers (which Glowforge uses) are distributed all over and they’re laser focused on performance. What’s trivially easy to fix about that?
1 Like
Per Glowforge it is supposed to be local at some point
I got this error this afternoon, too. Just for a short while. I was actually running something and when it finished, I was loading another file in and it gave me that ‘queue’ thing.
1 Like
I’ve got that message a couple of times this week, but it didn’t last long. I figure it has something to do with the latest updates being sent out en masse… but I could be wrong.
Ref.?
1 Like
Bingo
7 Likes
I’d love to see where you saw this cuz the only thing I’ve ever seen Dan say is that if 
ever went away as a company they’d make sure the firmware was available for tinkering (which they did in a GitHub drop a couple years back)
One example:
1 Like
Cloud services are designed to be “on demand” and scalable. You need more, you simply buy more. It is literally as easy as buying anything else on-line. It’s the whole point of Cloud Services - you pay for what you need and get more by paying more as your needs increase. Google and Amazon and Microsoft and everyone else aren’t going to “run out”. They have tons of “resiliency” since their business model is fundamentally based on not running out of capacity.
Like Cable TV. You want more channels, you just go on line and order them and your cable box suddenly has more channels. You’re running an app on Google and you need more processing capacity you just order it and it’s there.
So if “Queue Full” is a resource bandwidth problem, fixing it is as easy as buying more bandwidth. Trivial. If however “Queue Full” is a problem with Google’s service, that has nothing to do with whether or not GF has enough bandwidth. It’s to do with busted hardware or a software bug that needs to be fixed. Nothing GF can do about that except complain just like the rest of us.
4 Likes
Which is exactly what we’ve been saying for 3 hours…
2 Likes
I am so sorry for the periods this month where the service was slow or unavailable. You can see them all here: https://statuspage.glowforge.com.
We’re in the process of deploying improvements now to address this problem and improve our uptime in the future.
I’m going to close this thread - if you run into any trouble, go ahead and post a new topic.
5 Likes