It's amazing how quickly these cloud-dependent systems can be taken down and effectively relegate people back to the dark ages. Quite frightening, actually.
The YouTube TV Thread
- Thread starter IcEWoLF
- Start date
- Latest activity Latest activity:
- Replies 1K
- Views 154K
-
WELCOME TO THE NEW SERVER!
If you are seeing this you are on our new server WELCOME HOME!
While the new server is online Scott is still working on the backend including the cachine. But the site is usable while the work is being completes!
Thank you for your patience and again WELCOME HOME!
CLICK THE X IN THE TOP RIGHT CORNER OF THE BOX TO DISMISS THIS MESSAGE
It's easy to forget all the growing pains cable TV and satellite TV had in their first few years. When I was growing up in the early 70s, it seemed the cable TV service would go out more frequently than the electricity.
Yes but nobody depended on cable TV beyond entertainment. Google's platform includes services that could be essential to schools, hospitals, law enforcement etc. It's an infrastructure, not just a media platform.
If Google publishes the full post-mortem of the event, it's sure to read much like the failure that took out AWS S3 in late 2017: Amazon Corrects Massive AWS S3 Cloud Outage - Vendors React : @VMblog
Automation and autoscaling are great, but there is always the looming risk of cascading failure if the wrong underpinning infrastructure is under-sized. Whether it's being operated by humans, or running programs written by humans, semi-automated systems are not immune from system logic faults.
The infrastructure is already divided into regions and availability zones. Impact was mostly limited to US-East with some marginal degradation to US-Central. The megacorp I work for makes extensive use of GCP resources -- we shifted everything to US-West and stayed operational throughout the entire event.
I'm advocating that they need to partition their capacity such that they can guarantee that critical traffic gets through. Breaking things in to geographical regions or population centers doesn't answer that need.
Few who depend on cloud services have the luxury of a fallback.
It was a cascading network failure that took out everything in US-East, including Google internal systems that their operations teams use. The closest analogy would be the 2003 Niagra-Mohawk Northeast power blackout. I'm sure they would have liked to kept power going to hospitals, nursing homes, and other critical infrastructure -- but huge systemic failures don't work like that.
Just like we couldn't have the stable power infrastructure we have today if we didn't build massive grids, you can't scale compute infrastructure that can meet demand without building massive pools of dynamically allocatable resources. Bryan Cantrill (Joyent CTO) gave a talk about these types of failures a couple years ago -- these systems have tons of heavy automation, and are subject to operator or system failures that can cause things to go deeply sideways. This is worth a watch if you want to know a little more about these kinds of outage scenarios:
What made it particularly bad for most Google services, they continued to take traffic at local region ingress because their re-routing system were taken offline. In our case, we maintain our DNS control outside of GCP, so we were available nationwide even though we restricted our compute footprint to only operating in the US-West1 region.
Just like we couldn't have the stable power infrastructure we have today if we didn't build massive grids, you can't scale compute infrastructure that can meet demand without building massive pools of dynamically allocatable resources. Bryan Cantrill (Joyent CTO) gave a talk about these types of failures a couple years ago -- these systems have tons of heavy automation, and are subject to operator or system failures that can cause things to go deeply sideways. This is worth a watch if you want to know a little more about these kinds of outage scenarios:
What made it particularly bad for most Google services, they continued to take traffic at local region ingress because their re-routing system were taken offline. In our case, we maintain our DNS control outside of GCP, so we were available nationwide even though we restricted our compute footprint to only operating in the US-West1 region.
My former company's AWS-based transcoding service keeps having capacity issues. They're actually looking at switching to Azure for that functionality as they seem to have the most reliable offering at the moment. The key is building for portability and multi-cloud, which, at the end of the day, probably means it will be more expensive than just hosting things yourself in a co-lo or on-premises.
We have the Hallmark Movies Now! app,there are no commercials,we like it a lot.
Hulu Live TV does have more channels we could add in their add on section on their website.
Google's having a bad week so far: Major Google Fi outage prevented people from making or receiving calls
Well that did not take long,Spectrum had an issue within their bandwidth that affected both my Hulu Live and Vudu.
And Hulu Live has audio issues,plus their guide is not as good and complete as YTTVs,so it's back to YTTV for me.
I've noticed over in the YTTV Facebook group lately, some of the folks who cancelled YTTV in protest over the 43% price increase for legacy members last month are slowly finding their way back after discovering the grass isn't any greener with the other services.
They had to understand the price point was not sustainable. Sports and locals alone probably cost more to deliver than the legacy price. I had legacy pricing, but wasn't surprised to see the major jump. We got the benefit of a cheap package for a good amount of time. Now, let's hope the price increases don't come as rapidly as they did with cable/sat.
Sounds like there are no winback offers for those who want to play the cancel game. Maybe that will keep the prices stable for all users.
Sounds like there are no winback offers for those who want to play the cancel game. Maybe that will keep the prices stable for all users.
I just learned the hard way that if you want your Google Pay balance to go toward your YTTV subscription, you have to have at least $50 in your account and manually set your YTTV subscription to be charged to your GP balance (set to "primary"). I recently loaded a $50 GP gift card I had gotten for $45, then made an unexpected mobile purchase for $10. When it came time to bill my YTTV subscription the other day, it charged $50 to my credit card. This is unlike how the Sony Wallet works (for all its faults) where any account balance is used before your credit card is charged for the remaining fee.
Well Hulu Live does have History(We like Ancient Aliens),but as I said their guide is an issue for me and their audio issues.
I thought that Disney might make it better since they took over to buy it.
Sure there is with OTT providers there are no"got ya" fees.
Subscriptions prices can go up,but it's still cheaper with OTT providers.
The guide issue may be fixed very soon, as there is a update on the way that changes it up somewhat. I do agree that is one issue with the service, but even as it is now, it is WORLDS better than it was before they added the guide they have.
This is what it will look like after the update. Still similar, but with more information, two weeks of program info, and the tabs moved to the side, with what hopefully is a customization "my channels" option.
- Nov 29, 2003
- 14,302
- 18,366
I have posted here before the advantage of using gift cards to pay for OTT services and use them to get a discount on those services, every year between Black Friday and x-mas gift cards go on sale between 20% off ( itunes, google play, sony ) to 40% off (netflix, hulu, hbo, cbs, et), I stock up every year and plan to do so this year.
itunes for example this week are 15% off at Costco so I bought a couple of $100 packs to cover buying 4K movies, I also wait to buy till 2 weeks after disc street date, then itunes makes new 4K movies $9.99 which turns into $8.50 for me because of the cards.
