Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Tuesday March 24 2020, @11:23PM   Printer-friendly
from the you-do-not-own-your-cloud-computer dept.

'Azure appears to be full': UK punters complain of capacity issues on Microsoft's cloud:

Customers of Microsoft's Azure cloud are reporting capacity issues such as the inability to create resources and associated reliability issues.

Outage-tracking website Down Detector shows quite a few reports about UK Azure issues today, yet the official Azure Status page is all green ticks. The inability to provision resources does not count as an outage as such – though it is more than an annoyance since it is not always feasible to create the resource in an alternative Azure region. Some types of resource have to be same region in order to work correctly without a lot of reconfiguration.

Windows Virtual Desktop (WVD), a handy solution for remote workers, is one example. One user complained on Twitter that "Azure seems to be full" when trying to allocate a VM for WVD, though it appears to be a test deployment (if the name WVD-TEST-0 is anything to go by). The error reads "Allocation failed. We do not have sufficient capacity for the requested VM size in this region." The region is UK South.

[...] Note that Azure is a huge service and it would be wrong to give disproportionate weight to a small number of reports. Most of Azure seems to be working fine. That said, capacity in the UK regions was showing signs of stress even before the current crisis, so it is not surprising that issues are occurring now.


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 1, Interesting) by Anonymous Coward on Wednesday March 25 2020, @01:28PM (2 children)

    by Anonymous Coward on Wednesday March 25 2020, @01:28PM (#975444)

    3 reasons to rent server infrastructure from Amazon, Azure, Google, DigitalOcean, Linode, Scaleway, Vultr, etc...:

    1. Your infrastructure needs rapidly scale up and down. This is the whole reason Amazon started renting out servers, because they had enough infrastructure to handle the holiday rush but those machines spent most of the rest of the year idle. If you're hosting your own infrastructure and 90% of your machines are idle 90% of the time, it's far cheaper to rent elsewhere. ( So actually, I am shocked that Netflix has not moved *off* Amazon. Netflix's saving from, say, shutting off two thirds of their US infrastructure in the middle of the night is probably far less than they would save by running their own data centers. )

    2. Your infrastructure needs are relatively small. Unless your company is a tiny startup, the cost savings between spending $5k per month on your own infrastructure and $10k per month rented are probably not worth pursuing. Being able to do all of your machine management with APIs and web control panels and never needing to send someone to a colocation center will probably save you more in time and salary costs than you save with cheaper costs.

    3. It's a one-stop shop. Servers, storage, DNS, networking, firewalls, email - all in one place.

    Starting Score:    0  points
    Moderation   +1  
       Interesting=1, Total=1
    Extra 'Interesting' Modifier   0  

    Total Score:   1  
  • (Score: 2) by Runaway1956 on Wednesday March 25 2020, @01:53PM (1 child)

    by Runaway1956 (2926) Subscriber Badge on Wednesday March 25 2020, @01:53PM (#975454) Journal

    That's all well and good - there are likely savings to be had, if you shop around. But, there are more factors to the equation that need to be mentioned.

    1. Do I really really trust $cloud provider with my data? (Obviously, Uncle Sam trusts the winning bidder.)

    2. Is my ISP reliable? That is, can I count on 95% uptime, or 99%, or do I have that proverbial five nines?

    3. Do I trust all of my data to the internet? Sure, there's encryption, there are firewalls, there are blacklists, yada yada yada - but is all of it really secure, or are there a dozen zerodays that will hit me next month?

    Maybe I'm just cynical, but the cloud does introduce points of failure that don't exist if I keep my data inhouse.

    • (Score: 0) by Anonymous Coward on Wednesday March 25 2020, @02:12PM

      by Anonymous Coward on Wednesday March 25 2020, @02:12PM (#975459)

      Plenty of companies hosting their own infrastructure lost all of their data to bad backups, hackers, or phishing. And even if you host your own you still have to set up redundancy (two or more geographically distant data centers) and ISP redundancy, and maintain your own server and code security updates. None of those concerns magically go away, plus you have to handle rotating out and replacing obsolete hardware yourself.

      Granted, I would never use Azure. I think Microsoft's history is the biggest proof anyone could ever need that in a capitalist system sales and marketing are many orders of magnitude more important than quality and ethics. And if my employer let me choose, I would go for a split cloud infrastructure - run everything on one provider with a hot standby in another. Then your security concerns increase, but even if Amazon worldwide goes down unless the whole internet is out I can spin up my Digital Ocean (or whatever).