MUST READ: Meaningful Availability

Defining service availability using the famous X nines (and all the hacks like “planned downtime doesn’t count”) is pretty useless in a highly distributed system where the only thing that really matters is the user experience, not ping response times. One should ask what precisely should we be measuring, and how could we make sure we can act on the measurements

More details in a concise analysis of the Meaningful Availability paper by the one-and-only The Morning Paper.

Adding the Fallback Pool to the Load Balancing UI and other significant UI enhancements

The Cloudflare Load Balancer was introduced over three years ago to provide our customers with a powerful, easy to use tool to intelligently route traffic to their origins across the world. During the initial design process, one of the questions we had to answer was ‘where do we send traffic if all pools are down?’ We did not think it made sense just to drop the traffic, so we used the concept of a ‘fallback pool’ to send traffic to a ‘pool of last resort’ in the case that no pools were detected as available. While this may still result in an error, it gave an eyeball request a chance at being served successfully in case the pool was still up.

As a brief reminder, a load balancer helps route traffic across your origin servers to ensure your overall infrastructure stays healthy and available. Load Balancers are made up of pools, which can be thought of as collections of servers in a particular location.

Over the past three years, we’ve made many updates to the dashboard. The new designs now support the fallback pool addition to the dashboard UI. The use of a fallback pool is incredibly helpful in Continue reading

Juniper MX Upgrades Causing Overheating

Juniper changed the way they do temperature management on MX240 and MX480 chassis devices, somewhere between 15.1 and 17.3. The net result is that your chassis might run hotter after you upgrade, which can lead to the system shutting down some optics. Probably not what you want. Luckily there’s a few hidden commands you can use to change this behavior

“Optics will be disabled…”

Post upgrade, you might see higher temperatures reported by show chassis fpc. This system was reporting temperatures in the low 30s, now it reports 50:

1
2
3
4
5
6
7
8
9
lindsayh@MX240> show chassis fpc
                     Temp  CPU Utilization (%)   CPU Utilization (%)  Memory    Utilization (%)
Slot State            (C)  Total  Interrupt      1min   5min   15min  DRAM (MB) Heap     Buffer
  0  Empty
  1  Online            50     22          1       22     22     22    2048       38         21
  2  Empty

{master}
lindsayh@MX240>

On its own, that’s OK, until you start seeing log messages like this:

1
FPC 1 temperature over 50 degrees C; non-high-temperature tolerant optics will be disabled in 58 seconds if condition persists

Yeah that’s not good, especially when it carries out the threat, and Continue reading

Juniper MX Upgrades Causing Overheating

Juniper changed the way they do temperature management on MX240 and MX480 chassis devices, somewhere between 15.1 and 17.3. The net result is that your chassis might run hotter after you upgrade, which can lead to the system shutting down some optics. Probably not what you want. Luckily there’s a few hidden commands you can use to change this behavior

“Optics will be disabled…”

Post upgrade, you might see higher temperatures reported by show chassis fpc. This system was reporting temperatures in the low 30s, now it reports 50:

1
2
3
4
5
6
7
8
9
lindsayh@MX240> show chassis fpc
                     Temp  CPU Utilization (%)   CPU Utilization (%)  Memory    Utilization (%)
Slot State            (C)  Total  Interrupt      1min   5min   15min  DRAM (MB) Heap     Buffer
  0  Empty
  1  Online            50     22          1       22     22     22    2048       38         21
  2  Empty

{master}
lindsayh@MX240>

On its own, that’s OK, until you start seeing log messages like this:

1
FPC 1 temperature over 50 degrees C; non-high-temperature tolerant optics will be disabled in 58 seconds if condition persists

Yeah that’s not good, especially when it carries out the threat, and Continue reading

Cloud Foundry Foundation Adopts KubeCF Runtime

Daily Roundup: AWS Pledges $20M to COVID-19 Testing

Heavy Networking 507: Build And Run A Multi-Cloud Network Architecture With Aviatrix (Sponsored)

Heavy Networking 507: Build And Run A Multi-Cloud Network Architecture With Aviatrix (Sponsored)

Cloud networking is a challenge, to say the least. And when you've got deployments running in different public clouds, your challenges multiply. On today's Heavy Networking, sponsor Aviatrix joins us to discuss their cloud networking architecture, which provides a consistent operational environment on top of cloud-native constructs. Our guests are Hammad Alam and Shahzad Ali, both Principal Cloud Solutions Architects at Aviatrix.

The post Heavy Networking 507: Build And Run A Multi-Cloud Network Architecture With Aviatrix (Sponsored) appeared first on Packet Pushers.

Fast Friday Thoughts on Where We Are

It’s been a crazy week. I know the curse is “May you live in interesting times,” but I’m more than ready for things to be less interesting for a while. It’s going to take some time to adjust to things. From a networking perspective, I have a few things that have sprung up.

• Video conferencing is now a big thing. Strangely, Cisco couldn’t make video the new phone. But when people are stuck at home now we need to do video again? I get that people have a need to see each other face-to-face. But having worked from home for almost seven years at this point I can tell you video isn’t a necessity. It’s a nice option, but you can get a lot accomplished with video calls and regular emails.
• Along side this is the fact that the push to put more video out there is causing applications to reach their breaking points. Zoom, which is fairing the best out of all of them so far, had some issues on Thursday morning. Tripling the amount of traffic that’s going out and making it very sensitive to delay and jitter is going expose a lot of flaws in the system.
• Continue reading

AWS Pledges $20M to COVID-19 Testing as Cloud Giants, Chipmakers Fight Virus With HPC

Enea Taps Ampere for Arm-based Edge uCPE

Microsoft Skins AWS in Forrester Cloud Ranking

Cisco Donates OpenRoaming as WiFi Standard

Invitation: Virtual Happy Hour The Second – TONIGHT 30 Mar 2020 2100BST/1600EST/1300PST

An open videoconference to be nice to each other, blow the Corona blues away.

The post Invitation: Virtual Happy Hour The Second – TONIGHT 30 Mar 2020 2100BST/1600EST/1300PST appeared first on EtherealMind.

The Serverlist: Workers Secrets, Serverless Supremacy, and more!

Check out our thirteenth edition of The Serverlist below. Get the latest scoop on the serverless space, get your hands dirty with new developer tutorials, engage in conversations with other serverless developers, and find upcoming meetups and conferences to attend.

Sign up below to have The Serverlist sent directly to your mailbox.

BGP – the right tool for so many jobs

The post BGP – the right tool for so many jobs appeared first on Noction.

Weekly Wrap: Coronavirus No Match for Oracle’s Larry Ellison

BGP+SPF for Hyperscale/Massively Scale Datacenter Deployment

Remote Work Isn’t Just Video Conferencing: How We Built CloudflareTV

At Cloudflare, we produce all types of video content, ranging from recordings of our Weekly All-Hands to product demos. Being able to stream video on demand has two major advantages when compared to live video:

1. It encourages asynchronous communication within the organization
2. It extends the life time value of the shared knowledge

Historically, we haven’t had a central, secure repository of all video content that could be easily accessed from the browser. Various teams choose their own platform to share the content. If I wanted to find a recording of a product demo, for example, I’d need to search Google Drive, Gmail and Google Chat with creative keywords. Very often, I would need to reach out to individual teams to finally locate the content.

So we decided we wanted to build CloudflareTV, an internal Netflix-like application that can only be accessed by Cloudflare employees and has all of our videos neatly organized and immediately watchable from the browser.

We wanted to achieve the following when building CloudflareTV:

• Security: make sure the videos are access controlled and not publicly accessible
• Authentication: ensure the application can only be accessed by Cloudflare employees
• Tagging: allow the videos to be categorized so they can Continue reading

Opinion : Engineers, Solutions and Problems

The tension between iteration and guarantees.

The post Opinion : Engineers, Solutions and Problems appeared first on EtherealMind.