Juniper Default ARP Policer

Juniper devices have a default ARP policer that drops ARP requests and responses over 150kbps. By default, this is an aggregate policer that applies to all interfaces. This can lead to unexpected behavior when high levels of ARP on one interface lead to BGP session drops on another interface. You can’t change the default policer limits, but you can create a new policer, with higher limits.

Problem: IPv4 BGP Session Flaps on PNI

I was investigating a problem reported by one of our Transit providers. Once a day or so, our IPv4 BGP session with them would flap. The interface itself was stable, and the IPv6 session remained up. One particular site was seeing this more than others. The sites used different platforms, but were running the same code version.

The curious thing was the logs - we saw log messages saying that we had a notification message saying NOTIFICATION received from 192.0.2.188 (External AS 64498): code 4 (Hold Timer Expired Error). The syslog included this hold timer 30s, hold timer remain 0s, last sent 2s. So our router thought it was sending regular KEEPALIVE messages, but the remote end thought it had missed too many.

Looking Continue reading

Juniper Default ARP Policer

Juniper devices have a default ARP policer that drops ARP requests and responses over 150kbps. By default, this is an aggregate policer that applies to all interfaces. This can lead to unexpected behavior when high levels of ARP on one interface lead to BGP session drops on another interface. You can’t change the default policer limits, but you can create a new policer, with higher limits.

Problem: IPv4 BGP Session Flaps on PNI

I was investigating a problem reported by one of our Transit providers. Once a day or so, our IPv4 BGP session with them would flap. The interface itself was stable, and the IPv6 session remained up. One particular site was seeing this more than others. The sites used different platforms, but were running the same code version.

The curious thing was the logs - we saw log messages saying that we had a notification message saying NOTIFICATION received from 192.0.2.188 (External AS 64498): code 4 (Hold Timer Expired Error). The syslog included this hold timer 30s, hold timer remain 0s, last sent 2s. So our router thought it was sending regular KEEPALIVE messages, but the remote end thought it had missed too many.

Looking Continue reading

Money Moves: April 2020

Nokia faced a hostile takeover bid; Google eyeing a D2iQ purchase; T-Mobile to slash $30M in cloud...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

Rakuten Mobile ‘Working Like Crazy’ to Deploy 5G Next Month

"We might be slowed down on the number of base stations that we are able to construct, but I...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

Interesting: Hugo with Docsy and AWS Amplify

Mat Jovanovic decided to follow my lead and migrate his blog from Blogger to Hugo, using Docsy theme, AWS Amplify as the CI/CD pipeline, and AWS S3 as the hosting platform.

Nice job… but he did way more than that - he documented the whole process, including tool selection, setup, and Blogger migration.

Thank you Mat! Every time I see someone publishing blog posts about open-source tools on Medium I’ll send them a link to your blog (with a comment “this is how you should blog about open-source solutions").

BPF Compiler Collection – BCC in short

Network Tracing sometimes is really important, although most of the times tcpdump utility is quite handy there are other tools that can make life much easier.

while am no expert in eBPF and scripts, i do know how to use bcc-tools in some scenarios.

https://github.com/iovisor/bcc/blob/master/INSTALL.md#ubuntu—binary

Alright What is the scenario:

Let’s start with something small, you want to measure TCP connection latency.

Other interesting options, you can capture lifetime, stats and most importantly TCP-Retransmissions

Ebpf filters are safer and more powerful to implement, give this a consideration during any Linux troubleshooting scenarios.

-Rakesh

Daily Roundup: Cisco Patches Firewall Bugs

Cisco patched firewall bugs; Amazon threw another JEDI tantrum fit ; and Dish fought off critics,...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

Heavy Networking 516: Is LISP The Overlay Of The Future?

On today's Heavy Networking podcast, guest Cory Steele visits the podcast to make the case that overlays such as LISP offer unique benefits for the network. Greg Ferro disagrees, and makes the case for protocols like QUIC, TLS, and IPSec, and argues for the concept of end-to-end connectivity as the IP network was intended.

The post Heavy Networking 516: Is LISP The Overlay Of The Future? appeared first on Packet Pushers.

Getting Started with pyATS (and Genie)

What is Python Automated Test System (pyATS)? None of the answers I found to this question really made much sense to me initially. A Python3 based Test Automation and Validation Framework developed by Cisco (but open and extensible to any vendor) is probably the best short answer but still too vague. Add in Genie because, READ MORE

The post Getting Started with pyATS (and Genie) appeared first on The Gratuitous Arp.

Internet During Shutdown: Do We Need More Internet?

The Internet Society India Chennai Chapter organized a virtual roundtable in March, a few days after the Indian government announced a three-week nationwide lockdown. The virtual roundtable was a conversation on the importance of keeping the Internet open, and on the ways in which the Internet community could contribute to COVID-19 response and recovery in India and around the globe.
The virtual roundtable brought together a wide range of Internet stakeholders, including Andrew Sullivan, Jane Coffin, Mike Godwin, Yrjö Länsipuro, Wolfgang Kleinwächter, Olivier Crepin-Leblond, Sébastien Bachollet, Samiran Gupta, and Glen McKnight, as well as members of the Chapter from civil society and the private sector.

Some key highlights and takeaways from the virtual roundtable include the following:

The COVID-19 pandemic underscores the importance of the Internet. Without access, people are unable to communicate with family members and health workers, and participate in online learning and remote work.

The pandemic has clearly exposed the inequalities in Internet access and affordability – the digital divide across the region. Connecting the billions of people who are not yet connected must be a priority. At the same time, their privacy and autonomy must be protected.

Internet technologies can help us fight against the pandemic. Continue reading

Dish Fights Off Critics, Claims 5G Efforts on Track

Despite the calamity brought on by COVID-19, Dish still plans to launch 5G service in a single...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

Weekly Wrap: Nutanix Furloughs 25% of Workforce Citing COVID-19

SDxCentral Weekly Wrap for May 8, 2020: Nutanix to furlough 25% of its workforce; IBM wears Red Hat...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

‘SD-WAN Has to Evolve,’ Says Cisco Exec

Steven Wood, Cisco’s principal engineer of enterprise architectures and SD-WAN, made the...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

Amazon Files Another JEDI Protest, Microsoft Says ‘Stop Asking for Do-Over’

Microsoft called Amazon’s latest JEDI protest “yet another attempt to force a re-do because...

Read More »

© SDxCentral, LLC. Use of this feed is limited to personal, non-commercial use and is governed by SDxCentral's Terms of Use (https://www.sdxcentral.com/legal/terms-of-service/). Publishing this feed for public or commercial use and/or misrepresentation by a third party is prohibited.

The Devil Is In The Licensing

If you don’t already know that I’m a co-host of a great podcast we do at Gestalt IT, here’s a great way to jump in. This episode was a fun one to record and talk about licensing:

Sometimes I have to play the role of the genial host and I don’t get to express my true opinion on things. After all, a good podcast host is really just there to keep the peace and ensure the guests get to say their words, right?

Double Feature

I once said that every random feature in a certain network operating system somehow came from a million-dollar PO that needed to be closed. It reflects my personal opinion that sometimes the things we see in code don’t always reflect reality. But how do you decide what to build if you’re not listening to customers?

It’s a tough gamble to take. You can guess at what people are going to want to include and hope that you get it right. Other times you’re going to goof and put something your code that no one uses. It’s a delicate balance. One of the biggest traps that a company can fall into is waiting for their Continue reading

CUBIC and HyStart++ Support in quiche

CUBIC and HyStart++ Support in quiche

quiche, Cloudflare's IETF QUIC implementation has been running CUBIC congestion control for a while in our production environment as mentioned in Comparing HTTP/3 vs. HTTP/2 Performance). Recently we also added HyStart++  to the congestion control module for further improvements.

In this post, we will talk about QUIC congestion control and loss recovery briefly and CUBIC and HyStart++ in the quiche congestion control module. We will also discuss lab test results and how to visualize those using qlog which was recently added to the quiche library as well.

QUIC Congestion Control and Loss Recovery

In the network transport area, congestion control is how to decide how much data the connection can send into the network. It has an important role in networking so as not to overrun the link but also at the same time it needs to play nice with other connections in the same network to ensure that the overall network, the Internet, doesn’t collapse. Basically congestion control is trying to detect the current capacity of the link and tune itself in real time and it’s one of the core algorithms for running the Internet.

QUIC congestion control has been written based on many years of TCP Continue reading

1 … 963 964 965 966 967 … 3,843