![](https://secure.gravatar.com/avatar/533df7a4152bdd49983da377fabafe74.jpg?s=120&d=mm&r=g)
Jeremy caught the issue and reached out to Amazon a couple of weeks ago regarding the areas. It might help for those having issues to reach out to Amazon directly to knock it loose on their side. On Fri, Sep 6, 2024, 12:36 PM Isaiah Olson <iolson@paulbunyan.net> wrote:
It looks like Amazon’s port Ethernet6/1/1 on the MICE Core switch has some pretty significant errors.
*From:* MICE Discuss <MICE-DISCUSS@LISTS.IPHOUSE.NET> *On Behalf Of *Miles McCredie *Sent:* Friday, September 06, 2024 12:12 PM *To:* MICE-DISCUSS@LISTS.IPHOUSE.NET *Subject:* Re: [MICE-DISCUSS] [EXTERNAL] - [MICE-DISCUSS] Odd AWS transfer issue
Maybe someone could check for errors on the MICE ports for Atomic Data and Amazon.
Assuming the MICE ports are clean, it is probably worth following up with the Amazon NOC, peering-to@amazon.com. They did some work on their Minneapolis peering routers recently though I’m not sure what they were doing.
[image: Image removed by sender.]
*Miles McCredie * Principal Network Engineer II-Core IP
Office: 6052755192 Miles.McCredie@midco.com
*Midco.com*
Let's go beyond.
*From:* MICE Discuss <MICE-DISCUSS@LISTS.IPHOUSE.NET> *On Behalf Of *Danny Meister *Sent:* Friday, September 6, 2024 11:41 AM *To:* MICE-DISCUSS@LISTS.IPHOUSE.NET *Subject:* [EXTERNAL] - [MICE-DISCUSS] Odd AWS transfer issue
*CAUTION:* This email originated from outside of *MIDCO*. *Do not click links or open attachments unless you recognize the sender and know the content is safe.*
Good Afternoon,
I've been trying to track down a strange packet loss issue when a customer of mine is downloading files from S3 in the east region. We peer with AWS via MICE and traffic to east routes via MICE.
Could any of you folks peered with AWS via MICE try downloading this file a bunch of times and reports the results?
http://ofw-atomic-data-test-region-us-east-1.s3-website-us-east-1.amazonaws....
Basically, we see fast transfers maybe 80% of the time and then it will be slow, like sub 1Mb slow (we can replicate the issue with iperf to an EC2 instance also). Captures show lots of retransmits, etc during a slow transfer. If I shut down my AWS peering the problem goes away. I was able to test from another client of mine connected to Arvig and I'm pretty sure I can replicate the issue from there. So, I'm not convinced it's just me. On the other hand, if there is a problem, I'm surprised no one else has noticed!
Thanks!
*Danny Meister *
Infrastructure Engineer
612.466.2071 | DMeister@atomicdata.com
[image: Logo] <https://www.atomicdata.com/?utm_campaign=signature&utm_source=EmailSignature>
250 Marquette Avenue <https://www.google.com/maps/search/250+Marquette+Avenue?entry=gmail&source=g> South | Minneapolis, MN | 55401
*atomicdata.com* <https://www.atomicdata.com/?utm_campaign=signature&utm_source=EmailSignature> | *Services* <https://www.atomicdata.com/services?utm_campaign=Signature&utm_source=Email> | *Blog* <https://www.atomicdata.com/blog?utm_campaign=Signature&utm_source=Email> *Atomic Data Welcomes Chris Heim as CEO* <https://www.atomicdata.com/atomic-data-welcomes-chris-heim-as-ceo/?utm_source=EmailSignature>
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1