October 2022 - mice-discuss - lists.micemn.net

Re: Anyone lose peering with CloudFlare this morning at 4:13 am?

by Jeremy Lumby

I have been told that they have replaced the same Juniper switch two different times, and the two different replacements keep showing the same errors as the original. This is corroborated with their MICE link coming up briefly twice since they originally rerouted. From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of David Farmer Sent: Monday, October 31, 2022 2:48 PM To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Anyone lose peering with CloudFlare this morning at 4:13 am? Any word from Cloudflare? They will have been in re-routed mode for three weeks on Wednesday. They have theoretically completed two maintenance events since then; https://www.cloudflarestatus.com/incidents/txnkvw6frp25 https://www.cloudflarestatus.com/incidents/pqwsjx82wdt2 Thanks On Thu, Oct 20, 2022 at 11:08 AM Jeremy Lumby <jlumby(a)mnvoip.com> wrote: Yes, the last time they had a failure in Minneapolis it took them just shy of 2 weeks to get back online. From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Steve Savoy Sent: Thursday, October 20, 2022 10:39 AM To:MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Anyone lose peering with CloudFlare this morning at 4:13 am? I am pretty new to the CloudFlare peering. Anyone else go through the “Re-routed” scenario before? Will it eventually become “Operational”? Thank you, Steve From: MICE Discuss <MICE-DISCUSS(a)LISTS.IPHOUSE.NET> On Behalf Of David Farmer Sent: Wednesday, October 12, 2022 11:59 AM To:MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Anyone lose peering with CloudFlare this morning at 4:13 am? If you check Cloudflare’s status page you will see Minneapolis is in “Re-routed” state. https://www.cloudflarestatus.com On Wed, Oct 12, 2022 at 10:45 Steve Savoy <stevesavoy(a)wctatel.com> wrote: Yes – ours went down at the same time. Steve From: MICE Discuss <MICE-DISCUSS(a)LISTS.IPHOUSE.NET> On Behalf Of Frank Bulk Sent: Wednesday, October 12, 2022 10:42 AM To:MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: [MICE-DISCUSS] Anyone lose peering with CloudFlare this morning at 4:13 am? Both IPv4 and IPv6. I didn’t see any exceeding of prefix limits. Frank Oct 12 04:13:10 207.32.15.14 (sxct-north-2.sxcy.fbnt.netins.net <http://sxct-north-2.sxcy.fbnt.netins.net>) SiouxCity-Fibernet-Arista Rib: %BGP-5-ADJCHANGE: peer 206.108.255.45 (AS13335.micemn.net <http://AS13335.micemn.net>) (AS 13335 CLOUDFLARENET, US) old state Established event HoldTime new state Idle Oct 12 04:13:10 207.32.15.14 (sxct-north-2.sxcy.fbnt.netins.net <http://sxct-north-2.sxcy.fbnt.netins.net>) SiouxCity-Fibernet-Arista Rib: %BGP-5-ADJCHANGE: peer 2001:504:27::3417:0:1 (AS13335.micemn.net <http://AS13335.micemn.net>) (AS 13335 CLOUDFLARENET, US) old state Established event HoldTime new state Idle ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 Confidentiality Notice: This correspondence is the property of Winnebago Cooperative Telecom Association and is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 -- =============================================== David Farmer Email:farmer@umn.edu <Email%3Afarmer(a)umn.edu> Networking & Telecommunication Services Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414-3029 Cell: 612-812-9952 =============================================== --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 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 -- =============================================== David Farmer Email:farmer@umn.edu <Email%3Afarmer(a)umn.edu> Networking & Telecommunication Services Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414-3029 Cell: 612-812-9952 =============================================== ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

Oct. 31, 2022

Re: Anyone lose peering with CloudFlare this morning at 4:13 am?

by David Farmer

Any word from Cloudflare? They will have been in re-routed mode for three weeks on Wednesday. They have theoretically completed two maintenance events since then; https://www.cloudflarestatus.com/incidents/txnkvw6frp25 https://www.cloudflarestatus.com/incidents/pqwsjx82wdt2 Thanks On Thu, Oct 20, 2022 at 11:08 AM Jeremy Lumby <jlumby(a)mnvoip.com> wrote: > Yes, the last time they had a failure in Minneapolis it took them just shy > of 2 weeks to get back online. > > > > *From:* MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] *On Behalf > Of *Steve Savoy > *Sent:* Thursday, October 20, 2022 10:39 AM > *To:* MICE-DISCUSS(a)LISTS.IPHOUSE.NET > *Subject:* Re: [MICE-DISCUSS] Anyone lose peering with CloudFlare this > morning at 4:13 am? > > > > I am pretty new to the CloudFlare peering. Anyone else go through the > “Re-routed” scenario before? Will it eventually become “Operational”? > > > > Thank you, > > Steve > > > > *From:* MICE Discuss <MICE-DISCUSS(a)LISTS.IPHOUSE.NET> *On Behalf Of *David > Farmer > *Sent:* Wednesday, October 12, 2022 11:59 AM > *To:* MICE-DISCUSS(a)LISTS.IPHOUSE.NET > *Subject:* Re: [MICE-DISCUSS] Anyone lose peering with CloudFlare this > morning at 4:13 am? > > > > If you check Cloudflare’s status page you will see Minneapolis is in > “Re-routed” state. > > > > https://www.cloudflarestatus.com > > > > > > On Wed, Oct 12, 2022 at 10:45 Steve Savoy <stevesavoy(a)wctatel.com> wrote: > > Yes – ours went down at the same time. > > > > Steve > > > > *From:* MICE Discuss <MICE-DISCUSS(a)LISTS.IPHOUSE.NET> *On Behalf Of *Frank > Bulk > *Sent:* Wednesday, October 12, 2022 10:42 AM > *To:* MICE-DISCUSS(a)LISTS.IPHOUSE.NET > *Subject:* [MICE-DISCUSS] Anyone lose peering with CloudFlare this > morning at 4:13 am? > > > > Both IPv4 and IPv6. I didn’t see any exceeding of prefix limits. > > > > Frank > > > > Oct 12 04:13:10 207.32.15.14 (sxct-north-2.sxcy.fbnt.netins.net) > SiouxCity-Fibernet-Arista Rib: %BGP-5-ADJCHANGE: peer 206.108.255.45 ( > AS13335.micemn.net) (AS 13335 CLOUDFLARENET, US) old state Established > event HoldTime new state Idle > > Oct 12 04:13:10 207.32.15.14 (sxct-north-2.sxcy.fbnt.netins.net) > SiouxCity-Fibernet-Arista Rib: %BGP-5-ADJCHANGE: peer 2001:504:27::3417:0:1 > (AS13335.micemn.net) (AS 13335 CLOUDFLARENET, US) old state Established > event HoldTime new state Idle > > > > > ------------------------------ > > To unsubscribe from the MICE-DISCUSS list, click the following link: > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 > > Confidentiality Notice: This correspondence is the property of Winnebago > Cooperative Telecom Association and is for the sole use of the intended > recipient(s) and may contain confidential and privileged information. Any > unauthorized review, use, disclosure or distribution is prohibited. If you > are not the intended recipient, please contact the sender by reply email > and destroy all copies of the original message. > > > ------------------------------ > > To unsubscribe from the MICE-DISCUSS list, click the following link: > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 > > -- > > =============================================== > David Farmer Email:farmer@umn.edu > Networking & Telecommunication Services > Office of Information Technology > University of Minnesota > 2218 University Ave SE Phone: 612-626-0815 > Minneapolis, MN 55414-3029 Cell: 612-812-9952 > =============================================== > > > ------------------------------ > > 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 > -- =============================================== David Farmer Email:farmer@umn.edu Networking & Telecommunication Services Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414-3029 Cell: 612-812-9952 ===============================================

Oct. 31, 2022

Re: Proposal: MICE Static MAC address Only

by Mike Johnston

Well articulated, as always, David. Thank you. I am in favor of the use of static MAC ACL, even if that requires manual changes to that ACL, and the additional coordination that may consume. A few thoughts, though.... Maybe 1G links are exempt from this? Since a couple second switching loop from a 1G port is not as impactful? ????? Thoughts from others, please. Using a static MAC ACL could make troubleshooting a failure more problematic. For example, replacing a line card to "see if that fixes the issue" could send a tech down a trail of red hearings. Even if the member is fully aware that they need to coordinate the MAC change, this sort work is often being done in response to an unexpected outage. The added time spend on the coordination could be seen as frustrating. But, for some members it's not a big deal, and can use other links until their MICE connectivity is restored. I've heard that some members are using a virtual MAC address, which is probably the way to avoid needing to coordinate in the first place.

Oct. 31, 2022

MICE - IXP Manager - Change Control Access

by Justin Krejci

Spinning off a new thread from the static/mac-limit/port-security conversation. The IXP Manager can generate an instruction command and makes it available to the Arista(s). The switch(es) could pull the config using a script that further sanitizes/validates the config its pulling before applying. This would prevent IXP Manager from having its own direct access to the switch. This could also be the case for pulling data from the switch(es) as well. Patterned after a TR-069 or similar concept, the switch could phone home to the IXP manager (once a minute cronjob for example) and see if there is anything the IXP manager wants done. A simple text based file (json, etc) is generated by the IXP manager and the Arista would then just query this file and parse out instructions and perform actions as instructed or ignore them if malformed or otherwise not allowed. Obviously this is would be only for Arista switch(es) managed by MICE. I don't have any IXP Manager experience beyond simple user but I know Arista can do this fairly easily. It could also just be a RESTful API even. Maybe something like this already exists I have no idea but I would highlight the two things this concept does 1: Allows IXP manager to affect change without direct access to the switch(es) 2: Allows the Arista switch(es) to ultimately decide to accept or ignore instructions to make config changes, send info, perform other actions like bounce my member port, apply TCP/179 filtering to my port, show me some log data, etc. This concept could also apply to the route servers too since they also run a *nix OS like Arista. This would not be a simple/quick idea but I feel would check various boxes for people: air gap, allows automation/self-service, and if automation fails it results in a fallback to manual only changes like we are doing today anyways, etc. Since I know MICE does have money, it could be dev work hired out to someone or a team to produce and then MICE retains rights to the resultant code. We have enough people across our membership that I am sure we could validate the code is sound and also perform bugfixes as needed going forward, perhaps even make it public, GPL style. Maybe IXP Manager already has some of this functionality? Last second additional thought, simple instructions (eg add a second MAC address, show me log data for my port) get automatically approved. Bigger changes (remove/change my MAC address, some other action) need to get some sort of manual approval by the tech team so it can be scrutinized first. None of this needs to be done in IXP Manager either but some other custom, purpose built web portal. Maybe I'm way overthinking this. Food for thought at any rate. -Justin

Oct. 31, 2022

Re: Proposal: MICE Static MAC address Only

by David Farmer

I have similar concerns as Andrew. However, I'm cognizant of Jeremy's concerns as well. Nevertheless, if we are going to fix our old problems, we have to be willing to accept that we are likely to create other new problems. The hard part is deciding if the old problems, the known problems, are worse than the potential new problems, the potentially unknown problems, and which of the potentially new problems we are willing to accept. We seem to have a consensus that this particular old problem needs to be addressed; what is not clear is if we have a consensus on accepting the potential new problems and which of those potential problems we are willing to accept. Note either way we are accepting new problems; we are either accepting potential new problems caused by the automation of IX Manager. Otherwise, we are accepting the potential new problems caused by a static MAC ACL list with only manual changes; I expect delays in changes primarily. Both options are going to produce problems. And if they are delays impacting one of the large content peers they could have as significant of an impact as the original problem we are trying to solve. Thanks On Mon, Oct 31, 2022 at 9:02 AM Andrew Hoyos <hoyosa(a)gmail.com> wrote: > I disagree with static MAC address filtering unless we beef up operational > ability to change/adjust in a reasonable timeframe, or offer a self service > portal to do so (ie IXP Manager implementation first) > > Sent from my iPhone > > On Oct 31, 2022, at 9:59 PM, Steve Howard <showard(a)paulbunyan.net> wrote: > > > Below is a message indicating the current policy as decided by the MICE > Board in December 2019 is that the MAC address limit is 1. > > > The subject of Jay's message indicates "Static" MAC, but, the text seems > to indicate a limit of 1. I like the 1 address limit, but, am not a fan of > Static. Was the current issue caused by somebody not having a proper MAC > address limit installed? > > > On 12/13/19 18:00, Richard Laager wrote: > > The board has approved the change in MAC address limit from 5 to 1. > > NOTE: This is orthogonal to the dedicated remote switch question, which > is still pending. Non-dedicated remote switches can, for example, > enforce this per-VLAN. > > We have already confirmed that most participants are using only 1 MAC > address, and some that were using more than 1 have addressed that after > being contacted. As discussed below, this change will be rolled out in a > way to keep everyone working, by grandfathering if necessary. > > On 10/28/19 3:25 PM, Richard Laager wrote: > > We have discussed this a bit in the past, and this came up at the last > UG. I'm looking for feedback from the group before formally proposing > this to the board for a vote. > > I first brought this up to the technical committee, CCing the board. I > have heard no objections. Jeremy is "in favor of it 100%". > > Based on our quick conversation after the UG, I _think_ Anthony is in > favor as well; we both made notes to follow up on this. > > ----- > > I am proposing that MICE change its MAC address limit from 5 MACs per > port to 1 MAC per port. 1 MAC address is enough for normal scenarios and > this would further limit the potential for problems on the fabric. This > restriction is one that SIX and AMS-IX both have, with the latter being > known for their excellent configuration guide for participants:https://www.ams-ix.net/ams/documentation/config-guide > > To be clear, this is still only for end ports, not ports facing remote > switches, of course. The remote switches are responsible for enforcing > the current MAC limit, and would be responsible for changing those > settings as described here. > > If someone is swapping their MICE-facing router, the resulting port flap > on the MICE/remote switch will clear the limit anyway, if they are > directly connected. If they are unable to flap the port (e.g. because > they have someone else's layer 2 gear in the middle), they will have to > either work with that carrier or their switch operator (MICE/remote) to > flap the port. This is a non-zero burden, but I think it's pretty > minimal in practice. Additionally, if someone plans to make an equipment > swap, we would increase the limit in advance, temporarily, upon request. > > For reference, SIX goes further and locks you to a particular MAC > address in an ACL, so you have to contact them to change your MAC. I am > not proposing that. Still, I've been through that a couple of times, and > even that isn't really too big of a deal. So I don't think the change to > a limit of 1 MAC at MICE will be particularly onerous. > > There are a couple of participants who have two IP address sets (one > IPv4 and one IPv6) on the same port. For those ports, the limit > obviously must be at least 2 MACs, but I propose 3 MACs as the limit. If > they're using two IPs, they probably are doing so to get some redundancy > out of it. For this particular use case, needing to flap the port > defeats that redundancy goal, as it breaks the other router. Having a > MAC limit of 3 would allow them to swap a device without needing to flap > the port. We would grandfather these setups until they are looking for a > port-type upgrade; at that point they would need to get to one IP > address set per port (by splitting into two ports or reducing to one IP > address set) or request an exception as outlined below. The exception > process would allow us to better evaluate this use case at that time. > > There are some members who have multiple MACs showing up now, despite > using only one device. For those, I am proposing we set the MAC limit to > however many MACs are currently showing up on their port. That is, they > would be grandfathered at their current situation for now. At the time > of this change, we would encourage them to investigate and fix the > multiple MACs issue at their leisure. In the future, if they are doing a > equipment swap (especially like-for-like), we would again encourage them > to address it, but would otherwise leave their higher limit in place. If > they do a port-type upgrade, we would not bring the grandfathering over. > If they needed, they could ask for an official exception at that time. > > There may be cases where people cannot reasonably fix their routers to > use only 1 MAC, other scenarios I haven't considered, or things that may > come up in the future. For that reason, I think we should allow for the > possibility of exceptions upon request to the board. In practice, I > would expect the board would confer with the technical committee and/or > the technical committee would be bringing these to the board anyway. The > goal is to strike a reasonable balance between protecting the fabric > without preventing networks from connecting. > > > > > > On 10/31/22 08:27, Jay Hanke wrote: > > I propose we change the MICE required port security settings to be > hard filtered for a single static mac address per port. > > Over the years, we've had several incidents where floods have occurred > prior to port being error-disabled during a looping event. > > > > > ------------------------------ > > 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 > -- =============================================== David Farmer Email:farmer@umn.edu Networking & Telecommunication Services Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414-3029 Cell: 612-812-9952 ===============================================

Oct. 31, 2022

Re: [EXTERNAL] Re: [MICE-DISCUSS] Proposal: MICE Static MAC address Only

by Chris Wopat

Regarding the method of 'updating' the mac address list, there should probably be the ability to address maintenance, ie temporarily add a 2nd mac address to permit list. See SIX's faq. Unshore how handled internally, but one can email with a pending change (ie router replacement) to have 2 permitted for a period so maintenance can go smoothly. https://www.seattleix.net/faq -- Chris Wopat Network Engineer, WiscNet wopat(a)wiscnet.net 608-210-3965

Oct. 31, 2022

Re: Proposal: MICE Static MAC address Only

by Jeremy Lumby

I like that idea -----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Jay Hanke Sent: Monday, October 31, 2022 9:27 AM To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Proposal: MICE Static MAC address Only The switch access could be air-gapped. ie there could be a script that generates the new mac-address list and then it could be manually applied. On Mon, Oct 31, 2022 at 9:22 AM Jeremy Lumby <jlumby(a)mnvoip.com> wrote: > > I have mixed feelings about giving IXP manager access to the switch. The main reason is that if a member is proactive they can request the second MAC ahead of time, and it can be temporarily added for the transition. As it is most requests are processed in a couple hours. IXP manager is a program that none of the volunteers have a great grasp on, and lack of a timely security update on a portal that is made available to the entire internet in an invitation to get hacked. If it has direct access to the switch things can go wrong quickly. While I am concerned as it is about IXP manager having access to the route servers, they are an extra and not 100% necessary for the IX to function. > > -----Original Message----- > From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Jay Hanke > Sent: Monday, October 31, 2022 9:05 AM > To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET > Subject: Re: [MICE-DISCUSS] Proposal: MICE Static MAC address Only > > The error disable takes 1-2 seconds to clamp down and disable the > port. This results in a short, but severe, broadcast storm. > > The churn rate on members changing mac addresses is fairly low. > > I agree that we should automate the process so it can be done via > self-service by the members. > > On Mon, Oct 31, 2022 at 8:59 AM Steve Howard <showard(a)paulbunyan.net> wrote: > > > > > > Below is a message indicating the current policy as decided by the MICE Board in December 2019 is that the MAC address limit is 1. > > > > > > The subject of Jay's message indicates "Static" MAC, but, the text seems to indicate a limit of 1. I like the 1 address limit, but, am not a fan of Static. Was the current issue caused by somebody not having a proper MAC address limit installed? > > > > > > On 12/13/19 18:00, Richard Laager wrote: > > > > The board has approved the change in MAC address limit from 5 to 1. > > > > NOTE: This is orthogonal to the dedicated remote switch question, which > > is still pending. Non-dedicated remote switches can, for example, > > enforce this per-VLAN. > > > > We have already confirmed that most participants are using only 1 MAC > > address, and some that were using more than 1 have addressed that after > > being contacted. As discussed below, this change will be rolled out in a > > way to keep everyone working, by grandfathering if necessary. > > > > On 10/28/19 3:25 PM, Richard Laager wrote: > > > > We have discussed this a bit in the past, and this came up at the last > > UG. I'm looking for feedback from the group before formally proposing > > this to the board for a vote. > > > > I first brought this up to the technical committee, CCing the board. I > > have heard no objections. Jeremy is "in favor of it 100%". > > > > Based on our quick conversation after the UG, I _think_ Anthony is in > > favor as well; we both made notes to follow up on this. > > > > ----- > > > > I am proposing that MICE change its MAC address limit from 5 MACs per > > port to 1 MAC per port. 1 MAC address is enough for normal scenarios and > > this would further limit the potential for problems on the fabric. This > > restriction is one that SIX and AMS-IX both have, with the latter being > > known for their excellent configuration guide for participants: > > https://www.ams-ix.net/ams/documentation/config-guide > > > > To be clear, this is still only for end ports, not ports facing remote > > switches, of course. The remote switches are responsible for enforcing > > the current MAC limit, and would be responsible for changing those > > settings as described here. > > > > If someone is swapping their MICE-facing router, the resulting port flap > > on the MICE/remote switch will clear the limit anyway, if they are > > directly connected. If they are unable to flap the port (e.g. because > > they have someone else's layer 2 gear in the middle), they will have to > > either work with that carrier or their switch operator (MICE/remote) to > > flap the port. This is a non-zero burden, but I think it's pretty > > minimal in practice. Additionally, if someone plans to make an equipment > > swap, we would increase the limit in advance, temporarily, upon request. > > > > For reference, SIX goes further and locks you to a particular MAC > > address in an ACL, so you have to contact them to change your MAC. I am > > not proposing that. Still, I've been through that a couple of times, and > > even that isn't really too big of a deal. So I don't think the change to > > a limit of 1 MAC at MICE will be particularly onerous. > > > > There are a couple of participants who have two IP address sets (one > > IPv4 and one IPv6) on the same port. For those ports, the limit > > obviously must be at least 2 MACs, but I propose 3 MACs as the limit. If > > they're using two IPs, they probably are doing so to get some redundancy > > out of it. For this particular use case, needing to flap the port > > defeats that redundancy goal, as it breaks the other router. Having a > > MAC limit of 3 would allow them to swap a device without needing to flap > > the port. We would grandfather these setups until they are looking for a > > port-type upgrade; at that point they would need to get to one IP > > address set per port (by splitting into two ports or reducing to one IP > > address set) or request an exception as outlined below. The exception > > process would allow us to better evaluate this use case at that time. > > > > There are some members who have multiple MACs showing up now, despite > > using only one device. For those, I am proposing we set the MAC limit to > > however many MACs are currently showing up on their port. That is, they > > would be grandfathered at their current situation for now. At the time > > of this change, we would encourage them to investigate and fix the > > multiple MACs issue at their leisure. In the future, if they are doing a > > equipment swap (especially like-for-like), we would again encourage them > > to address it, but would otherwise leave their higher limit in place. If > > they do a port-type upgrade, we would not bring the grandfathering over. > > If they needed, they could ask for an official exception at that time. > > > > There may be cases where people cannot reasonably fix their routers to > > use only 1 MAC, other scenarios I haven't considered, or things that may > > come up in the future. For that reason, I think we should allow for the > > possibility of exceptions upon request to the board. In practice, I > > would expect the board would confer with the technical committee and/or > > the technical committee would be bringing these to the board anyway. The > > goal is to strike a reasonable balance between protecting the fabric > > without preventing networks from connecting. > > > > > > > > > > On 10/31/22 08:27, Jay Hanke wrote: > > > > I propose we change the MICE required port security settings to be > > hard filtered for a single static mac address per port. > > > > Over the years, we've had several incidents where floods have occurred > > prior to port being error-disabled during a looping event. > > > > > > > > ________________________________ > > > > To unsubscribe from the MICE-DISCUSS list, click the following link: > > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 > > > > -- > Jay Hanke, President > South Front Networks > jayhanke(a)southfront.io > Phone 612-204-0000 -- Jay Hanke, President South Front Networks jayhanke(a)southfront.io Phone 612-204-0000

Oct. 31, 2022

Re: Proposal: MICE Static MAC address Only

by Jay Hanke

Yes, that was the intention. On Mon, Oct 31, 2022 at 9:26 AM Jeremy Lumby <jlumby(a)mnvoip.com> wrote: > > I think the best way to phrase it is “static MAC ACL” applied to all member ports. There could be reasons to do static MAC entries in the forwarding table (so traffic does not flood when a large link drops) but we are not talking about that here/yet. > > > > From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Steve Howard > Sent: Monday, October 31, 2022 8:59 AM > To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET > Subject: Re: [MICE-DISCUSS] Proposal: MICE Static MAC address Only > > > > > Below is a message indicating the current policy as decided by the MICE Board in December 2019 is that the MAC address limit is 1. > > > The subject of Jay's message indicates "Static" MAC, but, the text seems to indicate a limit of 1. I like the 1 address limit, but, am not a fan of Static. Was the current issue caused by somebody not having a proper MAC address limit installed? > > On 12/13/19 18:00, Richard Laager wrote: > > The board has approved the change in MAC address limit from 5 to 1. > > > > NOTE: This is orthogonal to the dedicated remote switch question, which > > is still pending. Non-dedicated remote switches can, for example, > > enforce this per-VLAN. > > > > We have already confirmed that most participants are using only 1 MAC > > address, and some that were using more than 1 have addressed that after > > being contacted. As discussed below, this change will be rolled out in a > > way to keep everyone working, by grandfathering if necessary. > > > > On 10/28/19 3:25 PM, Richard Laager wrote: > > We have discussed this a bit in the past, and this came up at the last > > UG. I'm looking for feedback from the group before formally proposing > > this to the board for a vote. > > > > I first brought this up to the technical committee, CCing the board. I > > have heard no objections. Jeremy is "in favor of it 100%". > > > > Based on our quick conversation after the UG, I _think_ Anthony is in > > favor as well; we both made notes to follow up on this. > > > > ----- > > > > I am proposing that MICE change its MAC address limit from 5 MACs per > > port to 1 MAC per port. 1 MAC address is enough for normal scenarios and > > this would further limit the potential for problems on the fabric. This > > restriction is one that SIX and AMS-IX both have, with the latter being > > known for their excellent configuration guide for participants: > > https://www.ams-ix.net/ams/documentation/config-guide > > > > To be clear, this is still only for end ports, not ports facing remote > > switches, of course. The remote switches are responsible for enforcing > > the current MAC limit, and would be responsible for changing those > > settings as described here. > > > > If someone is swapping their MICE-facing router, the resulting port flap > > on the MICE/remote switch will clear the limit anyway, if they are > > directly connected. If they are unable to flap the port (e.g. because > > they have someone else's layer 2 gear in the middle), they will have to > > either work with that carrier or their switch operator (MICE/remote) to > > flap the port. This is a non-zero burden, but I think it's pretty > > minimal in practice. Additionally, if someone plans to make an equipment > > swap, we would increase the limit in advance, temporarily, upon request. > > > > For reference, SIX goes further and locks you to a particular MAC > > address in an ACL, so you have to contact them to change your MAC. I am > > not proposing that. Still, I've been through that a couple of times, and > > even that isn't really too big of a deal. So I don't think the change to > > a limit of 1 MAC at MICE will be particularly onerous. > > > > There are a couple of participants who have two IP address sets (one > > IPv4 and one IPv6) on the same port. For those ports, the limit > > obviously must be at least 2 MACs, but I propose 3 MACs as the limit. If > > they're using two IPs, they probably are doing so to get some redundancy > > out of it. For this particular use case, needing to flap the port > > defeats that redundancy goal, as it breaks the other router. Having a > > MAC limit of 3 would allow them to swap a device without needing to flap > > the port. We would grandfather these setups until they are looking for a > > port-type upgrade; at that point they would need to get to one IP > > address set per port (by splitting into two ports or reducing to one IP > > address set) or request an exception as outlined below. The exception > > process would allow us to better evaluate this use case at that time. > > > > There are some members who have multiple MACs showing up now, despite > > using only one device. For those, I am proposing we set the MAC limit to > > however many MACs are currently showing up on their port. That is, they > > would be grandfathered at their current situation for now. At the time > > of this change, we would encourage them to investigate and fix the > > multiple MACs issue at their leisure. In the future, if they are doing a > > equipment swap (especially like-for-like), we would again encourage them > > to address it, but would otherwise leave their higher limit in place. If > > they do a port-type upgrade, we would not bring the grandfathering over. > > If they needed, they could ask for an official exception at that time. > > > > There may be cases where people cannot reasonably fix their routers to > > use only 1 MAC, other scenarios I haven't considered, or things that may > > come up in the future. For that reason, I think we should allow for the > > possibility of exceptions upon request to the board. In practice, I > > would expect the board would confer with the technical committee and/or > > the technical committee would be bringing these to the board anyway. The > > goal is to strike a reasonable balance between protecting the fabric > > without preventing networks from connecting. > > > > > > > > On 10/31/22 08:27, Jay Hanke wrote: > > I propose we change the MICE required port security settings to be > > hard filtered for a single static mac address per port. > > > > Over the years, we've had several incidents where floods have occurred > > prior to port being error-disabled during a looping event. > > > > > > > > ________________________________ > > 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 -- Jay Hanke, President South Front Networks jayhanke(a)southfront.io Phone 612-204-0000

Oct. 31, 2022

Re: Proposal: MICE Static MAC address Only

by Jay Hanke

The switch access could be air-gapped. ie there could be a script that generates the new mac-address list and then it could be manually applied. On Mon, Oct 31, 2022 at 9:22 AM Jeremy Lumby <jlumby(a)mnvoip.com> wrote: > > I have mixed feelings about giving IXP manager access to the switch. The main reason is that if a member is proactive they can request the second MAC ahead of time, and it can be temporarily added for the transition. As it is most requests are processed in a couple hours. IXP manager is a program that none of the volunteers have a great grasp on, and lack of a timely security update on a portal that is made available to the entire internet in an invitation to get hacked. If it has direct access to the switch things can go wrong quickly. While I am concerned as it is about IXP manager having access to the route servers, they are an extra and not 100% necessary for the IX to function. > > -----Original Message----- > From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Jay Hanke > Sent: Monday, October 31, 2022 9:05 AM > To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET > Subject: Re: [MICE-DISCUSS] Proposal: MICE Static MAC address Only > > The error disable takes 1-2 seconds to clamp down and disable the > port. This results in a short, but severe, broadcast storm. > > The churn rate on members changing mac addresses is fairly low. > > I agree that we should automate the process so it can be done via > self-service by the members. > > On Mon, Oct 31, 2022 at 8:59 AM Steve Howard <showard(a)paulbunyan.net> wrote: > > > > > > Below is a message indicating the current policy as decided by the MICE Board in December 2019 is that the MAC address limit is 1. > > > > > > The subject of Jay's message indicates "Static" MAC, but, the text seems to indicate a limit of 1. I like the 1 address limit, but, am not a fan of Static. Was the current issue caused by somebody not having a proper MAC address limit installed? > > > > > > On 12/13/19 18:00, Richard Laager wrote: > > > > The board has approved the change in MAC address limit from 5 to 1. > > > > NOTE: This is orthogonal to the dedicated remote switch question, which > > is still pending. Non-dedicated remote switches can, for example, > > enforce this per-VLAN. > > > > We have already confirmed that most participants are using only 1 MAC > > address, and some that were using more than 1 have addressed that after > > being contacted. As discussed below, this change will be rolled out in a > > way to keep everyone working, by grandfathering if necessary. > > > > On 10/28/19 3:25 PM, Richard Laager wrote: > > > > We have discussed this a bit in the past, and this came up at the last > > UG. I'm looking for feedback from the group before formally proposing > > this to the board for a vote. > > > > I first brought this up to the technical committee, CCing the board. I > > have heard no objections. Jeremy is "in favor of it 100%". > > > > Based on our quick conversation after the UG, I _think_ Anthony is in > > favor as well; we both made notes to follow up on this. > > > > ----- > > > > I am proposing that MICE change its MAC address limit from 5 MACs per > > port to 1 MAC per port. 1 MAC address is enough for normal scenarios and > > this would further limit the potential for problems on the fabric. This > > restriction is one that SIX and AMS-IX both have, with the latter being > > known for their excellent configuration guide for participants: > > https://www.ams-ix.net/ams/documentation/config-guide > > > > To be clear, this is still only for end ports, not ports facing remote > > switches, of course. The remote switches are responsible for enforcing > > the current MAC limit, and would be responsible for changing those > > settings as described here. > > > > If someone is swapping their MICE-facing router, the resulting port flap > > on the MICE/remote switch will clear the limit anyway, if they are > > directly connected. If they are unable to flap the port (e.g. because > > they have someone else's layer 2 gear in the middle), they will have to > > either work with that carrier or their switch operator (MICE/remote) to > > flap the port. This is a non-zero burden, but I think it's pretty > > minimal in practice. Additionally, if someone plans to make an equipment > > swap, we would increase the limit in advance, temporarily, upon request. > > > > For reference, SIX goes further and locks you to a particular MAC > > address in an ACL, so you have to contact them to change your MAC. I am > > not proposing that. Still, I've been through that a couple of times, and > > even that isn't really too big of a deal. So I don't think the change to > > a limit of 1 MAC at MICE will be particularly onerous. > > > > There are a couple of participants who have two IP address sets (one > > IPv4 and one IPv6) on the same port. For those ports, the limit > > obviously must be at least 2 MACs, but I propose 3 MACs as the limit. If > > they're using two IPs, they probably are doing so to get some redundancy > > out of it. For this particular use case, needing to flap the port > > defeats that redundancy goal, as it breaks the other router. Having a > > MAC limit of 3 would allow them to swap a device without needing to flap > > the port. We would grandfather these setups until they are looking for a > > port-type upgrade; at that point they would need to get to one IP > > address set per port (by splitting into two ports or reducing to one IP > > address set) or request an exception as outlined below. The exception > > process would allow us to better evaluate this use case at that time. > > > > There are some members who have multiple MACs showing up now, despite > > using only one device. For those, I am proposing we set the MAC limit to > > however many MACs are currently showing up on their port. That is, they > > would be grandfathered at their current situation for now. At the time > > of this change, we would encourage them to investigate and fix the > > multiple MACs issue at their leisure. In the future, if they are doing a > > equipment swap (especially like-for-like), we would again encourage them > > to address it, but would otherwise leave their higher limit in place. If > > they do a port-type upgrade, we would not bring the grandfathering over. > > If they needed, they could ask for an official exception at that time. > > > > There may be cases where people cannot reasonably fix their routers to > > use only 1 MAC, other scenarios I haven't considered, or things that may > > come up in the future. For that reason, I think we should allow for the > > possibility of exceptions upon request to the board. In practice, I > > would expect the board would confer with the technical committee and/or > > the technical committee would be bringing these to the board anyway. The > > goal is to strike a reasonable balance between protecting the fabric > > without preventing networks from connecting. > > > > > > > > > > On 10/31/22 08:27, Jay Hanke wrote: > > > > I propose we change the MICE required port security settings to be > > hard filtered for a single static mac address per port. > > > > Over the years, we've had several incidents where floods have occurred > > prior to port being error-disabled during a looping event. > > > > > > > > ________________________________ > > > > To unsubscribe from the MICE-DISCUSS list, click the following link: > > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 > > > > -- > Jay Hanke, President > South Front Networks > jayhanke(a)southfront.io > Phone 612-204-0000 -- Jay Hanke, President South Front Networks jayhanke(a)southfront.io Phone 612-204-0000

Oct. 31, 2022

Re: Proposal: MICE Static MAC address Only

by Jeremy Lumby

I think the best way to phrase it is “static MAC ACL” applied to all member ports. There could be reasons to do static MAC entries in the forwarding table (so traffic does not flood when a large link drops) but we are not talking about that here/yet. From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Steve Howard Sent: Monday, October 31, 2022 8:59 AM To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Proposal: MICE Static MAC address Only Below is a message indicating the current policy as decided by the MICE Board in December 2019 is that the MAC address limit is 1. The subject of Jay's message indicates "Static" MAC, but, the text seems to indicate a limit of 1. I like the 1 address limit, but, am not a fan of Static. Was the current issue caused by somebody not having a proper MAC address limit installed? On 12/13/19 18:00, Richard Laager wrote: The board has approved the change in MAC address limit from 5 to 1. NOTE: This is orthogonal to the dedicated remote switch question, whichis still pending. Non-dedicated remote switches can, for example,enforce this per-VLAN. We have already confirmed that most participants are using only 1 MACaddress, and some that were using more than 1 have addressed that afterbeing contacted. As discussed below, this change will be rolled out in away to keep everyone working, by grandfathering if necessary. On 10/28/19 3:25 PM, Richard Laager wrote: We have discussed this a bit in the past, and this came up at the lastUG. I'm looking for feedback from the group before formally proposingthis to the board for a vote. I first brought this up to the technical committee, CCing the board. Ihave heard no objections. Jeremy is "in favor of it 100%". Based on our quick conversation after the UG, I _think_ Anthony is infavor as well; we both made notes to follow up on this. ----- I am proposing that MICE change its MAC address limit from 5 MACs perport to 1 MAC per port. 1 MAC address is enough for normal scenarios andthis would further limit the potential for problems on the fabric. Thisrestriction is one that SIX and AMS-IX both have, with the latter beingknown for their excellent configuration guide for participants:https://www.ams-ix.net/ams/documentation/config-guide To be clear, this is still only for end ports, not ports facing remoteswitches, of course. The remote switches are responsible for enforcingthe current MAC limit, and would be responsible for changing thosesettings as described here. If someone is swapping their MICE-facing router, the resulting port flapon the MICE/remote switch will clear the limit anyway, if they aredirectly connected. If they are unable to flap the port (e.g. becausethey have someone else's layer 2 gear in the middle), they will have toeither work with that carrier or their switch operator (MICE/remote) toflap the port. This is a non-zero burden, but I think it's prettyminimal in practice. Additionally, if someone plans to make an equipmentswap, we would increase the limit in advance, temporarily, upon request. For reference, SIX goes further and locks you to a particular MACaddress in an ACL, so you have to contact them to change your MAC. I amnot proposing that. Still, I've been through that a couple of times, andeven that isn't really too big of a deal. So I don't think the change toa limit of 1 MAC at MICE will be particularly onerous. There are a couple of participants who have two IP address sets (oneIPv4 and one IPv6) on the same port. For those ports, the limitobviously must be at least 2 MACs, but I propose 3 MACs as the limit. Ifthey're using two IPs, they probably are doing so to get some redundancyout of it. For this particular use case, needing to flap the portdefeats that redundancy goal, as it breaks the other router. Having aMAC limit of 3 would allow them to swap a device without needing to flapthe port. We would grandfather these setups until they are looking for aport-type upgrade; at that point they would need to get to one IPaddress set per port (by splitting into two ports or reducing to one IPaddress set) or request an exception as outlined below. The exceptionprocess would allow us to better evaluate this use case at that time. There are some members who have multiple MACs showing up now, despiteusing only one device. For those, I am proposing we set the MAC limit tohowever many MACs are currently showing up on their port. That is, theywould be grandfathered at their current situation for now. At the timeof this change, we would encourage them to investigate and fix themultiple MACs issue at their leisure. In the future, if they are doing aequipment swap (especially like-for-like), we would again encourage themto address it, but would otherwise leave their higher limit in place. Ifthey do a port-type upgrade, we would not bring the grandfathering over.If they needed, they could ask for an official exception at that time. There may be cases where people cannot reasonably fix their routers touse only 1 MAC, other scenarios I haven't considered, or things that maycome up in the future. For that reason, I think we should allow for thepossibility of exceptions upon request to the board. In practice, Iwould expect the board would confer with the technical committee and/orthe technical committee would be bringing these to the board anyway. Thegoal is to strike a reasonable balance between protecting the fabricwithout preventing networks from connecting. On 10/31/22 08:27, Jay Hanke wrote: I propose we change the MICE required port security settings to behard filtered for a single static mac address per port. Over the years, we've had several incidents where floods have occurredprior to port being error-disabled during a looping event. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

Oct. 31, 2022