- mice-discuss - lists.micemn.net

Re: Please Welcome Radio Link Internet to MICE

by Doug McIntyre

On Mon, Mar 17, 2014 at 08:01:41PM -0500, Jay Hanke wrote: > main switch: xe-0/0/17 > Radio Link Internet > AS: 53301 > 206.108.255.31/24 > 2001:504:27:0:0:d035::1/64 > daniel.petsinger(a)radiolinkinternet.com > > Prefix Limit 25 > > They will be providing an optic for their 10G fiber connection. > > Route-server and defaulting graphing please. I've got the route-servers setup for them, ready to go when they are (and when I've tested the brandnew -01 well enough to haul it back out to Cologix). Welcome aboard! -- Doug McIntyre <merlyn(a)iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 19, 2014

Akamai (20940) < > MICE IXP Upgrade to 2x10G

by Hannigan, Martin

Everyone, All paperwork has been completed and Akamai will be upgrading to 20G. FYI. Best Regards, Martin Hannigan Director, Network and Data Center Architecture Akamai Technologies, Inc. ASN 20940 ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 19, 2014

MICE User Group meeting next Wednesday

by Mike Hemphill

All, Just a quick meeting reminder for our upcoming UG meeting on Wednesday, March 26th at 4:00 PM in the 511 Building cafeteria. Please send Shaun, Reid and me any agenda items you may have or any topics you want to be sure are covered. We currently do not have a sponsor for food & beverages, if your company is willing to step up, please let us know. Thanks! Mike Mike Hemphill General Manager | *Cologix, Inc.* <http://www.cologix.com/> 511 11th Ave S, Suite 450 | Minneapolis, MN 55415 P: 1+612.333.1922 | M: 1+612.812.5242 mike.hemphill(a)cologix.com ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 19, 2014

Please Welcome Radio Link Internet to MICE

by Jay Hanke

main switch: xe-0/0/17 Radio Link Internet AS: 53301 206.108.255.31/24 2001:504:27:0:0:d035::1/64 daniel.petsinger(a)radiolinkinternet.com Prefix Limit 25 They will be providing an optic for their 10G fiber connection. Route-server and defaulting graphing please. Thanks! -- Jay Hanke CTO, CCIE #19093 Mankato Networks LLC PO Box 54 619 S Front St Mankato, MN 56001-3838 Google 530-618-2398 jayhanke(a)mankatonetworks.net http://www.mankatonetworks.com http://www.neutralpath.net ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 18, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Nevin Lyne

Doug, Thanks for the update, and the work on this. Glad I have a lot of spare 1950 era gear on the stack for re-use. -Nevin On Monday, March 17, 2014 3:10pm, "Doug McIntyre" <merlyn(a)IPHOUSE.NET> said: > On Fri, Mar 14, 2014 at 06:29:44PM -0500, Doug McIntyre wrote: >> On Thu, Mar 13, 2014 at 04:03:03PM -0500, Doug McIntyre wrote: >> > On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: >> > > We've noticed this being unresponsive several times this evening. >> > >> > This is an actual hardware motherboard failure. >> >> I brought out the new machine this early evening, and it came up >> without a hitch. > > > Just an update on this, as I'm sure people have noticed that this > machine died horribly over the weekend, without any remedy by remote > reboot. > > Same sort of issues, with PCI bus errors, and potential disk failure > log entries. > > I went and visited the machine before lunch, brought another good disk > controller, and now neither controller card will work in the slot, and > PCI errors on either controller. > > I can only assume that the original disk controller (which was moved over > due to the existing setup requiring that type of controller) is poison > and destroying PCIe slots and thus systems that it plugs into. > > So, I think we'll try yet again, and burn in a system for a few days > first before hauling it back out to MICE again. > > -- > Doug McIntyre <merlyn(a)iphouse.net> > ~.~ ipHouse ~.~ > Network Engineer/Provisioning/Jack of all Trades > > ######################################################################## > > 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

March 17, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Doug McIntyre

On Fri, Mar 14, 2014 at 06:29:44PM -0500, Doug McIntyre wrote: > On Thu, Mar 13, 2014 at 04:03:03PM -0500, Doug McIntyre wrote: > > On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: > > > We've noticed this being unresponsive several times this evening. > > > > This is an actual hardware motherboard failure. > > I brought out the new machine this early evening, and it came up > without a hitch. Just an update on this, as I'm sure people have noticed that this machine died horribly over the weekend, without any remedy by remote reboot. Same sort of issues, with PCI bus errors, and potential disk failure log entries. I went and visited the machine before lunch, brought another good disk controller, and now neither controller card will work in the slot, and PCI errors on either controller. I can only assume that the original disk controller (which was moved over due to the existing setup requiring that type of controller) is poison and destroying PCIe slots and thus systems that it plugs into. So, I think we'll try yet again, and burn in a system for a few days first before hauling it back out to MICE again. -- Doug McIntyre <merlyn(a)iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 17, 2014

NTIA announcement about ICANN transition to independent entity

by Mike O'Connor

hi all, i’m presuming most of you have seen the NTIA announcement by now. since i hang “MICE" on my nametag when i participate in ICANN meetings, i thought i’d push along some observations, and some links to detailed info that may not find its way into the press coverage. the announcement: http://www.ntia.doc.gov/press-release/2014/ntia-announces-intent-transition… WASHINGTON – To support and enhance the multistakeholder model of Internet policymaking and governance, the U.S. Commerce Department’s National Telecommunications and Information Administration (NTIA) today announces its intent to transition key Internet domain name functions to the global multistakeholder community. As the first step, NTIA is asking the Internet Corporation for Assigned Names and Numbers (ICANN) to convene global stakeholders to develop a proposal to transition the current role played by NTIA in the coordination of the Internet’s domain name system (DNS). the FAQ on ICANN’s site; http://www.icann.org/en/about/agreements/iana/functions-transfer-faqs-14mar… want to know a LOT about these issues (like PhD level)? Here’s Samantha Dickinson’s summary of a boatload of submissions from all over the place that’s going into the upcoming Net Mundial meeting in Brazil. read all these (like i am right now) and you’ll know where the proposals lie. http://linguasynaptica.com/netmundial-submissions-on-iana/ my view is mostly “keep calm and peer on.” i’m hoping the transition could be the same level of non-event that taking down the NSF backbone network was. on the other hand, we don’t know how this is actually going to turn out until it’s done. the fearmongers in Washington aren’t being helpful. my view “from the bottom of the bottom-up process” is that i’ll be an interested observer and i’ll probably continue to put most of my energy into the day-to-day policy stuff. that’s the stuff that makes ICANN legitimate. i was on a conference call this morning where the CEO (Fadi Chehadi) and the Chair of the Board (Steve Crocker - author of RFC 0001) talked to community members about what’s going on. the points i raised were mostly about bandwidth (the community is already really overloaded) and process (a good project plan would be nice). happy to try to answer questions, although i’m still coming up to speed myself — this came as a surprise to almost everybody. i’m headed to the ICANN meeting in Singapore (leaving this Wednesday, back the following Friday). the outlines will be clearer after that. mikey PHONE: 651-647-6109, FAX: 866-280-2356, WEB: www.haven2.com, HANDLE: OConnorStP (ID for Twitter, Facebook, LinkedIn, etc.) ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 15, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Mike Hemphill

Doug, Thank you! You are a great asset & supporter of the exchange. You are appreciated! Mike On Mar 14, 2014 6:29 PM, "Doug McIntyre" <merlyn(a)iphouse.net> wrote: > On Thu, Mar 13, 2014 at 04:03:03PM -0500, Doug McIntyre wrote: > > On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: > > > We've noticed this being unresponsive several times this evening. > > > > This is an actual hardware motherboard failure. > > I brought out the new machine this early evening, and it came up > without a hitch. > > I see most people's BGP sessions up, we have 923 prefixes announced into > -01 > now, and things are looking good. > > This should complete out this repair now. > > -- > Doug McIntyre <merlyn(a)iphouse.net> > ~.~ ipHouse ~.~ > Network Engineer/Provisioning/Jack of all Trades > > ######################################################################## > > 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

March 15, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Nevin Lyne

Thanks, Doug! On Friday, March 14, 2014 6:29pm, "Doug McIntyre" <merlyn(a)IPHOUSE.NET> said: > On Thu, Mar 13, 2014 at 04:03:03PM -0500, Doug McIntyre wrote: >> On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: >> > We've noticed this being unresponsive several times this evening. >> >> This is an actual hardware motherboard failure. > > I brought out the new machine this early evening, and it came up > without a hitch. > > I see most people's BGP sessions up, we have 923 prefixes announced into -01 > now, and things are looking good. > > This should complete out this repair now. > > -- > Doug McIntyre <merlyn(a)iphouse.net> > ~.~ ipHouse ~.~ > Network Engineer/Provisioning/Jack of all Trades > > ######################################################################## > > 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

March 14, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Doug McIntyre

On Thu, Mar 13, 2014 at 04:03:03PM -0500, Doug McIntyre wrote: > On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: > > We've noticed this being unresponsive several times this evening. > > This is an actual hardware motherboard failure. I brought out the new machine this early evening, and it came up without a hitch. I see most people's BGP sessions up, we have 923 prefixes announced into -01 now, and things are looking good. This should complete out this repair now. -- Doug McIntyre <merlyn(a)iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 14, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Nevin Lyne

Have about 2 dozen of them that have been retired, and in IPHouse's data center so easy to get to Doug. -Nevin On Thursday, March 13, 2014 7:44pm, "Brady Kittel" <bkittel(a)GMAIL.COM> said: > We probably have a couple 1950s lying around too if needed. Drop line if > you need me to take a look. > On Mar 13, 2014 4:03 PM, "Doug McIntyre" <merlyn(a)iphouse.net> wrote: > >> On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: >> > We've noticed this being unresponsive several times this evening. >> > >> > I'm into the machine, but it is behaving badly, and from past >> > experience, (and not direct evidence since it is remote and I can't >> > get the relavent logs), it feels like a hard drive failing on this >> > system. >> >> >> This is an actual hardware motherboard failure. >> Probably only the 2nd I've seen in my life of using Dell hardware. >> >> Lots of errors around the PCI bus, even with no cards installed. >> Not drives, memory, DRAC card. Nothing left to even try. >> >> Nevin@Arcus has already volunteered another PE1950 chassis if this was the >> case. We just have to coordinate getting it and swapping it in. >> >> The 2nd route-reflector should just keep up with our needs until this >> hardware can be fixed up. >> >> >> -- >> Doug McIntyre <merlyn(a)iphouse.net> >> ~.~ ipHouse ~.~ >> Network Engineer/Provisioning/Jack of all Trades >> >> ######################################################################## >> >> 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

March 14, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Brady Kittel

We probably have a couple 1950s lying around too if needed. Drop line if you need me to take a look. On Mar 13, 2014 4:03 PM, "Doug McIntyre" <merlyn(a)iphouse.net> wrote: > On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: > > We've noticed this being unresponsive several times this evening. > > > > I'm into the machine, but it is behaving badly, and from past > > experience, (and not direct evidence since it is remote and I can't > > get the relavent logs), it feels like a hard drive failing on this > > system. > > > This is an actual hardware motherboard failure. > Probably only the 2nd I've seen in my life of using Dell hardware. > > Lots of errors around the PCI bus, even with no cards installed. > Not drives, memory, DRAC card. Nothing left to even try. > > Nevin@Arcus has already volunteered another PE1950 chassis if this was the > case. We just have to coordinate getting it and swapping it in. > > The 2nd route-reflector should just keep up with our needs until this > hardware can be fixed up. > > > -- > Doug McIntyre <merlyn(a)iphouse.net> > ~.~ ipHouse ~.~ > Network Engineer/Provisioning/Jack of all Trades > > ######################################################################## > > 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

March 14, 2014

Re: DDoS Attacks

by David Farmer

On 3/13/14, 14:04 , Richard Laager wrote: > Last night, we got hit by a ~3 Gbps DDoS attack. It's been a while since > this has happened to us, so I'd like to make sure I'm still up on the > state of the art. > > Is there anything more to be done than the following? > 1. Identify the victim. > 2. Null route the victim. > 3. Propagate the null route to your upstreams (via BGP, if > supported, otherwise a phone call to their NOC). > 4. Move the victim to a new IP. 3a. Notify your (MICE) peer if they are participating in the attack. We for one want to respond and deal with issues to the extent possible. FYI, specifically for us abuse(a)umn.edu, sometimes we are not the quickest, but we really don't just ignore issues reported to us. > To avoid participating in at least some classes of DDoS attacks, we: > * long ago implemented uRPF (and/or similar ACLs) to block spoofed > outbound packets, as recommended by BCP 38 (RFC 2827). Yes, please implement BGP38 to the extent possible, we do. Yes, not everyone realistically can, but if everyone that can did, the world would have a better Internet. > * ensured our NTP servers (and any NTP servers of our customers) > are not responding to monlist queries. The openntpproject.org > website is useful here. They list vulnerable NTP servers by IP > range, or you can get all NTP servers by AS (replace YOUR_AS > with your AS, and optionally, add &csv=1) and then query with > ntpdc -n -c monlist IP: We addressed this in January for our authoritative public NTP servers. But, there were a lot of other NTP servers on campus, this took us a while to mitigate. We think we have this squelched now, but let us know if you see moles that need whacking, especially if the mole is bothering you. > http://openntpproject.org/searchby-asn.cgi?search_asn=YOUR_AS > * just this week started addressing customers with open DNS > resolvers, which can also be used in amplification attacks: > http://openresolverproject.org/searchby-asn.cgi?search_asn=YOUR_AS A continual battle for years, if not a decade. Again, let us know if you see moles that need whacking, especially if the mole is bothering you. :( > Is there anything else we should be doing? If you don't already, block other UDP ports that probably shouldn't talk to the Internet because they are also targets of UDP amplification attacks too, SNMP, Chargen, SSDP, QOTD, etc... https://www.us-cert.gov/ncas/alerts/TA14-017A -- ================================================ David Farmer Email: farmer(a)umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================ ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

Re: DDoS Attacks

by Frank Bulk

I've documented over 60 attacks to customers in our network in the last 22 months. Some are small, just 200 Mbps, the largest has been 2.6 Gbps. In 98% of all cases the attack is just six to eight minutes long, so by the time I respond to any alarms it's already over. So there's one more thing you can do. 5. Increase your uplink so that the small attacks end up affecting only the subset of access technology that's part of the targeted customer. Frank -----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Richard Laager Sent: Thursday, March 13, 2014 2:04 PM To: MICE-DISCUSS(a)LISTS.IPHOUSE.NET Subject: [MICE-DISCUSS] DDoS Attacks Last night, we got hit by a ~3 Gbps DDoS attack. It's been a while since this has happened to us, so I'd like to make sure I'm still up on the state of the art. Is there anything more to be done than the following? 1. Identify the victim. 2. Null route the victim. 3. Propagate the null route to your upstreams (via BGP, if supported, otherwise a phone call to their NOC). 4. Move the victim to a new IP. To avoid participating in at least some classes of DDoS attacks, we: * long ago implemented uRPF (and/or similar ACLs) to block spoofed outbound packets, as recommended by BCP 38 (RFC 2827). * ensured our NTP servers (and any NTP servers of our customers) are not responding to monlist queries. The openntpproject.org website is useful here. They list vulnerable NTP servers by IP range, or you can get all NTP servers by AS (replace YOUR_AS with your AS, and optionally, add &csv=1) and then query with ntpdc -n -c monlist IP: http://openntpproject.org/searchby-asn.cgi?search_asn=YOUR_AS * just this week started addressing customers with open DNS resolvers, which can also be used in amplification attacks: http://openresolverproject.org/searchby-asn.cgi?search_asn=YOUR_AS Is there anything else we should be doing? -- Richard ######################################################################## 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

March 13, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Michael Horwath

On Mar 13, 2014, at 4:03 PM, Doug McIntyre <merlyn(a)iphouse.net> wrote: > Nevin@Arcus has already volunteered another PE1950 chassis if this was the > case. We just have to coordinate getting it and swapping it in. That Nevin; he rocks! I nominate him as MICEMN mascot for 2014. -- Mike Horwath, reachable via drechsau(a)Geeks.ORG ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Doug McIntyre

On Thu, Mar 13, 2014 at 12:19:08AM -0500, Doug McIntyre wrote: > We've noticed this being unresponsive several times this evening. > > I'm into the machine, but it is behaving badly, and from past > experience, (and not direct evidence since it is remote and I can't > get the relavent logs), it feels like a hard drive failing on this > system. This is an actual hardware motherboard failure. Probably only the 2nd I've seen in my life of using Dell hardware. Lots of errors around the PCI bus, even with no cards installed. Not drives, memory, DRAC card. Nothing left to even try. Nevin@Arcus has already volunteered another PE1950 chassis if this was the case. We just have to coordinate getting it and swapping it in. The 2nd route-reflector should just keep up with our needs until this hardware can be fixed up. -- Doug McIntyre <merlyn(a)iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

Re: DDoS Attacks

by Mike O'Connor

hi all, there’s a recent report from ICANN’s SSAC (Security and Stability Advisory Committee) on DDoS attacks using DNS infrastructure that may be useful https://www.icann.org/en/groups/ssac/documents/sac-065-en.pdf here’s the Executive Summary to give you a sense of what’s in it. Executive Summary This document is intended for the Internet technical community, particularly authoritative and recursive Domain Name System (DNS) operators, network operators, the Internet Corporation for Assigned Names and Numbers (ICANN), and policy makers. It explores several unresolved critical design and deployment issues that have enabled increasingly large and severe Distributed Denial of Service (DDoS) attacks using the DNS. While DDoS attacks can exploit multiple characteristics of network infrastructure and operations, the prevalence and criticality of the DNS means that securing it is both challenging and urgent. These unresolved DNS issues and related DDoS attacks pose a real and present danger to the security and stability of the Internet. The first recommendation below is made to ICANN, while others are made to operators of Internet infrastructure and manufacturers. While in many instances they reflect actions not under ICANN’s control and actors not necessarily within ICANN’s usual community, they are meant to address the overall responsibilities of the multi-stakeholder community and encourage ICANN to take action where it is relevant to do so. In particular, this means ICANN should be looking for ways to increase the effectiveness of steps already being taken against DNS abuse and promoting the participation of others as well as pursuing the measures suggested here. The Security and Stability Advisory Committee (SSAC) strongly recommends that all types of DNS operators and network operators take immediate steps to mitigate the design and deployment issues that make large scale DDoS attacks possible. Specifically, the SSAC strongly recommends that: • ICANN should help facilitate an Internet-wide community effort to reduce the number of open resolvers and networks that allow network spoofing. This effort should involve measurement efforts and outreach. • All network operators should take immediate steps to prevent network address spoofing. • Recursive DNS server operators should take immediate steps to secure open recursive DNS servers. • Authoritative DNS server operators should support efforts to investigate authoritative response rate limiting. • DNS server operators should put in place operational processes to ensure that their DNS software is regularly updated and communicate with their software vendors to keep abreast of the latest developments. • Manufacturers and/or configurators of customer premise networking equipment, including home networking equipment, should take immediate steps to secure these devices and ensure that they are field upgradable when new software is available to fix security vulnerabilities, and aggressively replace the installed base of non-upgradeable devices with upgradeable devices. On Mar 13, 2014, at 2:04 PM, Richard Laager <rlaager(a)WIKTEL.COM> wrote: > Last night, we got hit by a ~3 Gbps DDoS attack. It's been a while since > this has happened to us, so I'd like to make sure I'm still up on the > state of the art. > > Is there anything more to be done than the following? > 1. Identify the victim. > 2. Null route the victim. > 3. Propagate the null route to your upstreams (via BGP, if > supported, otherwise a phone call to their NOC). > 4. Move the victim to a new IP. > > > To avoid participating in at least some classes of DDoS attacks, we: > * long ago implemented uRPF (and/or similar ACLs) to block spoofed > outbound packets, as recommended by BCP 38 (RFC 2827). > * ensured our NTP servers (and any NTP servers of our customers) > are not responding to monlist queries. The openntpproject.org > website is useful here. They list vulnerable NTP servers by IP > range, or you can get all NTP servers by AS (replace YOUR_AS > with your AS, and optionally, add &csv=1) and then query with > ntpdc -n -c monlist IP: > http://openntpproject.org/searchby-asn.cgi?search_asn=YOUR_AS > * just this week started addressing customers with open DNS > resolvers, which can also be used in amplification attacks: > http://openresolverproject.org/searchby-asn.cgi?search_asn=YOUR_AS > > Is there anything else we should be doing? > > -- > Richard > > ######################################################################## > > To unsubscribe from the MICE-DISCUSS list, click the following link: > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 PHONE: 651-647-6109, FAX: 866-280-2356, WEB: www.haven2.com, HANDLE: OConnorStP (ID for Twitter, Facebook, LinkedIn, etc.) ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

DDoS Attacks

by Richard Laager

Last night, we got hit by a ~3 Gbps DDoS attack. It's been a while since this has happened to us, so I'd like to make sure I'm still up on the state of the art. Is there anything more to be done than the following? 1. Identify the victim. 2. Null route the victim. 3. Propagate the null route to your upstreams (via BGP, if supported, otherwise a phone call to their NOC). 4. Move the victim to a new IP. To avoid participating in at least some classes of DDoS attacks, we: * long ago implemented uRPF (and/or similar ACLs) to block spoofed outbound packets, as recommended by BCP 38 (RFC 2827). * ensured our NTP servers (and any NTP servers of our customers) are not responding to monlist queries. The openntpproject.org website is useful here. They list vulnerable NTP servers by IP range, or you can get all NTP servers by AS (replace YOUR_AS with your AS, and optionally, add &csv=1) and then query with ntpdc -n -c monlist IP: http://openntpproject.org/searchby-asn.cgi?search_asn=YOUR_AS * just this week started addressing customers with open DNS resolvers, which can also be used in amplification attacks: http://openresolverproject.org/searchby-asn.cgi?search_asn=YOUR_AS Is there anything else we should be doing? -- Richard ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

Re: bgp flapping to 206.108.255.1

by Frank DiGravina

Hello, I am shutting our peering to 206.108.255.1 until I see notice that that your route-reflector issue has been addressed. Thanks, Frank -- Frank DiGravina University of Minnesota Networking & Telecommunications Network Operations Phone: 612-626-9074 Cell: 612-386-0449 E-Mail: digravin(a)umn.edu Helpline: 612-301-HELP ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

Re: micemn-01.micemn.net route reflector at 206.108.255.1

by Brady Kittel

I've also seen issues with failing drives not flagging because of a buggy firmware version on the raid controller or drive at least a few times on Dell servers around that vintage. On Mar 13, 2014 12:19 AM, "Doug McIntyre" <merlyn(a)iphouse.net> wrote: > We've noticed this being unresponsive several times this evening. > > I'm into the machine, but it is behaving badly, and from past > experience, (and not direct evidence since it is remote and I can't > get the relavent logs), it feels like a hard drive failing on this > system. > > Ie. it may run some of its process out of memory for some functions, > but other functions are probably locking up and not behaving correctly. > Even RAIDed as it is. > > Probably the best course of action is to bring out a fresh drive to it > and swap whatever drive is failing and let the mirror rebuild. I think > we've got some of the 80GB WD800AAJS that is part of the mirror on the > other route reflector around somewhere... > > -- > Doug McIntyre <merlyn(a)iphouse.net> > ~.~ ipHouse ~.~ > Network Engineer/Provisioning/Jack of all Trades > > ######################################################################## > > 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

March 13, 2014

micemn-01.micemn.net route reflector at 206.108.255.1

by Doug McIntyre

We've noticed this being unresponsive several times this evening. I'm into the machine, but it is behaving badly, and from past experience, (and not direct evidence since it is remote and I can't get the relavent logs), it feels like a hard drive failing on this system. Ie. it may run some of its process out of memory for some functions, but other functions are probably locking up and not behaving correctly. Even RAIDed as it is. Probably the best course of action is to bring out a fresh drive to it and swap whatever drive is failing and let the mirror rebuild. I think we've got some of the 80GB WD800AAJS that is part of the mirror on the other route reflector around somewhere... -- Doug McIntyre <merlyn(a)iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 13, 2014

XO issues Sunday morning?

by Frank Bulk

This isn't a MICE-related question, but since so many MICE members are in the Minnesota area I thought I would ask. Sunday morning from 1:11 am to about 4:20 am we saw an unusually high number of websites time out over IPv6. We monitor 50 IPv6-enabled web sites over IPv6, performing both an ICMP check (if possible) and then HTTP over IPv6. There were at least 17 sites during that time frame that timed out on our monitoring systems' HTTP GET request. One pattern emerged -- 14 of them traceroute through an XO link that goes through Minneapolis. Here's a sample: traceroute to ipv6.test.Level3.com (2001:1900:2018:3000::105) from 2607:fe28:0:1000::5, port 80, from port 36274, 30 hops max, 60 bytes packets 1 router-core.mtcnet.net (2607:fe28:0:1000::1) 0.395 ms 0.325 ms 0.319 ms 2 sxct.sxcy.mtcnet.net (2607:fe28:11:1002::197) 0.188 ms 0.135 ms 0.185 ms 3 v6-premier.sxcy-mlx.fbnt.netins.net (2001:5f8:7f0a:2::1) 1.612 ms 1.545 ms 1.613 ms 4 v6-ins-db1-te-13-2-219.desm.netins.net (2001:5f8:1:1::1) 6.667 ms 6.566 ms 8.002 ms 5 v6-ins-dc1-et-8-2.desm.netins.net (2001:5f8::1:1) 6.624 ms 7.995 ms 6.602 ms 6 2610:18:18b:c000::11 (2610:18:18b:c000::11) 12.108 ms 12.075 ms 12.137 ms 7 mcr1.minneapolis-mn.us.xo.net (2610:18::30b8) 21.431 ms 21.452 ms 21.421 ms 8 vb1710.rar3.chicago-il.us.xo.net (::ffff:216.156.0.169) 30.701 ms 23.639 ms 23.771 ms .... I then looked at the daily traceroutes we have on file for all 50 sites we monitor and it looks like 15 of them go through XO. So over 93% (14 out of 15) of the sites that go through XO timed out during the 1:11 am to 4:20 am time period. Did anyone with XO connectivity see issues Sunday morning? Frank ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 10, 2014

Re: 2 x 10GE Contact?

by Steve Howard

Hi Martin, I just wanted to make sure that somebody followed up with you. Did you get a response to your question? Thanks, Steve On 03/05/2014 02:55 PM, Hannigan, Martin wrote: > Who can I talk to about adding a second 10G? > > Best, > ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 7, 2014

2 x 10GE Contact?

by Hannigan, Martin

Who can I talk to about adding a second 10G? Best, , -M< ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

March 5, 2014

Re: Public Stats Aggregate View

by Michael Horwath

On Feb 26, 2014, at 4:23 PM, Anthony Anderberg <anthonyanderberg(a)NU-TELECOM.NET> wrote: >>> I believe it's donated server space. >> >> When I donated space it was called 'stats.micemn.net'. > > Oh ya, everyone just changed their bookmarks and it never got updated - good catch. > Steve@PBT can you fix that for us? +1 from the peanut gallery! -- Mike Horwath, reachable via drechsau(a)Geeks.ORG ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

Feb. 27, 2014