The CL portal loads for me, and I can log in, but it is slower than usual. Not
seeing traffic issues on our CL circuits.
-mel via cell
On Aug 30, 2020, at 5:23 AM, Drew Weaver via NANOG wrote:
Hello,
Woke up this morning to a bunch of reports of issues with connectivity had to
shut down
Started about 5:05am central, started clearing up for me about 7:15am.
My route from ATT in Chicago is still going through NYC to get back to
Chicago but at least packet loss isn't 70-100% anymore.
I also tried turning down sessions and still was seeing stale
announcements on other LGs.
That site seems to be just for their cloud products, is there one of these for
their actual network?
From: R. Leigh Hennig
Sent: Sunday, August 30, 2020 8:54 AM
To: Drew Weaver ; nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
Global impact with issues reported by Fastly,
>
> They’re not reachable so who knows if they’re even working on it.
>
Gonna go out on a limb here and assume that a lot of phones were ringing
and people are in fact working on whatever it is. :)
On Sun, Aug 30, 2020 at 8:34 AM David Hubbard
wrote:
> Same. Also, as reported on outages list,
Yikes, they really need to get their reflectors fixed so that we can dis-engage
from them.
-Original Message-
From: NANOG On Behalf Of Ben
Russell
Sent: Sunday, August 30, 2020 9:14 AM
To: 'nanog@nanog.org'
Subject: RE: Centurylink having a bad morning?
One of my legacy L3 peers is
In many situations it will make sense to keep the CPE provided by the ISP in a
configuration equivalent to a "bridge" (I know is not a bridge, I'm trying to
use a single word to describe it), so it runs things like the NAT for IPv4 and
the CLAT for IPv6, even DHCP, or RA, etc. It all depends on
Looks like this is causing problems with Meraki Cloud connectivity since level3
is a transit provider for Meraki.
From: NANOG on behalf of Mel
Beckman
Sent: Sunday, August 30, 2020, 08:47
To: Drew Weaver
Cc: nanog@nanog.org
Subject: Re: Centurylink having a
Reporting from Europe, any IP with them in the path is unreachable from various
providers. I guess they wanted to try IPv6-only.. :P IPv6 is fine, working
fine, IPv4 not at all..
Antonis
> On 30 Aug 2020, at 14:58, Tomas Lynch wrote:
>
> Flapping in Miami, Dallas, Atlanta, Los Angeles,
I am seeing some odd traffic deflections in the EU via 3356, but nothing in
the US so far.
Some scattershot oddball reports landing in our NOC, but nothing
conclusive.
On Sun, Aug 30, 2020 at 8:41 AM Mel Beckman wrote:
> The CL portal loads for me, and I can log in, but it is slower than
One of my legacy L3 peers is only advertising 498174 prefixes. Very strange.
Ben Russell
-Original Message-
From: NANOG On Behalf Of Drew
Weaver
Sent: Sunday, August 30, 2020 7:45 AM
To: 'Mel Beckman'
Cc: 'nanog@nanog.org'
Subject: RE: Centurylink having a bad morning?
Can't
I just brought one of my sessions back up to attempt to avoid the blackholing,
should be full feed, getting all of 850 v4 routes and 106 v6.
From: NANOG on behalf
of Tomas Lynch
Date: Sunday, August 30, 2020 at 9:41 AM
To: Drew Weaver
Cc: "nanog@nanog.org"
Subject: Re: Centurylink having a
Well, When I tried calling I got a fast busy, so that's nice.
On Sun, Aug 30, 2020 at 8:33 AM David Hubbard
wrote:
> Same. Also, as reported on outages list, what’s even worse is that they
> appear to be continuing to propagate advertisements from circuits whose
> sessions have been turned
I have checked all of our circuits in the LA area and are seeing no BGP issues,
but we don’t provide transit for CL, only use them for DIA. Traffic flows
appear normal, but early Sunday morning is normally the lowest time of the
week.
-mel beckman
> On Aug 30, 2020, at 5:29 AM, Mel Beckman
Flapping in Miami, Dallas, Atlanta, Los Angeles, Seattle and San Jose. It
is also affecting some data centers in Europe too. but haven't seen flaps
there, just suboptimal routing.
On Sun, Aug 30, 2020 at 8:53 AM Drew Weaver wrote:
> Saw the flapping in Cleveland but not in Cincinnatti or
Now if you call into CL you get a message stating their technicians are
working on an ip outage.
On Sun, Aug 30, 2020 at 6:56 AM Chase Christian wrote:
> Multiple BGP sessions with Level3 (DIA) started flapping at approx 03:00
> Pacific:
>
> Aug 30 03:05:13 rtr02 Rib: %BGP-3-NOTIFICATION: sent
Yeah, I am still seeing them announce our Ips even though we’ve shut down our
sessions with them.
From: NANOG On Behalf Of David
Hubbard
Sent: Sunday, August 30, 2020 8:28 AM
To: nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
Same. Also, as reported on outages list, what’s
Finally got through on their support line and spoke to level1. The only
thing the tech could say was it was an issue with BGP route reflectors and
it started about 3am(pacific). They were still trying to isolate the issue.
I've tried failing over my circuits and no go, the traffic just dies as L3
Saw the flapping in Cleveland but not in Cincinnatti or Ashburn…
From: Tomas Lynch
Sent: Sunday, August 30, 2020 8:45 AM
To: Mel Beckman
Cc: Drew Weaver ; nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
BGP sessions randomly flapping or having routing issues in different cities
I don’t think it is anywhere near back to normal but I can get to 2-3 more
sites than I could when I got called in.
From: NANOG On Behalf Of Andy
Brezinsky
Sent: Sunday, August 30, 2020 8:48 AM
To: nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
Started about 5:05am central,
Something just now changed in this situation and now it seems to have gotten
worse.
From: Jason Kuehl
Sent: Sunday, August 30, 2020 10:00 AM
To: Drew Weaver
Cc: R. Leigh Hennig ; nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
People are rebooting ghosting now.
Same. Also, as reported on outages list, what’s even worse is that they appear
to be continuing to propagate advertisements from circuits whose sessions have
been turned down. I validated ours still were via a couple looking glass
portals. Down Detector shows nearly every major service
One anomaly I’ve just notice: we can’t reach 209.244.0.1 (ns1.level3.com) from
anywhere on the Internet, including our CL connections. Traceroute fails at
4.16.30.129 (lag-101.ear2.LosAngeles1.Level3.net) with 65% packet loss:
traceroute to 209.244.0.1 (209.244.0.1), 30 hops max, 56 byte
Their route-servers just can’t keep up.
They are not getting our community-string announcements at all
From: NANOG On Behalf Of
Drew Weaver
Sent: Sunday, August 30, 2020 8:38 AM
To: 'David Hubbard' ; 'nanog@nanog.org'
Subject: RE: Centurylink having a bad morning? [EXTERNAL]
Yeah, I am
Can't get to pretty much anything on the Internet from my Tmobile device... not
really sure whats going on =)
-Original Message-
From: Mel Beckman
Sent: Sunday, August 30, 2020 8:43 AM
To: Drew Weaver
Cc: nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
One anomaly I’ve
We've been on hold for more than an hour trying to get an update.
We see the same behavior where they continue to announce our blocks
despite all the interfaces to them being hard down.
Scott Helms
On Sun, Aug 30, 2020 at 8:58 AM Jason Kuehl wrote:
>
> Well, When I tried calling I got a fast
People are rebooting ghosting now.
[image: image.png]
https://twitter.com/ir_kujoe/status/1300066569645707265
Seeing other reports of this too.
On Sun, Aug 30, 2020 at 9:45 AM Drew Weaver wrote:
> That site seems to be just for their cloud products, is there one of these
> for their actual
Multiple BGP sessions with Level3 (DIA) started flapping at approx 03:00
Pacific:
Aug 30 03:05:13 rtr02 Rib: %BGP-3-NOTIFICATION: sent to neighbor 4.35.X.Y
(AS 3356) 4/0 (Hold Timer Expired Error/Unspecified) 0 bytes
Aug 30 03:05:13 rtr02 Rib: %BGP-5-ADJCHANGE: peer 4.35.X.Y (AS 3356) old
state
I'm over in MA in a CL building, it's very much still broken. I shut down
the interfaces to CL and now just using Comcast.
On Sun, Aug 30, 2020 at 9:20 AM Andy Brezinsky wrote:
> Started about 5:05am central, started clearing up for me about 7:15am.
> My route from ATT in Chicago is still going
Once upon a time, Robert Blayzor said:
> Just to confirm we're seeing this on AS3356 and not AS209, correct?
Correct - we had problems with our 3356 connection but not our 209
connection.
--
Chris Adams
On 30/08/2020 20:08, Baldur Norddahl
wrote:
https://blog.cloudflare.com/analysis-of-todays-centurylink-level-3-outage/
Sounds like Flowspec possibly blocking
tcp/179 might be the cause.
But that is Cloudflare speculation.
On Sun, 30 Aug 2020 at 20:00, Baldur Norddahl wrote:
> Not really the point. BGP is designed such that if I take down the link, the
> prefixes MUST be withdrawn within reasonable time. The self healing aspect of
> the internet entirely depends on this. Clearly they have some kind of system
>
Hello,
Woke up this morning to a bunch of reports of issues with connectivity had to
shut down some Level3/CTL connections to get it to return to normal.
As of right now their support portal won't load:
https://www.centurylink.com/business/login/
Just wondering what others are seeing.
Its dead JIM. I also can't get in starting around 7 est
On Sun, Aug 30, 2020 at 8:19 AM Drew Weaver via NANOG
wrote:
> Hello,
>
>
>
> Woke up this morning to a bunch of reports of issues with connectivity had
> to shut down some Level3/CTL connections to get it to return to normal.
>
>
>
> As
BGP sessions randomly flapping or having routing issues in different cities
since ~5AM EST
On Sun, Aug 30, 2020 at 8:42 AM Mel Beckman wrote:
> The CL portal loads for me, and I can log in, but it is slower than usual.
> Not seeing traffic issues on our CL circuits.
>
> -mel via cell
>
> On Aug
On 8/30/20 8:14 AM, Drew Weaver via NANOG wrote:
> Woke up this morning to a bunch of reports of issues with connectivity
> had to shut down some Level3/CTL connections to get it to return to normal.
Just to confirm we're seeing this on AS3356 and not AS209, correct?
We have links to both and
How is that acceptable behaviour? I shall remember never to make a contract
with these guys until they can prove that they won't advertise my prefixes
after I pull them. Under any circumstances.
søn. 30. aug. 2020 15.14 skrev Joseph Jenkins :
> Finally got through on their support line and spoke
I’m not defending them but I am sure it isn’t intentional.
From: NANOG On Behalf Of
Baldur Norddahl
Sent: Sunday, August 30, 2020 9:28 AM
To: nanog@nanog.org
Subject: Re: Centurylink having a bad morning?
How is that acceptable behaviour? I shall remember never to make a contract
with these
This is on the Legacy Level 3 side which is generally much better and still
largely separate (according to my former co-workers). So say what you want
about CTL but this is an LVLT issue.
- Mike Bolitho
On Sun, Aug 30, 2020, 8:03 AM Ross Tajvar wrote:
> I've never heard a single positive word
I saw customers behind AS209’s ILEC Residential network flapping all
morning. It’s stable now. So LVLT 3356 is definitely feeding parts of 209
now.
On Sun, Aug 30, 2020 at 8:57 AM Ross Tajvar wrote:
> Not being intentional isn't really an excuseOutages are generally not
> intentional but
Once upon a time, Baldur Norddahl said:
> How is that acceptable behaviour? I shall remember never to make a contract
> with these guys until they can prove that they won't advertise my prefixes
> after I pull them. Under any circumstances.
Umm, then I guess you won't sign a contract with
AS3356 is the Level3 internet.
On Sun, Aug 30, 2020 at 8:09 AM Ian Bowers wrote:
> AS3356 is the one I've seen all the chatter about this morning.
>
> On Sun, Aug 30, 2020 at 10:03 AM Robert Blayzor
> wrote:
>
>> On 8/30/20 8:14 AM, Drew Weaver via NANOG wrote:
>> > Woke up this morning to a
Not being intentional isn't really an excuseOutages are generally not
intentional but we still like to use services that stay up most of the time.
On Sun, Aug 30, 2020 at 11:11 AM Drew Weaver wrote:
> I’m not defending them but I am sure it isn’t intentional.
>
>
>
> *From:* NANOG *On
CenturyLink a network which has a long history of troubles, merged with
Level3/TWT/Telcove/etc which had, IMOHO, excellent technical history. They've
created some bastard which is less than what Level3 once was. As legacy Level3,
I still have most of my original team, but I'm confident that
Yep
Regards
Ray Ludendorff
On Aug 30, 2020, at 16:16, Jason Kuehl wrote:
I'm over in MA in a CL building, it's very much still broken. I shut down the
interfaces to CL and now just using Comcast.
On Sun, Aug 30, 2020 at 9:20 AM Andy Brezinsky
mailto:a...@mbrez.com>> wrote:
Started about
If you have to have connectivity to them, you could always just instruct them
not to announce your routes beyond their AS; paid peering, and announce through
more reliable ASs such as 2914 and 1299. Many people do this. Otherwise, cut
ties with them and save yourself the headaches.
> On Aug
I've never heard a single positive word about them, and I've had my fair
share of issues myself (as an indirect customer). But it seems that lots of
people put them in their transit blend. Other than lack of options, why
would anyone use them? To me, it just seems like asking for trouble...but
How is that acceptable behavior?
It's not, the best part. There RCA will be terrible. "Bad Regex" or the
best I ever got was "Bad cable" just two words... My contact is ending
soon...
On Sun, Aug 30, 2020 at 10:29 AM Antonios Chariton
wrote:
> Reporting from Europe, any IP with them in the
Exactly. And asking that they somehow prove this won't happen again is
impossible.
- Mike Bolitho
On Sun, Aug 30, 2020, 8:10 AM Drew Weaver wrote:
> I’m not defending them but I am sure it isn’t intentional.
>
>
>
> *From:* NANOG *On Behalf
> Of *Baldur Norddahl
> *Sent:* Sunday, August 30,
On 30/08/2020 18:22, Joseph Jenkins
wrote:
Well at least it looks like the issue is starting
to resolve and stuff is coming back up.
On Sun, Aug 30, 2020 at 8:21
AM Matt Hoppes
An outage is what it is. I am not worried about outages. We have multiple
transits to deal with that.
It is the keep announcing prefixes after withdrawal from peers and
customers that is the huge problem here. That is killing all the effort and
money I put into having redundancy. It is sabotage
Is this what happens when your entire network is database driven?
I believe from this moment forward things are converging back to normal.
Kind regards,
Job
It’s just due to network size. Horrid service and reliability aside, if you
have enough eyeballs, application providers will want to directly peer, and if
you have enough app providers on net then access providers will want to peer.
With all the acquisitions, they have a ton of fiber in the
Also adding GLBC to this list but as a customer for several CL products
I wish we would have more protection and isolation for our region. Even
we are working in global BGP terms nowadays, that incident makes the
idea of a global CSMA/CD switch more attractive.
A similar event was on
Well at least it looks like the issue is starting to resolve and stuff is
coming back up.
On Sun, Aug 30, 2020 at 8:21 AM Matt Hoppes <
mattli...@rivervalleyinternet.net> wrote:
> Is this what happens when your entire network is database driven?
>
I've been burning before. I'll wait at least an hour before turning my
links back on.
On Sun, Aug 30, 2020 at 11:31 AM Job Snijders wrote:
> I believe from this moment forward things are converging back to normal.
>
> Kind regards,
>
> Job
>
--
Sincerely,
Jason W Kuehl
Cell 920-419-8983
I’ve got CL Fiber+ ent service with BGP for my own v4 and v6 space.
There’s been two outages in the several years it’s been installed - once when
the major nationwide outage happened, and the other when Track Utils touched
our dedicated strand and left us without service for two days while CL
Peace,
On Sun, Aug 30, 2020, 6:02 PM Ross Tajvar wrote:
> Other than lack of options, why would anyone use them?
>
Connectivity and latency (of Level3 which was acquired).
--
Töma
>
Big leaf SDWan is seeing the outage between gateways.
https://status.bigleaf.net/incidents/31r4wts0jlrr
Robert DeVita
Managing Director
Mejeticks
c. 469-441-8864
e. radev...@mejeticks.com
From: NANOG on behalf of Tom
Beecher
Sent: Sunday, August 30, 2020
Latest updates from my tickets:
08/30/2020 14:28:20 GMT - The IP NOC confirmed a routing issue and
commenced with troubleshooting efforts. Routing configuration adjustments
have been made and service affecting alarms are beginning to clear.
08/30/2020 11:38:15 GMT - The IP NOC is engaged in
My systems dropped CL about 03 this morning. Appears that a correction has been
applied as I am seeing more normal traffic from CL. Still no access to their
portal and the phone is a mess.
LQ. Marshall | Sr. Network Engineer | RegEd
-Original Message-
From: NANOG On Behalf Of
Drew
No clue. They’ve been progressively getting worse since 2010. I have no idea
why anyone chooses them and they shouldn’t be considered a Tier1 carrier with
the level of issues they have.
> On Aug 30, 2020, at 11:07 AM, Ross Tajvar wrote:
>
>
> I've never heard a single positive word about
AS3356 is the one I've seen all the chatter about this morning.
On Sun, Aug 30, 2020 at 10:03 AM Robert Blayzor
wrote:
> On 8/30/20 8:14 AM, Drew Weaver via NANOG wrote:
> > Woke up this morning to a bunch of reports of issues with connectivity
> > had to shut down some Level3/CTL connections
That might be because of this:
The IP NOC with the assistance of the Operations Engineering team confirmed
a routing issue to be preventing BGP sessions from establishing correctly.
A configuration adjustment was deployed at a high level, and sessions began
to re-establish with stability. As the
We (eyeball network in Germany) cancelled all our Level3 transit links last
year since it consisted mostly of traffic of their own CDN and they refused to
lower the price per mbit/s or set up peering for the CDN traffic. Now we
receive their CDN traffic via a different Tier1, for a fraction of
On Sun, Aug 30, 2020 at 5:21 PM Chris Adams wrote:
> Once upon a time, Baldur Norddahl said:
> > How is that acceptable behaviour? I shall remember never to make a
> contract
> > with these guys until they can prove that they won't advertise my
> prefixes
> > after I pull them. Under any
This is what happens when the design of 'god power' automation tools
doesn't take into account the concept of blast radius. It might be more
inconvenient to internally partition automated change management systems,
but it can also limit the effect of automation tools gone awry.
67 matches
Mail list logo