Cox cable "outages" and a solution
I'm posting this info here because I know quite a few of us have BYOI customers and may be affected by this. Cox users have been reporting a lot of outages lately in certain areas (it seems to cluster by whatever areas Cox is making changes in). I found that when there is a supposed outage, the only thing that fails is DNS. You can still ping/trace/connect with an IP address. I looked at the router's DHCP-assigned info and found that it was seeing an IPv6 DNS as its primary, with the usual two IPv4 servers as secondary/tertiary. Oddly, this didn't seem to happen right away, but would happen after the router had been up for a few hours. My best guess at the root problem is that Cox isn't properly handling IPv6 yet, even though they are advertising it to the router. Completely disabling IPv6 in the router fixes it.
I had a major issue with Cox when they were deploying ipv6. They set a rate limiter on packets, and any voip call would trigger it. Once the call ended, the speed would come back. The way we proved it was to ping from the router to the immediate hop after the modem... once we could prove it this way to the field tech, they got "tier 3" to call a network engineer and resolve it. Hopefully it's similar. Fred Posner The Palner Group, Inc. http://www.palner.com (web) +1-224-334-FRED (3733) direct On 03/19/2016 12:48 PM, Carlos Alvarez wrote:
I'm posting this info here because I know quite a few of us have BYOI customers and may be affected by this. Cox users have been reporting a lot of outages lately in certain areas (it seems to cluster by whatever areas Cox is making changes in). I found that when there is a supposed outage, the only thing that fails is DNS. You can still ping/trace/connect with an IP address. I looked at the router's DHCP-assigned info and found that it was seeing an IPv6 DNS as its primary, with the usual two IPv4 servers as secondary/tertiary. Oddly, this didn't seem to happen right away, but would happen after the router had been up for a few hours. My best guess at the root problem is that Cox isn't properly handling IPv6 yet, even though they are advertising it to the router. Completely disabling IPv6 in the router fixes it.
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
When Cox first launched IPv6 in my area, I'd lose v6 routing randomly every few days until I manually renewed the address on my router. Then I found out one tiny config difference between Cox and Comcast is that Cox doesn't support rapid commit renews. So I removed that from my router config and I haven't lost routing since. Also if your customers use a Motorola/Arris SB6183 model modem, the firmware Cox uses has a major bug where it drops packets. Cox is testing new firmware and rolling it out to those who request it. See: https://www.dslreports.com/forum/r30562682-Arris-SB6183-IPv6-TCP6-bug-Lookin... ~Jared On Sat, Mar 19, 2016 at 10:10 AM, Fred Posner <fred at palner.com> wrote:
I had a major issue with Cox when they were deploying ipv6. They set a rate limiter on packets, and any voip call would trigger it. Once the call ended, the speed would come back.
The way we proved it was to ping from the router to the immediate hop after the modem... once we could prove it this way to the field tech, they got "tier 3" to call a network engineer and resolve it.
Hopefully it's similar.
Fred Posner The Palner Group, Inc. http://www.palner.com (web) +1-224-334-FRED (3733) direct
On 03/19/2016 12:48 PM, Carlos Alvarez wrote:
I'm posting this info here because I know quite a few of us have BYOI customers and may be affected by this. Cox users have been reporting a lot of outages lately in certain areas (it seems to cluster by whatever areas Cox is making changes in). I found that when there is a supposed outage, the only thing that fails is DNS. You can still ping/trace/connect with an IP address. I looked at the router's DHCP-assigned info and found that it was seeing an IPv6 DNS as its primary, with the usual two IPv4 servers as secondary/tertiary. Oddly, this didn't seem to happen right away, but would happen after the router had been up for a few hours. My best guess at the root problem is that Cox isn't properly handling IPv6 yet, even though they are advertising it to the router. Completely disabling IPv6 in the router fixes it.
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
If you?ve deployed hard phones for these users, try configuring the phones to use your own DNS servers. From: VoiceOps [mailto:voiceops-bounces at voiceops.org] On Behalf Of Carlos Alvarez Sent: Saturday, March 19, 2016 11:49 AM To: voiceops at voiceops.org Subject: [VoiceOps] Cox cable "outages" and a solution I'm posting this info here because I know quite a few of us have BYOI customers and may be affected by this. Cox users have been reporting a lot of outages lately in certain areas (it seems to cluster by whatever areas Cox is making changes in). I found that when there is a supposed outage, the only thing that fails is DNS. You can still ping/trace/connect with an IP address. I looked at the router's DHCP-assigned info and found that it was seeing an IPv6 DNS as its primary, with the usual two IPv4 servers as secondary/tertiary. Oddly, this didn't seem to happen right away, but would happen after the router had been up for a few hours. My best guess at the root problem is that Cox isn't properly handling IPv6 yet, even though they are advertising it to the router. Completely disabling IPv6 in the router fixes it.
We have always deployed our phones to only use our own DNS and SNTP servers. The phones only accept the IP and gateway from DHCP. We found out about the issues because customers' IT people were asking us why the phones were working and "the internet" was not. After stringing them along on answers about voodoo and black magic, we figured it out. I posted here because not everyone doest things this way. On Mon, Mar 21, 2016 at 9:39 AM, Lee Riemer <LRiemer at bestline.net> wrote:
If you?ve deployed hard phones for these users, try configuring the phones to use your own DNS servers.
*From:* VoiceOps [mailto:voiceops-bounces at voiceops.org] *On Behalf Of *Carlos Alvarez *Sent:* Saturday, March 19, 2016 11:49 AM *To:* voiceops at voiceops.org *Subject:* [VoiceOps] Cox cable "outages" and a solution
I'm posting this info here because I know quite a few of us have BYOI customers and may be affected by this. Cox users have been reporting a lot of outages lately in certain areas (it seems to cluster by whatever areas Cox is making changes in). I found that when there is a supposed outage, the only thing that fails is DNS. You can still ping/trace/connect with an IP address. I looked at the router's DHCP-assigned info and found that it was seeing an IPv6 DNS as its primary, with the usual two IPv4 servers as secondary/tertiary. Oddly, this didn't seem to happen right away, but would happen after the router had been up for a few hours. My best guess at the root problem is that Cox isn't properly handling IPv6 yet, even though they are advertising it to the router. Completely disabling IPv6 in the router fixes it.
participants (4)
-
caalvarez＠gmail.com -
fred＠palner.com -
jared＠compuwizz.net -
LRiemer＠bestline.net