After yesterday (and starting again today) all the trunk issues with Bandwidth, we are looking at getting a set of backup trunks that do NOT use Bandwidth.
Does anyone know of a reliable SiP trunk provider that does not use Bandwidth?
1 Like
I honestly don’t know what they use as a back end, but we’ve had no trouble at all with Telnet Worldwide ( telnetww.com ) recently.
Thanks I’ll contact them and ask
I use Twilio for internal SMS messaging. Wonder who they utilize as their ULC? I’ve had good experiences thus far at least for messaging.
Just keep in mind that Telnet is a CLEC in MI. They do not do pricing like Flowroute or other ITSPs. They charge per path/channel. They are not Pay as you go. Outside of MI they use Bandwidth, Intelliquent (used to be Broadvox) and other carriers because they don’t have infrastructure outside of MI/WI.
Disclosure: I spent many years there and was part of their jump to SIP.
1 Like
Ok thanks, so they might not be the best option. I’ll have to keep looking
I’ve had good success with Flowroute. Vitelity does not use bandwidth either but they have been having their own issues. There are users here who swear by Skyetel, though I don’t have any experience with them.
Reading up a bit, it seems that Twilio’s ULC is AT&T. If that’s correct I might give them a look-see. Pay as you go SIP services ($0.005/min making calls, $0.0034/min taking calls), free DID ports, and $5/month DID’s. Seems reasonable. Any of y’all have experience using them for SIP services?
We are in the process of switching all of our voice to SIP trunks through an AT&T IP Flex circuit. I’m like 99% sure that’s completely independent from bandwidth.com.
Our port date for all of our phone numbers is in 2 weeks… just far enough out that we get to participate in all this bandwidth.com issues.
Well in the US we have Incumbent Carriers (ILECs), ATT and Verizon are ILECs. That means they directly own the infrastructure of the telecom network. We also have Competitive Carriers (CLECs). They generally do not own the infrastructure, the ruling in the 90’s that made CLECs a thing forces the ILECs to share their network with CLECs. TelnetWW is a CLEC in Michigan which means when you order a circuit from them, depending on where you are in the state, 75% of the time it means ATT is the one installing the circuit and X-connecting it back to Telnet in the CO.
There is a reason that telecom is a natural monopoly in the US. It costs billions to have the infrastructure to do this. In a lot of cases there are only a handful of ILECs that have control of the infrastructure so even other major carriers are at their mercy.
So while people are complaining about getting off of Bandwidth just remember that when ATT, Level 3, Verizon, et al have an outage that waterfalls down to everyone that depends on them. I mean it’s not like Level 3 hasn’t caused most the Internet to stop working at times.
Oh yeah and while Bandwidth was getting hit yesterday let’s not forget that Telnyx had an outage. Level 3 had issues (could’ve been a DDoS) and Verizon had issues that I am aware off because calls to/from Verizon where having issues until about 8AM this morning.
So I have to ask, if ATT has an outage (oh and they do) are people going to have to jump to someone else that doesn’t use ATT in their network? At what point is a carrier going to be 100% perfect and avoid any outages?
I shudder to think how some people would have handle things like this 15-20 years ago when you couldn’t just port in 24 hours to some place else. Not because you couldn’t port but because you really didn’t have a lot of carrier options and this “be connected in 5 minutes” stuff was highly untrusted when it did exist.
we’ve been relatively fond of DIDLogic as one of our providers.
Simple setup, “decent” rates. nice use of multiple caller IDs.
That indeed makes it challenging in terms of bringing some true fault tolerance into a specific site location. Our headquarters can have primary fiber Internet coming in from AT&T. And then if I want to add a secondary failover, my vendor selections are limited to who has presence for our service address. Dropping in a Charter/Spectrum fiber circuit is more or less pointing back to the ULC, which is AT&T I believe. If there is a major event affecting AT&T’s network then HQ is dead in the water until it’s resolved.
I think a lot of folks (myself included) are getting antsy since it seems as lately the amount of bad actors wreaking some havoc is increasing in frequency and scope. But stopping to consider the big picture is a smart idea.
one item we have been investigating is an LTE failover for our sip trunking.
for the cost, i can get 10-50gigs a month of LTE data for sub-$200/mo, or pay-as-you-go data.
Assuming a call is 56kbps up/down plus overhead, lets say 100kbps symmetrical, LTE (in theory. 100.8mbps for a 64QAM cchannel…). I can rock almost 1,000 concurrent calls over LTE with minimal issues.
is it Janky as hell? yes. is it a horrible idea? double yes. Is it better than an outage? very much yes.
1 Like
I was looking into something similar for next year, as I will have Cisco Meraki units at each site. Popping in a cellular modem as a secondary failover at least diversifies things a bit if the primary circuit is having trouble.
1 Like
Interesting information! I’ve been a client of theirs since the PRI days & before - almost 20 years - and never knew any of that.
Then there’s a good chance I might even know who your company is. 
Probably even worked tickets for you guys.
While I agree to most of this, it is still a good idea (OP’s original post) to get a backup trunk from a provider that does not use the same upstream carriers as your primary.
We have all read this reddit comment from that redditor who ported their numbers from VoIPms to another carrier and his customers are now fuming that they AGAIN have an outage…
I personally think, that if you don’t have a TFN then it is not worth porting and jumping between carriers. Just wait it out… If you don’t have issues with your incoming calls, you may have issues when calling other numbers… 40% of US service providers are using Bandwidth.
Bandwith DDoS has resulted (for me) in the last 4 days of noticing almost no issues during the 4 days and only 4 customer complaints. Monday was the only day during the 4-day attack I saw actual issues with my calls from Bandwidth and that was about 10-15% of the incoming calls. The complaints I had from customers were about intermittent issues and not constant issues. For the most part they were understanding and let it go.
Well I do understand that end users are going to flip their minds over outages, the amount of people saying their having mass customer churn of this (the wholesale/resale people) I would have to venture other issues are already happening with them. Been doing this a long time and have had multi-day outages, I’ve never seen a huge churn from my customers during that time. I have had churn but nothing like what people are painting over the VoIP.ms outage or even this one.
So I would say these people are either over dramatizing things or they have other issues that contributed to these outages being the straw on the back.
1 Like
Things did seem to settle down for my sites, although BW still was reporting ongoing impact. The one element that we only heard about second-hand and after the fact was besides the occasional choppy audio there were issues where external callers couldn’t route to ring into our sites via our assigned DID’s. The site users wouldn’t know about this really, although a few external callers kept trying and eventually rang in and remarked about what had happened.
Great observation about jumping ship. If the VoIP DDoS vector will become another new norm, then porting over to a different provider won’t likely change anything. I was looking at Twilio out of curiosity. More due to cost savings than anything. Their ULC appears to be BW just like my current provider anyway.
The only client I moved during this was one that already planned to submit a port out to a new carrier on October 1st. We just moved up the timeline by two weeks.