anyNode Discontinues Chicago Location

[TheLinuxBug]

I just received this e-mail not to long ago from anyNode, it looks like they will be discontinuing their Chicago location. They are sending me this even though I have an account in Detroit.  I am a little concerned now with the sustainability of their services if they are not even able to sustain what I thought would be their more widely used location. Anyhow, what do you guys think?

-------------------------------------------------------------------------------------

Hello anyNode customers,

We have decided to cease service in our Chicago location on August 31st
at 5:00 PM EDT. What does this mean for you?

First, we will be assigning everybody a new IP address today (August
13th). This address will function alongside your existing address in the
216.107.159.* range until our cutover date. On the cutover date, your
216.107.159 address will be removed and the new 162.217.176.* address
will become your primary.

Second, your routes and latency will change. As our Detroit location's
network is superior to our current Chicago location, many users will
notice an improvement over their existing connection.

We suggest you migrate any DNS records and services over to the new
addresses as soon as possible. The existing addresses will continue to
work until August 31st.

You can test our Detroit network with the following address and test file:
IP: 162.217.179.1
File: http://help.activesolutionsmi.com/100mb.test

If you have any questions or concerns please submit a ticket in our
helpdesk.
--
Seamus Caveney
anyNode.net - System Administrator
NOC: 313.566.4166

-------------------------------------------------------------------------------------

I am also curious why they didn't announce this here first?

 

[drmike]

Well, it's a shame providers move around so quickly (then again, maybe they've been there for eons and we just started seeing them around here).  Usually when this happens (and the hated IP change) I tend to just go shopping elsewhere.  I detest service changes.

Another company that bailed on Continuum Data Center.   Seems to be semi-common.    Though oddly with my upstreams, Continuum gets the best throughput compared to just about any other facility.

Problem with Detroit, is traffic is backhauled likely from Chicago, so unneeded latency.   Better network?  Maybe.   I saw Level-3 routes to their Detroit installation.

Hoping this is a good things for anyNode and not a sign of problems.

 

[terafire]

Don't give them too much of a hard time, they're just looking for a better bandwidth provider.

 

[Francisco]

Continuum is rough.

While they've offloaded some of their DDOS issues onto blacklotus, it's simply taken way too long to get that far.

They've had huge network issues, almost always ddos for years now.

Francisco

 

[Aldryic C'boas]

I am also curious why they didn't announce this here first?

I applaud them for not, actually.  Seamus kinda looks at things like this the way I do - obligation is informing your clients, then unaffiliated mediums such as forums/etc.  I had a service with another provider for awhile... and it would always piss me off that I'd find out about changes/etc on a forum online a day or more before actually getting an official email from their system as a client.

 

[TheLinuxBug]

I applaud them for not, actually.  Seamus kinda looks at things like this the way I do - obligation is informing your clients, then unaffiliated mediums such as forums/etc.  I had a service with another provider for awhile... and it would always piss me off that I'd find out about changes/etc on a forum online a day or more before actually getting an official email from their system as a client.

OK, at that rate, lets remove the word "first".  How about, I am surprised as active as they have been here that they did not announce this here once they sent out the e-mail.  Also, it would have been nice to have more details as to why they made the choice to leave.  Now, I do not use their Chicago DC, what people above have said makes sense, as I know Continuum isn't that great of a DC, but it still would have been better to hear it is for service quality reasons than "We are just leaving, you have 15 days to be migrated, k thnx bye".  They do mention better connectivity, and that is great, but at my first read it sounded like more of a "Hey we can't afford this, plus XYZ is better anyhow, so we are moving".

Heck, they haven't even come and said anything in this thread yet?

(Now you will probably tell me that they are busy worrying about the  change over, right?)

Cheers!

 

[drmike]

Lots of providers ---- in their defense --- aren't hawks for every mention of their company online.   

I am amazed at how on top of social things some providers are.   It's a high bar to reach and to me, in a small company, points to idleness (slow business?).

I see no harm. Short term notice being the only blargh point.

Continuum is still a place not perfected and that likely led to this.  Other providers more known around here, like Backupsy have left Continuum recently.  Consider it a right of passage.

 

[Francisco]

Heck, they haven't even come and said anything in this thread yet?

I linked this off to Seamus so he'll get a reply soon.

His idle time is 4 hours though so it's likely he's asleep since he's Eastern.

From what he said in #frantech the amount of users over there wasn't enough to justify physically moving the nodes to Detroit, so instead he's just vzmigrating people.

Francisco

 

[scv]

I just received this e-mail not to long ago from anyNode, it looks like they will be discontinuing their Chicago location. They are sending me this even though I have an account in Detroit.  I am a little concerned now with the sustainability of their services if they are not even able to sustain what I thought would be their more widely used location. Anyhow, what do you guys think?

[snip]

I am also curious why they didn't announce this here first?

My apologies to you (and everybody else with one of our services) - I should've included an explanation as to why we've decided to shut down service in Chicago in the announcement and email. As a few people have said in the thread, it's entirely due to Continuum's poor network. We've had our equipment colocated there since early January this year and the entire time it's been less than stellar. High latency and packet loss is no uncommon event. I spent the better part of last month trying to find colo in Chicago that was both cost-effective and offered the quality we want to deliver to our clients. While we did get some tempting offers, our Detroit location has sold much better than our Chicago location. The logical move was to cease service in Chicago and focus on Detroit - not only due to better sales but due to the fact the network is significantly more stable.

We just received an ASN and IP allocations from ARIN this past month. At some point in the next 6 months we're required to return all previously existing IP space to our upstreams. Migrating our customers to our own IP space at the same time we do a facility move kills two birds with one stone. We actually just finished adding the new addresses to all customer instances so they have a two week window to update their services and DNS - both the old and new addresses will remain active like we mentioned in the email. Hopefully on the cutover date all of our clients will have completed their migrations and it'll be relatively pain-free for them.

Well, it's a shame providers move around so quickly (then again, maybe they've been there for eons and we just started seeing them around here).  Usually when this happens (and the hated IP change) I tend to just go shopping elsewhere.  I detest service changes.

Another company that bailed on Continuum Data Center.   Seems to be semi-common.    Though oddly with my upstreams, Continuum gets the best throughput compared to just about any other facility.

Problem with Detroit, is traffic is backhauled likely from Chicago, so unneeded latency.   Better network?  Maybe.   I saw Level-3 routes to their Detroit installation.

Hoping this is a good things for anyNode and not a sign of problems.

We wanted to avoid any sort of major outage like this, especially since we've only been publicly offering services for a short time, but Continuum is just unacceptable for us[1]. Our DC in Detroit has some bandwidth backhauled to Chicago but only as a backup. They have 10Gbps links from Level3 and Cogent + 1Gbps from Comcast directly in Detroit, 10Gbps links to Equinix Chicago and Ashburn, 10Gbps Global Crossing in Chicago, and 10Gbps Savvis in Ashburn[2]. The entire time we've been at Continuum I've seen nothing but the same HE.net 10GbE on outbound routes, and only very rarely inbound routes through Ubiquity Servers.

I linked this off to Seamus so he'll get a reply soon.


His idle time is 4 hours though so it's likely he's asleep since he's Eastern.


From what he said in #frantech the amount of users over there wasn't enough to justify physically moving the nodes to Detroit, so instead he's just vzmigrating people.


Francisco

Francisco's right - we only have 6 nodes in Chicago and the amount of used disk space is low enough that it's faster to vzmigrate everybody to temp nodes compared to taking a 10 hour round trip to pick up the gear.

I'll be updating the official announcement with an explanation as to the move - hopefully not too many people think this is budget related. We've been at least breaking even since our first month of sales (May) and things have only improved since then. I just finished building two new nodes today that will be used for KVM services we're planning to introduce next month.

If anybody does have concerns about anything at all, feel free to PM me or shoot me an email directly and I'll do my best to answer your questions. Again, my apologies for the rather abrupt announcement without an explanation.

I've updated the announcement available on our site to describe the reasoning behind the move - https://billing.anynode.net/announcements.php?id=7

[1] - http://monitor.tortois.es/smokeping/smokeping.cgi?target=anyNode.Chicago - courtesy of kaniini

[2] - http://www.123.net/about/data-centers/ - we're in DC3

 

[drmike]

Thanks Mr. Seamus!   Total sense made on this one.

I like the Detroit location -- metro Detroit I mean.  It's not the same location as everyone else.

123.net (killer domain) has a pretty good network  From my location I hit it via Chicago and that issue is my own provider upstream that routes everything out to Chicago, albeit moronic.

 

[scv]

Thanks Mr. Seamus!   Total sense made on this one.

I like the Detroit location -- metro Detroit I mean.  It's not the same location as everyone else.

123.net (killer domain) has a pretty good network  From my location I hit it via Chicago and that issue is my own provider upstream that routes everything out to Chicago, albeit moronic.

Although we aren't directly in Detroit proper Southfield is a mere 15 minutes north of Detroit and a part of the metro area. Something along the same lines as 'Buffalo' being 'New York'   

123.net has been very reliable for us. Our Hosted PBX product is running at 123.net and with all our VoIP services coming from Level3, it's proven to be a boon. Their network has been nothing short of outstanding and their technicians are top notch.

 

[wlanboy]

Lots of providers ---- in their defense --- aren't hawks for every mention of their company online.   

I am amazed at how on top of social things some providers are.   It's a high bar to reach and to me, in a small company, points to idleness (slow business?).

I see no harm. Short term notice being the only blargh point.

Continuum is still a place not perfected and that likely led to this.  Other providers more known around here, like Backupsy have left Continuum recently.  Consider it a right of passage.

Yup, the sozial game is not that easy. And many providers underestimate the customer perception on how the company is handling things.

My apologies to you (and everybody else with one of our services) - I should've included an explanation as to why we've decided to shut down service in Chicago in the announcement and email. As a few people have said in the thread, it's entirely due to Continuum's poor network.

It is ok. I had some tickets about network problems and you always tried to solve the issue. Chicago and Detroit are quite near - so I don't know where the cant comes from.

I mean ... you are not moving from Chicago to L.A. - but from Chicago to Detroit.

 

[kaniini]

Out of curiosity, why would it be necessary for a company to announce on vpsBoard things which only affect legacy customers?  I would figure that notifying the legacy customers of the change would be the first step, and then actually doing the migration is the next step.

Continuum is crap, we looked into moving many years ago when RapidXen was affected by the 26-hour steadfast network outage.  We found that steadfast was superior, so continued with them.