amuck-landowner

Colo@ and SecureDragon Clifton, New Jersey Issues / Migration

drmike

drmike

100% Tier-1 Gogent
As per SecureDragon's Twitter:

https://twitter.com/SecureDragonLLC/status/448761993017978880

"Apparently our NJ servers are being sent to another data center. No notice was given to us and we're not sure where they are going."

Others commented on here about the situation....

What is going on?

The facility in Clifton, NJ, is said to be: COLO@ (Telx NJR2)

@qps said:

"I believe they are co-located with Colo@. The move announcement was made some time ago. Colo@ is co-located with FortressITX/Dedicatednow/Pwebtech, who is moving from Telx in Clifton to I think Digital Realty Trust in Weehawken."
 
Last edited by a moderator:
Patrick

Patrick

INIZ.COM
Verified Provider
It seems Dedicated Minds had planned the move, possibly new/recent clients who joined Colo@ were not notified and still accepted orders to old facility?

I remember them pulling back the date due to bad weather last month. 
 
S

Steven F

New Member
Verified Provider
January 17th, 2014 - Dedicated Minds sent out an e-mail that they would be migrating on February 9th.

February 5th, 2014 - Dedicated Minds sent out an e-mail that they would be postponing the migration.

March 10th, 2014 - Dedicated Minds sent out an e-mail that they would be migrating on March 26th.

March 26th, 2014 - Colo@ migrates the Dedicated Minds servers, expecting to take a few hours, but it takes much longer.

The migration was planned, but it seems to have been botched. I can't comment on SecureDragon, though.
 
KuJoe

KuJoe

Well-Known Member
Verified Provider
After purposely going to bed as soon as I got home this morning for fear of losing my temper, I am still furious right now at this whole situation. Since we still have servers in production in their other data centers I'm just going to hold my tongue and pretend this never happened.
 
Last edited by a moderator:
wlanboy

wlanboy

Content Contributer
Today I got a lot of statuscake emails about "your vps is down" ... "your vps is up" ... "your vps is down" ...
 
KuJoe

KuJoe

Well-Known Member
Verified Provider
After taking a moment to collect my thoughts and type up a ticket to the COLO@ management. I'm able to better reflect on the situation and instead of focusing on all of the negative things we've experienced thus far, I'll focus on the good things that came from this:

1) I found a bandwidth calculation bug in Wyvern when a node is offline for an extended period of time.

2) Only one client VPS was actually online on our NJ node at this time (the rest were pending migration or payment luckily).

3) We've realized that the saying "all your eggs in one basket" doesn't just mean all of your VPS nodes in one location but also all of your VPS node with one company across multiple locations. We are looking at diversifying our future locations even in markets where COLO@ is available.
 
drmike

drmike

100% Tier-1 Gogent
I won't push.  Sorry something got dropped/mishandled Joe.

Colo@ gets good words from most people I trust.

Perhaps honest human mistake?
 
KuJoe

KuJoe

Well-Known Member
Verified Provider
Auggie (Director of Business Development) responded to our ticket and made things right by us and assured us that these oversights will not happen again. I cancelled my request to have our AZ server shipped back to us so we do plan on continuing our business relationship with COLO@.

I'm glad too because I am very happy with their Chicago location (as are a lot of our clients).

Their LA location is nice but we receive far to many network down alerts for some of our IPs that are most likely false positives since our other IPs on the same server are online. You can see what I mean here. For "-Los Angeles" it shows 97.747% uptime for the past 30 days yet our ca1ovz01 node has 100% uptime. "-Los Angeles" is the gateway IP for ca1ovz01 so if the router only had 97.747% uptime then it would be impossible for the server on that router to have 100% uptime. We have 2 additional tests that are not publicly available that backs up the 100% uptime for ca1ovz01 so the 97.747% uptime for the router are all false alerts.

Their Atlanta location has been excellent so far (100% uptime, good network, speedy support since that's where they are based out of). No complaints once the server and network was online.

New Jersey was the one location we've been looking forward to the most which is probably why we've stuck it out for this long. We'll see how things go from here on out, hopefully as well as our Chicago and Atlanta servers.
 
Last edited by a moderator:
Jack

Jack

Active Member
Haha! Don't worry Wiresix do this to all clients I had it a few months ago with Atlanta.
 
Wintereise

Wintereise

New Member
> "-Los Angeles" is the gateway IP for ca1ovz01 so if the router only had 97.747% uptime then it would be impossible for the server on that router to have 100% uptime

This is normal ICMP de-prioritization in most carrier grade routing equipment. When the PFE is routing actual packets linerate, it'll discard ICMP/similar probe packets -- some rate limit as well.

If I were you, I'd remove that monitor off that IP.
 
Last edited by a moderator:
KuJoe

KuJoe

Well-Known Member
Verified Provider
Wintereise said:
> "-Los Angeles" is the gateway IP for ca1ovz01 so if the router only had 97.747% uptime then it would be impossible for the server on that router to have 100% uptime

This is normal ICMP depriotization in most carrier grade routing equipment. When the PFE is routing actual packets linerate, it'll discard ICMP/similar probe packets -- some rate limit as well.

If I were you, I'd remove that monitor off that IP.
Click to expand...
That's strange that it's not like this for any other location. I'll make some adjustments to our monitoring.
 
Last edited by a moderator:
KuJoe

KuJoe

Well-Known Member
Verified Provider
Thanks for opening this thread @drmike, we got a really nice offer for a new location that we're shipping a server to tomorrow and should be online next Friday after a 48 hour burn-in. DIVERSITY!


Edit: We're not moving away from COLO@, just want to make that clear to avoid confusion.
 
Last edited by a moderator:
You must log in or register to reply here.
Top
amuck-landowner