Our Server Is Extremely Unreliable & Undependable

[Main Street James]

Our VPS, which is on the kvothe.nocdirect.com node, has been down several times a day since Wed, Dec 7th. And by "down" I mean unresponsive, inaccessible and unavailable via ssh, sftp, http, ftp, telnet, etc, etc. The kvothe.nocdirect.com node even shows as down on the JPC Network Status page. That's 11 days in a row. Every time we report it we get some different answer, excuse or reason. Obviously none of them have been the correct diagnosis or we wouldn't have been suffering with this sub-par performance and support for 11 days.

I've gotten so many incorrect answers as to what's going on with the node it's been like a bad episode of House.

We've opened tickets; we've called and yelled (when the tech support phone number actually works); we've been contacted by the tech support manager. And yet we continue to go down. Sometimes for just 10 - 15 minutes, sometimes for hours at a time.

Our node has a 98.36% uptime for December (which doesn't include the downtime for today). That's 6 1/2 hours of downtime. But, being unresponsive, inaccessible and unavailable via ssh, sftp, http, ftp, telnet, etc, does not necessarily qualify as "down" based on the JPC definition, so our true availability is much worse.

We have lost two more clients because they could not access the web apps they count on to run their businesses. This only a month after we lost our biggest client due to JPC's failure to address a server issue we had been dealing with for over a month.

After several days of trying to call JPC's attention to the chronic problems with the kvothe.nocdirect.com node, we received an offer to move to a different node. When we asked what, if anything, we would need to do to facilitate this escape from the kvothe.nocdirect.com node we were told "nothing". At the last minute we were told that we would have had to change all of out IP's. That means changing every DNS entry for every one of the dozens of domains that point to our VPS. This would have taken about two hours … and if I'm going to spend two hours changing DNS entries it will be to point them to a new hosting company.

I don't know what else we can do. We've reported the problem numerous times. We've provide information directly to the tech support manager. We've been waiting 11 days. But, as I write this, we are down again.

Ticket # 13461099

[JPC-JamesG]

Hello,
I am very sorry to hear about your problems you have been experiencing, obviously, I want to assure you that I am doing everything I can to ensure this gets resolved.

I have notified our tech dept abotu your ticket and they are working on it as I write this.

Let me point something out regarding the IP situation. The reason that you will get a different set of IPs is becasue different nodes reside in different VLANS where the IPs are routed. If it was posible to transfer you to a different node in the same VLAN, then the IPs could go with you. I do understand your frustration regarding this and know it is a big hassle to change your DNS zones, however, DNS zonees on the server would actually be updated automatically for you, so you would really only have to update the nameserver IP addresses for those domains that require that, unless you are using a third party for DNS hosting.

We will look thoroughly into the issues caused on the node and advise you of exactly what has happened this past week or so.

Our techs will be in contact with you shortly on your ticket and we will see if we can get this resolved in a timely and affective mannor for you.

[Main Street James]

I'm aware of the reasons behind changing all of our IP's in order to escape the node we're one, but when we asked what we needed to do (if anything), we were told "nothing". Then, when we were casually told we'd have to change IP's and asked "Please confirm if you are fine with IP change", we had two thoughts:

• "what else would we need to do that they just aren't telling us?" since our confidence in the "support" that this problem has received - as well as other serious server issues we've encountered - couldn't be much lower.
• "how long will we actually be down?" considering that we don't know when the move would be complete, what the new IP's would be, and propagation of the new DNS entries across the globe will take hours.

Every time we report this problem we get a different answer as to what is wrong and why JPC thinks that it is fixed. Charlie Brown, Lucy and that football comes to mind.