Time for a new DC in CA?

Status
Not open for further replies.
M

magic

New Member
Now fortunately I don't have anything hosted in CA but from all the other folks here on this board it seems that they are having recurring network/router problems in that DC. Is KH considering any options to move to a new DC if the service quality there does not improve?
 
First, I would like to apologize for the issue today along with the whole KH team. It was unfortunate but as you can see we were doing all we humanly could the entire time. I can assure you we will do what we feel is best for our customers so that shouldn't be questioned. Moving DC's isn't as simple as picking up and leaving so that just isn't an option. We have quite a bit of experience and will do what we feel is best. Thanks for understanding.

Regards,
Joel
 
CA has regular outages of < 10 minutes.

I wouldn't normally claim on the SLA but I think an outage of over 3 hours at CA is unacceptable and KH need to send a clear message to the network providers at CA about it.
 
I totally agree 3 hrs is not right and can assure you the message we sent was very clear. We truely care how this affected customers and if such an event happens again, actions will be taken. Please understand we care and are very sorry.

Jay
 
Hello,

Still regarding the last outage at CA.

The currently network provider at CA is Market Post Tower, right? So if it stops, the entire CA datacenter goes down.

Isn't there a way to have two providers? Market Post Tower and another one? So when one goes down the other keeps the CA datacenter alive even with a slow network.
 
I think they use 2 network providers already - cognent and verio.


The problem was a router I think - why it took over 3 hours to rectify is absurd though.
 
I used to have a couple servers at One Wilshire, and never experienced anything like this. Looks like the MPT boys need to take some lessons from their LA counterparts. ;)
 
I'm not sure if any of you guys really understand what's going on here...and it's very unfortunate that you obviously think you do.

Basically, MPT provides space and power. The 90+ networks listed on the site are available for KH to connect to. There are also larger providers that have a mix of a selection of the 90 available network providers within the facility...KH goes through one of these providers. Their provider has 2 carriers, Verio & Cogent.

The outage has nothing to do with MPT directly. However, one would assume that many who are within the facility were affected, but the actual issue has nothing to do with MPT.
 
gsv said:
I'm not sure if any of you guys really understand what's going on here...and it's very unfortunate that you obviously think you do.
Click to expand...

I think it's the other way around GSV. In any event, we used to have quite a few servers in a handful of datacenters, and the datacenter ALWAYS provided the routing/connections to whatever providers. It was the datacenter who provided support for connectivity/router issues, etc.

I have no idea how MPT handles theirs, or what type of support relation they have with KH, but I do know that One Wilshire handled (or used to handle) all connectivity issues with their clients.

So yes, if MPT handles the routers/connectivity, it IS their problem.
 
KnownHost-Jay said:
I totally agree 3 hrs is not right and can assure you the message we sent was very clear. We truely care how this affected customers and if such an event happens again, actions will be taken. Please understand we care and are very sorry.

Jay
Click to expand...

Well... It just happened again. I love the communication with you guys, but this much downtime is unacceptable.
 
I totally understand and am just as unhappy. We will not sick back as I said before. This will be addressed if that means moving to a new DC.

-Jay
 
Seems okay to me right now. We will open a thread for this if there is an issue as we always do.

Joel
 
If you see a problem again please submit the tracert to Support as everything seems normal right now. Thanks!

Joel
 
Status
Not open for further replies.
Top