100% uptime on Cloud VPS?

**John** · #1 (**permalink**) 17-05-10, 08:46 AM

Hello All,

This thread is designed to address the concerns a few customers have in relation to our Cloud VPS products.

There have been a few complaints in regards to "downtime" on a Cloud VPS. All of these complaints share the same cause, which is the customer's VPS attempting to use more resources than has been allocated.

As an example, please view the below specs of our VMware STARTER, which are displayed on our website:

CPU 800 Mhz
Hard Drive Space 25 GB
Monthly Bandwidth 600 GB
Guaranteed RAM 512 MB

As you can see, just like a PHYSICAL server, there are hardware limits. You can not use more than that has been allocated.

Perhaps some confusion has come from those customers who are used to a Virtuozzo VPS. On a Virtuozzo VPS the hardware Node will allow you to use "burstable" resources when they are available on the server. This is often a result of another customer not using their full entitlement. On a Cloud VPS, you have your allocation and you are limited to it.

If your VPS attempts to use more than say 512 RAM, when you have the starter package, it is quite likely that your services will crash and you may therefore associate this with "downtime". Is it downtime? Quite simply, no. You physically can not use more than you have. This is especially true for those customers using the Windows platform.

If a customer needs more resources, they will need to upgrade to a higher plan. I paste below the specifications of our VMware BUSINESS and VMware ENTERPRISE:

VMware BUSINESS
CPU 1.5 Ghz
HDD 45 GB
Bandwidth 800 GB
RAM 1 GB

VMware ENTERPRISE
CPU 2.2 Ghz
HDD 65 GB
Bandwidth 1000 GB
RAM 2 GB

As you can see, they each have different resource allocations. If a customer needs more resources, an upgrade will be needed.

Please let me know if you have any questions or concerns. I will be quite happy to address any questions you have.

sipero · #2 (**permalink**) 19-05-10, 03:59 PM

Hi,

In my case prior to migrating to the VPS Cloud I was using a lower spec dedicated server with only 1 Gb of ram. In the past month since the migration I've had more downtime than the past 3 years put together.

I chose the highest spec VPS with 2 Gb of ram so did not expect to have these issues.

Today my VPS is down again, and this time I've been told this is due to a problem with the host node.

My questions are this/

1. When I contact you for support which department should I chose it seems what ever one I chose it simply get transferred to another one and is different everythime.

2. You fall over and other redundancy that this is meant to use so why is my VPS and all web sites down for more than 1 hour and at this point of time still offline

Jonathan Hamon

**Sebastian** · #3 (**permalink**) 19-05-10, 07:34 PM

Hello,

We have checked everything on VMware Cluster and all VMs are up and running fine. Some of the customers are able to access their VMs, it seems there is a problem with the network. You should be able to ping your VM but you might not be able to SSH or RDP. As per the feedback got on Live Chat and phone calls many of the users are able to access their VMs through few ISPs. We are also checking this issue with network engineers. We are trying our level best to solve it for you.

sipero · #4 (**permalink**) 19-05-10, 07:41 PM

Hi,

None of my sites are working, pinging the VPS fails, and I cannot access via RDP either. Just coming up to 5 hoiurs now.

Jonathan Hamon

**James** · #5 (**permalink**) 19-05-10, 08:05 PM

Please accept our apologies for the inconvenience that this is causing and be assured that we are working hard to resolve the issue.

sipero · #6 (**permalink**) 19-05-10, 08:17 PM

Hi,

With your so called
Distributed
Resource Scheduler (DRS)
• VMware High
Availability (HA)
• VMware Fault Tolerance

Together with the Cloud feature why has this happened.

Cloud VPS hosting is fast becoming one of the most popular forms of web hosting available mainly because it offers high availability in the form of 100% uptime. Cloud hosting environment makes use of a number of different servers meaning that there will always be at least one server available to cope up with the load that is going to be put on the cluster or to replace a crashed node. The hosting is spread across the entire Cloud and each server is generally assigned to perform a single task meaning that a single web hosting package will never be relying on a single server in a cloud hosting environment.

I'm sorry if this seems harsh but words mean nothing, it results that count. This is destroying my business.

My ticket is still showing as unassigned and coming up to 6 hours and still no resolution.

Jonathan Hamon begin_of_the_skype_highlighting*****end_of_the_sky pe_highlighting

**Sebastian** · #7 (**permalink**) 19-05-10, 09:46 PM

Hello,

There was no problem with HA and FT configuration. The problem was with the routing rules which were swapped on the router, our network engineers solved this issue. You should able to access your VM now, we have already verified all VMs and they are responding normal now. Earlier we were able to access all VMs from some US based ISPs and Virgin Network.

Our network engineers are going to find out the root cause of this issue in next 12 -24 hours. We will keep you updated.

solarsystems · #8 (**permalink**) 05-06-10, 02:12 AM

hi sipero how are you finding your Cloud service now?
I have been facing several issues since moving from a semi dedicated server and still not that happy.
In the begging the server move processed was never completed properly and took me over 40 hours to get back up and running due to very poor service and planning from webhostinguk.
I now face issues where RDP and plesk are down and no one knows why, the best bit is they say the server is not down!! Well without RDP or Plesk how else can I use it?
When plesk is down the emails are down and I host for 40 sites so I get 40 angry customers calling.
I sometimes find the support desk would rather blame something you have done rather than just get on and fix it.
Not sure what I should do from here on?

sipero · #9 (**permalink**) 05-06-10, 09:55 PM

Hi,

There were a few further problems since my last posting but for around the last 10 days things seem to have been stable.

I just it stays this way. Previously for 3 years these sites were being hosted on a low spec dedicated server with 1 gb ram. I expected this cloudvps with 2 gb ram to be substantially superior.

**John** · #10 (**permalink**) 06-06-10, 10:40 PM

Quote:

Originally Posted by solarsystems

hi sipero how are you finding your Cloud service now?
I have been facing several issues since moving from a semi dedicated server and still not that happy.
In the begging the server move processed was never completed properly and took me over 40 hours to get back up and running due to very poor service and planning from webhostinguk.
I now face issues where RDP and plesk are down and no one knows why, the best bit is they say the server is not down!! Well without RDP or Plesk how else can I use it?
When plesk is down the emails are down and I host for 40 sites so I get 40 angry customers calling.
I sometimes find the support desk would rather blame something you have done rather than just get on and fix it.
Not sure what I should do from here on?

Hello,

Just about every "downtime" report I have come across on our Cloud Hosting has been a result of over utilization of the allocated resources.

Do you have a ticket? If so, please PM me the ticket ID and I will get the head of our VPS Department to look into it for you.

sipero · #11 (**permalink**) 06-06-10, 10:56 PM

Hi,

That may well be the case but in my case what it does not explain is how my sites were running fine on a lower spec CPU with 1 Gb ram and when I migrated to this platform I chose the high end package with 2 Gb ram but still had all these problems.

Jonathan Hamon

**John** · #12 (**permalink**) 06-06-10, 11:04 PM

Quote:

Originally Posted by sipero

Hi,

That may well be the case but in my case what it does not explain is how my sites were running fine on a lower spec CPU with 1 Gb ram and when I migrated to this platform I chose the high end package with 2 Gb ram but still had all these problems.

Jonathan Hamon

Hi,

If you have an open ticket, please message me the ticket ID.

solarsystems · #13 (**permalink**) 10-06-10, 09:23 AM

Ah down again and thought I would just update this 100% uptime service thread.
The online support guys are just taking on board that the issue but saying "allow me some time to check " leaving me to twidle my thumbs while i get loads of voicemail from angry customers. Come on you must understand that downtime on a 100% uptime server is not on.

solarsystems · #14 (**permalink**) 10-06-10, 09:30 AM

Ah a response
-------------------------
Philip S: there seems to network problem at DC the DC engieers are looking into the issue further.

garry: ok but thought there was a fall back on your cloud servers?

Philip S: i am not able to tell you an exact time now

Philip S: the whole DC network have problem

Philip S: there eingeers are looking into the issue

Philip S: Please allow us some time to check
--------------------------

Not really a 100% uptime service.

Anyone know who I should contact to dispute the SLA

**John** · #15 (**permalink**) 10-06-10, 09:35 AM

Hello,

It is still being looked into. What happened shouldn't be able to happen. It looks like a number of power bars had a fuse fitted which was under the bars rated power. These came direct from a reputable supplier, so it is quite concerning.

However, all services are now back online and have been for the past 10 or so minutes. Please come onto live chat if you are still having problems.