High server load

[squidier]

Re: High server load

I've updated the status page. Please let me know if you still see any sluggishness with the servers.

My server experienced another high load this morning.

You have asked us to report back re sluggishness, we have - so now what?

Going by yesterday's experience, this slowness precedes the server grinding to a halt completely.

[woodpecker]

Re: High server load

I find the lack of response by eukhost disgraceful, no response here, by email or by ticket, I'm quite happy to start via social media if you prefer but last time I did that you got teary about it.

I'm looking at migrating elsewhere, had enough of this nonsense!

[squidier]

Re: High server load

Now 4 incidents of high load since 9am this morning....

[GarryK]

Re: High server load

Hi!

I've asked one of our Senior Administrators to look into this for you, they'll get back to you shortly.

Regards,
Garry K.

[cbjroms]

Re: High server load

I am also very disappointed with the lack of response from feedabck@ & Tech Support over the last few days. Keep being promised relies asap and the nothing for 24 hours. What is going on?

[ollybee]

Re: High server load

Hi,

I am a senior administrator at eUKhost, let me update you on what has been going on.

As you may know we are migrating our VMware servers to a new cluster. The bottom line is the SAN on the new cluster is not performing as expected or indeed in the same way it did during our testing phase. On Monday the storage required a complete restart and this caused 1.5 - 2hours outage realistically until all VM's where back up and running normally. Since then we have been suffering latency spikes which have been causing the high loads on the guest VM's. We are currently working with our storage vendor to find a solution to these issues. We are also putting together a plan B in case the issues can't be resolved to our satisfaction.

We have also suffered a denial of service attack on Tuesday morning which will have affected customers on other platforms. DDOS attacks are a fact of life but this one was affected customer more widely due to the way the network was configured to accommodate the VMware live migrations between data centers.

Let me apologise again for the outages and lack of information. Please be assured we are taking this seriously and working hard on a solution

Oliver

[RyanPS]

Re: High server load

Our monitoring system suggests that you have disabled the virtual agents required for your system to run effectively. Please speak to support who will be happy to install these virtual agents for you. If other readers are unsure if they have these virtual agents, please also contact the support team who can check whether or not these are installed and install them if not.

[woodpecker]

Re: High server load

"As you may know we are migrating our VMware servers to a new cluster"

We know since you collapsed our websites, but we didn't know, do you think we are psychic?

Current dire status of VMC5 I/O

root@cloudserver [~/ioping-0.6]# ./ioping -RD .

--- . (ext3 /dev/root) ioping statistics ---
863 requests completed in 3000.0 ms, 295 iops, 1.2 mb/s
min/avg/max/mdev = 0.6/3.4/237.9/9.5 ms

Other cloud platforms are in the region of circa 15000 iops on this test

[ollybee]

Re: High server load

Hi,

We know since you collapsed our websites, but we didn't know, do you think we are psychic?

Fair comment. We didn't mail out all customers since the migrations would be transparent. My formal job title is Enterprise support manager, I had discussed this with some of my customers and many others have come to know since but I accept you may not have done.

We had been migrating customers to the new cluster for over 2 weeks before these issues came to light and everything was running smoothly up to that point. It's frustrating for us as we had used similar tools to ioping to run a synthetic workloads on the storage from test VM's across the new hosts to simulate the workload from the original VMware cluster, there were no issues during testing either. The system has plenty of capacity as it was designed designed with growth, performance and redundancy in mind.

It's taking time to try and zero in on where the problems lies as everything is new compared to the old cluster and the issues can't be reproduced on demand.

Oliver

[Site]

Re: High server load

How is it looking Oliver?

Are the servers in VMC5 performing better now?

[Rock]

Re: High server load

How is it looking Oliver?

Are the servers in VMC5 performing better now?

Hi Site!

The servers are all stable & performing much better now.

[squidier]

Re: High server load

I have problems again with VM5 - Can't get any help from support, more than 24 hours of being fobbed off now. 2 live chats, 2 tickets, server still down.
Please help?....

[squidier]

Re: High server load

36 hours down now. Support keep rebooting server but refuse to work out what the underlying cause is.

[Ryan]

Re: High server load

36 hours down now. Support keep rebooting server but refuse to work out what the underlying cause is.

I'd request you to share the ongoing ticket IDs so I can escalate them to one of our senior techs for a thorough check.

[squidier]

Re: High server load

I'd request you to share the ongoing ticket IDs so I can escalate them to one of our senior techs for a thorough check.

xni-496-57631