VMHosts

VMHosts http://www.vmhosts.co.uk "Cloud hosting within reach" Thu, 16 May 2013 08:51:42 +0000 en-US hourly 1 http://wordpress.org/?v=3.5.1 16/05/2013 – Planned Generator Maintenance Notification http://www.vmhosts.co.uk/16052013-planned-generator-maintenance-notification/ http://www.vmhosts.co.uk/16052013-planned-generator-maintenance-notification/#comments Thu, 16 May 2013 08:51:42 +0000 admin http://www.vmhosts.co.uk/?p=1254
Why — Pulsant follow a stringent Planned Maintenance regime. This work forms a major part of this regime and regular maintenance is done to ensure all services are kept to the highest standard.

When —- The planned maintenance work will take place on the following date, no interruption to service is expected, but all maintenance carries an “At Risk” as the level of resiliency during maintenance may be reduced. Only one Generator will be worked on at any time. The planned maintenance window is planned for 09:00am to 17:00pm on Wednesday 22nd May 2013.]]> http://www.vmhosts.co.uk/16052013-planned-generator-maintenance-notification/feed/ 0 10/05/13 Host connectivity issues affecting some VMs – RESOLVED http://www.vmhosts.co.uk/100513-host-connectivity-issues-affecting-some-vms/ http://www.vmhosts.co.uk/100513-host-connectivity-issues-affecting-some-vms/#comments Fri, 10 May 2013 13:34:37 +0000 admin http://www.vmhosts.co.uk/?p=1241 Date: 10/05/13
Time first reported: 14:00
Time resolved: 14:50

Details: We have experienced an issue with connectivity to one from one of our physical hosts, which is affecting some customers VMs

Further Action: A hardware failure on one of our hosts caused a fail-over to occur. The hardware failure was quickly identified and a fix put in place. All customer VMs running on that host successfully failed over to other hosts in the cluster. All customer VMs will be kept off the host as a precautionary measure, until we are satisfied that the host can be returned to production.]]> http://www.vmhosts.co.uk/100513-host-connectivity-issues-affecting-some-vms/feed/ 0 VMware HP Storage CIM agent Warning http://www.vmhosts.co.uk/vmware-hp-storage-cim-agent-warning/ http://www.vmhosts.co.uk/vmware-hp-storage-cim-agent-warning/#comments Tue, 30 Apr 2013 16:11:58 +0000 admin http://www.vmhosts.co.uk/?p=1223 Hardware Status

We contacted HP who found an issue with the HP-smx-provider. Here is the fix provided by HP:

Step 1

Remove the below “VIBs” from the local ESXi host by typing “esxcli software vib remove –n hp-smx-provider” for each vib listed below.

hp-smx-provider
char-hpilo
char-hpcru
hp-ams
hpacucli
hpbootcfg
hponcfg
hpnmi

Step 2

Reboot the host after removing the VIBs and download the older version of ESXi 5 offline bundle and the version is 1.3-12 (4 Sep 2012). For more information Click Here.

Step 3

Reboot the ESXi HOST once again after installation of the Offline Bundle and check the status in the vCenter.]]> http://www.vmhosts.co.uk/vmware-hp-storage-cim-agent-warning/feed/ 0 04/04/2013 – Routine UPS maintenance notification – Clear http://www.vmhosts.co.uk/04042013-routine-ups-maintenance-notification/ http://www.vmhosts.co.uk/04042013-routine-ups-maintenance-notification/#comments Thu, 04 Apr 2013 16:19:05 +0000 admin http://www.vmhosts.co.uk/?p=1215
In keeping with the manufacturer’s recommendations, and the Pulsant planned and preventative maintenance regime, we will be carrying out routine UPS maintenance at our MHD1 Datacentre.

The first portion of maintenance involves the replacement of AC/DC capacitors on one UPS module. This UPS module forms only part of a resilient UPS array, so power to equipment won’t be interrupted during the capacitor replacement process. However, system redundancy will be reduced temporarily during the works and therefore power will be deemed “at risk” for a portion of the maintenance window.

Once the capacitor replacements have been completed, we will commence the second part of the process, which involves routine preventative maintenance on the UPS array. No outage is expected, however there will be increased risk while this work is carried out. Pulsant’s specialist UPS contractors will be on site to complete the capacitor replacements and maintenance, supervised by the Datacentre management team throughout.

The work will take place during the following time window: Tuesday 9th April, commencing at 21:00BST and ending by 05:00BST the following morning.]]> http://www.vmhosts.co.uk/04042013-routine-ups-maintenance-notification/feed/ 0 Uses for Virtual Machine Hosting http://www.vmhosts.co.uk/uses-for-virtual-machine-hosting/ http://www.vmhosts.co.uk/uses-for-virtual-machine-hosting/#comments Tue, 26 Mar 2013 17:05:04 +0000 admin http://www.vmhosts.co.uk/?p=1187

Email services including Microsoft Exchange
Remote Desktop Services
Hosting high traffic websites
SQL applications
Document management including Microsoft SharePoint services
DNS hosting
VOIP Services
VPN Server
E-commerce

As you might expect there are many more uses for virtual private servers, these are just a few common examples, but rest assured, if I can be installed, you bet it can almost certainly be hosted by us.]]> http://www.vmhosts.co.uk/uses-for-virtual-machine-hosting/feed/ 0 Load balancing CAS servers in Exchange 2013 http://www.vmhosts.co.uk/load-balancing-cas-servers-in-exchange-2013/ http://www.vmhosts.co.uk/load-balancing-cas-servers-in-exchange-2013/#comments Mon, 25 Mar 2013 17:38:37 +0000 admin http://www.vmhosts.co.uk/?p=1141 Exchange 2010 and the CAS array

I’m going to start this blog on the assumption that you have some understanding about what a CAS array in Exchange 2010 is and what it does. But If not, then don’t worry, I’m going to sum it up quickly now.

The CAS array, which was introduced in Exchange 2010, was a way of grouping one or more client access servers in an active directory site together, so that they could be seen and accessed by clients as a single entity. In other words, a CAS array had one IP address and FQDN DNS name. Usually a hardware (or software) load balancer was also configured to allow for effective load balancing between the servers in the CAS array. Windows NLB could also be used to provide load balancing, however a dedicated hardware load balancer was, and still is recommended.

This was an excellent addition to Exchange and something that I always implement when designing a new Exchange 2010 topology.

CAS arrays essential information

Without a load balancer as mentioned above, a CAS array will not load balance traffic
CAS arrays are active directory site specific. You can’t have a CAS array that spans more than one AD site
You can configure a CAS array even if you only have one CAS server. In fact I’d recommend this – it makes adding additional CAS server into the mix very easy!
No need to put the CAS array FQDNs on the SSL certificate
Configure CAS arrays before configuring new databases (if possible)

A Problem with site resiliency..

Whilst CAS arrays are excellent at protecting against a failure of a single CAS server, if a whole site was to fail then some manual intervention was needed to provide fail-over. DNS records of the CAS array would have to be manually updated to point to a functioning CAS server (or array) at another site. And if you forgot to set your time to live (TTL) value to a low number then it could be at least an hour before all the clients get the updated IP address.

No more CAS Arrays with Exchange 2013

So, step forward Exchange 2013. In Exchange 2013 some changes have been made to address this problem and make it easier for the administrator to work with.

This was done partly by removing the concept of the CAS array. I know I’ve spent the first half of this blog waxing lyrical about the CAS array, but it’s no longer needed anymore – and I’ll tell you why.

Goodbye MAPI – Hello HTTPS

In previous versions of Exchange, outlook connections across the internal network took place by establishing a RPC MAPI connection to the mailbox servers. In Exchange 2010 the CAS servers handled these connections on behalf of the clients, so that the clients didn’t need to connect directly to the mailbox servers anymore. Whilst this was an improvement on previous versions, it meant that the CAS servers and the Mailbox servers were joined at the hip so to speak.

In Exchange 2013, all outlook client connectivity is made via RPC-over-HTTPS (or OutlookAnywhere). The main (huge) benefit of this is that now the CAS and Mailbox servers can act independently from each other. Now when a client connects to a CAS server, the CAS server queries active directory and finds the relevant Mailbox server.

So how does this affect site resilience?

With Exchange 2013, a single namespace can have multiple VIP (Virtual IP addresses). For example the namespace mail.company.com could have the every IP address of every CAS server attributed to it. In the event of a CAS servers IP address becoming unavailable, the clients will automatically fail over to a working IP address! Sure fail-over takes about 20 seconds, but who’s going to notice that in the real world??

The below diagram illustrates a 2 site example of this site resiliency

Hold on I though this article was about Load balancing Exchange 2013?!

Well it still is… Hardware load balancers will still play an important part in making the CAS servers highly available as they are able to detect when a CAS server has failed and will stop servicing clients, however with the Exchange 2013 design changes the configuration of the load balancers just got a whole lot easier J ]]> http://www.vmhosts.co.uk/load-balancing-cas-servers-in-exchange-2013/feed/ 0 13/03/2013 at risk notification http://www.vmhosts.co.uk/13032013-at-risk-notification/ http://www.vmhosts.co.uk/13032013-at-risk-notification/#comments Wed, 13 Mar 2013 16:20:34 +0000 admin http://www.vmhosts.co.uk/?p=1127
What —- Pulsant will be performing core network maintenance, this is an at risk notification, your service should be unaffected. Services at risk: All South London, Milton Keynes, Maidenhead and Newbridge based Static & BGP transit.

Why — The maintenance is required as part of on-going network improvements

When —- The maintenance will take place on the following dates / times: 20/03/2013 – 22:00 – 23:00 21/03/2013 – 22:00 – 23:00]]> http://www.vmhosts.co.uk/13032013-at-risk-notification/feed/ 0 Connectivity Issue 08/03/13 – resolved http://www.vmhosts.co.uk/connectivity-issue-080313/ http://www.vmhosts.co.uk/connectivity-issue-080313/#comments Fri, 08 Mar 2013 17:48:19 +0000 admin http://www.vmhosts.co.uk/?p=1085 Date: 08/03/2013
Time first reported: 16:54
Time service restored: 17:40

Details: Some customer VM’s have had experienced stability problems. When VM’s were restarted they were unable to start.

Latest Update: The issue has now been resolved. Please check your VM services, some VM’s may require a restart.

Cause: One of the storage nodes servicing read requests caused inconsistencies. This appears to have been caused by a faulty Raid Card. All writes are mirrored to a second storage node to keep data safe. We forced the IO connections to the partner node. Read IO requests are now in a consistent state and VM’s are now responding, although some may need a reboot depending on the read activity prior to the failure.

Although hardware failures are very rare they can happen. We are extremely sorry for any inconvenience caused. Normally the system would have failed over all storage requests to the remaining storage nodes but in with this type of failure the system was unable to detect the read inconsistencies. Due to the storage technology we are using we were able to recovery quickly and without any long restore requests.

If you have any questions or are still experiencing any problems with connectivity please contact our support department ]]> http://www.vmhosts.co.uk/connectivity-issue-080313/feed/ 0 24/02/13 storage connectivity issues affecting some VM’s -Resolved http://www.vmhosts.co.uk/240213-storage-connectivity-issues-affecting-some-vms/ http://www.vmhosts.co.uk/240213-storage-connectivity-issues-affecting-some-vms/#comments Sun, 24 Feb 2013 16:21:31 +0000 admin http://www.vmhosts.co.uk/?p=1074 Date: 24/02/13
Time first reported: 13:44
Time service restored: 16:15

Details: We have experienced an issue with connectivity to one datastore from a single physical host. This caused some VM’s to power off or go into a non responding state. We have moved these VM’s to other hosts.

Further Action: We are continuing to troubleshoot the storage connectivity issues and will update this post with further details soon.

We have not be able to isolate the cause of the storage interruption. Our storage array showed active connections and connections to other volumes on the same array were fine. We have opened a ticket with the iSCSI HBA manufactures. This host is currently our of production ]]> http://www.vmhosts.co.uk/240213-storage-connectivity-issues-affecting-some-vms/feed/ 0 Potential Security Vulnerability found in Linux Systems http://www.vmhosts.co.uk/potential-security-vulnerability-found-in-linux-systems/ http://www.vmhosts.co.uk/potential-security-vulnerability-found-in-linux-systems/#comments Sat, 23 Feb 2013 14:37:00 +0000 admin http://www.vmhosts.co.uk/?p=1067

Information

According to recent reports there is a rootkit which leaves a backdoor on the system and gains the full root access to Linux-based servers.

The root cause of the rootkit is not known yet.
Please visit the following link for more information and to check if your system is vulnerable:
http://kb.parallels.com/en/115589]]> http://www.vmhosts.co.uk/potential-security-vulnerability-found-in-linux-systems/feed/ 0