Tag Archives: Vcloud Director

Cloud Computing and the E-commerce Industry

The unprecedented growth of the Internet and the phenomenal commercial opportunity it has unveiled, has propelled e-commerce to experience growth rates that have never been witnessed before. The web hosting features and resources required for e-commerce functionality call for uninterrupted performance, reliability, and data security among other factors. This is perhaps why we have all noticed the adoption of cloud computing by the e-commerce industry at an accelerated pace. However, not all cloud hosting service providers deliver services at the same level. This is why it is a good idea to conduct a quick assessment of your needs as an e-commerce enterprise when benchmarking various cloud delivery services and platforms against your needs. We present three important considerations. ecom

Choosing a Cloud Hosting Company

There are several talking points to consider when you feel it is time to take your e-commerce enterprise to the next progressive milestone by transitioning to the cloud. Some of these factors and considerations may not come to you as a surprise because of the frequent press they receive. Others may be interesting enough to peak your curiosity. They are all, however, quite significant and deserve equitable if not equal attention as you plan the migration.

Access Speeds of Cloud Hosting Servers

Experts describe the speed of access to web pages as the single most popular reason why e-commerce companies pursue the cloud migration pathway. Amazon increased its overall gross revenue by 1% for every 100 milliseconds of improvement in the speed with which its flagship website served up web pages. Inferior access speeds typically lead to traffic losses which invariably translate into lost revenue. According to a study conducted by the Aberdeen Group, page load delays cost e-commerce enterprises up to $117 million in lost sales annually. Shopping cart abandonment has also been linked to poor server speeds. Owing to its distributed computing capability, near instant scalability during peak times, and built-in redundancies that literally guarantee a zero downtime performance record, cloud servers maintain consistent access speeds regardless of how much traffic they are experiencing at any given point in time. This functionality is even more critical during the holiday shopping season in November and December each year when e-commerce activity is at an all time high. Many leading edge cloud vendors currently use Tier 3 a+ data centers to deliver optimal performance while fully supporting dedicated e-commerce applications such as shopping carts, inventory management software, CRM, live chat and help desk solutions simultaneously without any compromise to access speeds.

Security of Financial Data

Although traditional web hosting companies provide a level of data security that is industry compliant, cloud service providers have seemingly raised the bar through superior technology and self-governance. You will experience no difficulty in locating a cloud hosting company  is ISO 27001 compliant. Many cloud vendors also achieve SysTrust certification. For credit card and financial data processing in a fully secure and encrypted environment, the cloud hosting company you eventually select should be able to design PCI-DSS solutions. It should use the highest level of SSL encryption available which currently stands at 256 kb encryption. There are several other data security measures employed by most cloud hosting service providers such as biometric screening of personnel, two-factor authentication, and IDS which are now industry standards in the cloud.

The Trust Factor

Customer perception has a great deal to do with e-commerce success. If your enterprise can communicate to your customer community through newsletters and email alerts that they will never experience down time, always receive access to your product pages at consistent speeds, and never have to worry about their privacy and credit card data security, all thanks to state-of-the-art 21st century cloud technologies, chances are high that your customer attrition rates will seldom become a cause for concern all things being equal. Share the credentials of the cloud hosting vendor you eventually select with your customers as yet another confidence and trust building measure. Educate them about the multiple layers of protection the cloud provides them. You are sure to experience positive feedback from your customers sooner than later.

Concluding Thoughts

Managed cloud hosting services provided by qualified vendors allows e-commerce enterprises to focus on their core activities related to the sale of products and services on the Internet while technology related issues and challenges are handled by cloud vendors. Moreover, the pay-as-you-go model facilitates improved levels of resource management and usually generates long-term savings. It is therefore no wonder that the e-commerce industry continues to experience unbridled growth worldwide. In a recent study published by ComScore, Q1 2014 saw desktop e-commerce spending rise 12 percent year-over-year to $56.1 billion, marking the eighteenth consecutive quarter of positive year-over-year growth and fourteenth consecutive quarter of double-digit growth. mCommerce spending on smartphones and tablets added $7.3 billion for the quarter, up 23 percent vs. year ago, for a digital commerce spending total of $63.4 billion in the first quarter of 2014.

Has your e-commerce enterprise finally decided to connect with the cloud? What are some of the other factors you will take into consideration as you plan a cloud migration strategy? We would be very interested in hearing from you about your experience so far through your comments below.

Find out more about StratoGen

 

Science Logic Application Monitoring

StratoGen uses Science Logic Application Monitoring (SLAM) for its customers.  Customers have full access to this market-leading tool, plus StratoGen uses SLAM to help customers work through performance issues.

Extendibility

SLAM, like other monitor products, displays application metrics on dashboards.  But what makes this product different is the large collection of custom dashboards, called PowerApps, that the vendors of different applications and devices have written for Science Logic’s customers.  Plus the customer can program their own dashboards.

The customer can define custom dashboards by using the regular SNMP information output by many devices.  The customer can also program application calls using SQL, to query the database, or web services SOAP/XML, to either check the application for performance or execute actual transactions, to make sure the application is working correctly.

Features

The best way to illustrate this product is to show you some dashboards and list some of the available metric by product (e.g. Apache or EMC) and to list some of the available PowerPacks.

First we group application monitoring by type of software or hardware monitored:

Networking

Network monitoring relies mainly upon the SNMP reporting abilities of network devices and PowerPacks written by the vendor.  They show network performance and bottlenecks plus generate alerts whenever someone makes a configuration change.  Plus it can draw the topology on the dashboard.

Science Logic 1

 Here are just a couple of available metrics:

  • Cisco CM no bandwidth avail
  • Packets dropped

Server Health

Monitoring the servers is fairly standard stuff.  Service Logic also has a server monitoring product EM7 that we wrote about here http://www.stratogen.net/blog/science-logic-em7-server-monitoring-by-walker-rowe-guest-blogger-computer-technology-writer/

Science Logic 2

  • Busiest servers
  • CPU load

Custom App

You can monitor custom apps, meaning anything you have written yourself and deployed to an application servers or even purchased software, at the operating system level, the application server level, or by executing specific SQL queries, Hadoop MapReduce jobs, or calling the application using web services.

  • Processes running or not (meaning they are down)
  • App server performance
  • LDAP and Active Directory load balancing and performance
  • Database performance

Apache and IIS

Web server metrics include those listed below and the others you would normally see in an Apache, Nginx, or IIS server.

  • CPU load
  • Requests per second
  • HTTP connections

Microsoft Exchange and Outlook Web Access

For your own organization there is the need to make sure that Exchange is performing correctly and know how many messages outbound are being blocked by spam filters.  That would show the need to contact SpaumHaus.org or any of the other organizations that maintain domain blacklists. Plus a surge in outbound viruses would indicate an infection in your systems that might not have been flagged by your security monitoring.

  • Speed of delivery and connection failure
  • Messages rejected due to viruses and blacklists
  • Inbound spam surge in volume

Messaging

Messaging is the transport layer by which data objects are sent to the application that requested them.  This is used instead of direct JNDI lookups, RCP, COBRA, and DCOM.  If a queue builds up in the queue, then applications will be stuck waiting for data.

  • Number of messages in queue
  • Queue stopped
  • Change in configuration

DNS

DNS lookup time affects app performance. With Science Logic Application monitoring you can check latency for the DNS servers that you host or the DNS servers hosted by your ISP.  This could point to the need to perhaps switch to using a public DNS server like Google (8.8.8.8 and 8.8.4.4) if you find any latency issues.  It could also point out the need for your customers to do that as well or contact their ISP for performance issues.

  • Lookup time for A (address), MX(mail), and NS (name server records)

Storage

With Service Logic you can monitor cloud or local storage.  This lets you verify that your storage is operating at the speed in which you need it.  Monitoring points to failed disks, which must be replaced, poorly distributed replication strategies, disk under or overutilization, and controller health.  If works for solid state storage as well as HDD.

  • pool usage
  • disk controller performance
  • IOPs
  • Reads and writes
  • Write and read errors

PowerApps

A collection of PowerApps is a PowerPack. These re grouped by type of monitoring and the vendor or customer who wrote them.  A PowerApp includes alarms, dynamic dashboards, and metrics.  Customers can build their own PowerAPps to monitor their custom applications plus they can download those written by third-parties and software and hardware vendors from the ScienceLogic portal.  Most of these are written by software and hardware vendors tailored to their specific systems. We show a sample of what is available below.

First in this download screen on the portal you can see how PowerApps are grouped:

Science Logic 3

Here are some samples by vendor and by type of PowerPack

EMC

  • LUN performance
  • Storage processor performance
  • Raid group performance

F5 BIG-IP

  • Temperature
  • LTM service discovery
  • DCM caching

Security (All vendors):

  • Tipping Point
  • NetScreen (policy, session graph, VPN tunnel
  • Fortinet (disk usage, config, memory)

Science Logic 4

As you can see Science Logic Application Monitoring is a flexible product that works with lots of products out-of-the-box plus it can be extended by the customer and has been extended by vendors who have written PowerApps.

StratoGen makes extensive use of this product in their cloud services.

http://www.stratogen.net/products/vmware-hosting.html

Constantly Syncing Inventory

When performing an upgrade of vCloud Director 1.5 to 5.1 we ran into this issue to do with synchronisation.

Constantly Syncing Inventory

A vCloud Director cell may fail to finish the synchronisation with a vCenter Server.  This is an issue where vCloud Director is constantly stating ‘syncing inventory’ in the vCenters section of the system>Manage & Monitor page.
Syncing Inventory
You may find that a simple restart of the affected cell services may fix the issue.  If you are running a multi-cell environment you can do this by quiescing the currently active cell and then stopping and restarting the vCD services.

First disable the cell and pass the active jobs to the other cells.

Display the current state of the cell to view any active jobs.
#  /opt/vmware/vcloud-director/bin/cell-management-tool -u <USERNAME> cell --status

Then Quiesce the active jobs.
#  /opt/vmware/vcloud-director/bin/cell-management-tool -u <USERNAME> cell --quiesce true

Confirm the cell isn’t processing any active jobs.
#  /opt/vmware/vcloud-director/bin/cell-management-tool -u <USERNAME> cell --status

Now shut the cell down to prevent any other jobs from becoming active on the cell.
#  /opt/vmware/vcloud-director/bin/cell-management-tool -u <USERNAME> cell --shutdown

Then restart the services.
# service vmware-vcd restart

If you are not running multiple cells you can just restart the service but it will cause a loss of service during the restart.  A typical restart takes around 2-5 minutes.  You can monitor the progess of the restart by tailing the cell.log file.
# tail -f /opt/vmware/vcloud-director/logs/cell.log

Once it say’s 100%, it is done.

If restarting the services doesn’t help try rebooting the cell.  Use the same commands as above to pass active tasks over to the other cells first before rebooting.
When the cell restarts check and see if the cell will reconnect and finish the sync.  If not check the log /opt/vmware/vcloud-director/logs/vcloud-container-info.log.  Look for the following error.  ORA-01013: user requested cancel of current operation.

You can do this as follows.
# less /opt/vmware/vcloud-director/logs/vcloud-container-info.log

Then press / and type in “user requested cancel of current operation” to go to the location in the log where this entry is recorded.

The reason for ORA 1013: error can be:

  • caused by the user – actually canceling the operation
  • caused by a response to congruent errors
  • the result of timeouts

When  vCloud Director sync is taking place, after processing the updates vCD performs database insertions.  Sometimes while persisting these updates vCloud Director will stop the sync and restart it, hence the constant sync.

Here is how to get around the issue.
1. Take a snapshot of the cell.
2. Quiesce the services of the cells using the cell-management-tool and then stop the services with service vmware-vcd stop as described above.
3. Open vi and add this line to /opt/vmware/vcloud-director/etc/global.properties
database.defaultQueryTimeout=300

# vi /opt/vmware/vcloud-director/etc/global.properties

4. Start the vCloud Director services again.
# service vmware-vcd start

If you do the above for all cells then the setting should be applied.