|
|
|
Forums » List all forums » » Forum: General Discussion » » » Thread: Disaster recovery and redundancy |
| Posted by calliduscloudnexus at Sep 26, 2017, 4:35:47 PM |
|
Disaster recovery and redundancy Hi, What best practices and infrastructure do you have in place to manage redundancy if your primary server goes down and for disaster recovery if your primary data source is corrupted or destroyed? Thanks. |
| Posted by calliduscloudnexus at Sep 26, 2017, 7:03:49 PM |
|
Re: Disaster recovery and redundancy Also, we need a summary of backup processes and a description of your company's organization and leadership. It's okay if you want to send these offline. Thanks. |
| Posted by support at Sep 26, 2017, 10:18:06 PM |
|
Re: Disaster recovery and redundancy I'm happy to post a little infrastructure info here, b/c it's in our members' best interest to know how things work (to the extent is doesn't compromise security, of course) After about 17 years of operation running on physically colocated servers, this Spring localendar migrated to a cloud environment. This environment is maintained and managed by Leidos, formerly known as SAIC https://en.wikipedia.org/wiki/Leidos Our Master Service Agreement with Leidos states that we are guaranteed 99.5% uptime. However, that is just infrastructure uptime for our environment. It obviously doesn't cover outside network infrastructure that might affect your ability to connect to localendar and it definitely doesn't cover what happens if we do something like deploying a new release with a bug that causes an outage. Leidos provides us with multiple virtual machines as well as a series of daily backup images. We also take our own regular backups of the localendar databases for additional backup purposes and store them on our internal network. We do not do intra-day backups, which means in an outage their could be a theoretical loss of one day's data. In the event of a DR/BR scenario, Leidos has cloud infrastructure hosted in multiple locations. We currently maintain active virtual servers in 2 of these locations so that we could restore to an alternate environment on existing VMs, rather than waiting for Leidos to spin up new ones. We manage our DNS independently from Leidos (through Register.com) which means any outage that requires an IP change for localendar.com could take anywhere from minutes to days in a worse-case scenario (https://support.managed.com/kb/a604/dns-propagation-and-why-it-takes-so-long-explained.aspx) There are solutions to reduce this time, but they introduce other problems that we feel are not an appropriate risk. That's a quick overview. If you look at the registration of our name http://ewhois.org/www/localendar.com you will see we have been around for some time (Nov 2, 1999). A lot has changed on the web since then, and will continue to do so. We will make future infrastructure changes when we feel they are appropriate for the site. ---------------------------------------- Marc Higgins Support Associate, localendar.com Follow us on Twitter! http://www.twitter.com/localendar_news |
| Posted by support at Sep 26, 2017, 10:21:21 PM |
|
Re: Disaster recovery and redundancy Our founder is Michael Ogrinz: https://www.linkedin.com/in/ogrinz/ Beyond that, things get a little personal and we won't divulge much more (publicly or privately) to protect our staff's privacy in the same way that we protect and respect yours. Can I ask why you are requesting this information? ---------------------------------------- Marc Higgins Support Associate, localendar.com Follow us on Twitter! http://www.twitter.com/localendar_news |
| Posted by calliduscloudnexus at Sep 29, 2017, 6:24:34 PM |
|
Re: Disaster recovery and redundancy Thank you very much for the information. I will pass this onto my management. The reason why we need it is that is what we ask of all vendors - and what our customers ask of us - as far as security and infrastructure. If InfoSec walks down and asks about security for third-party applications and platforms that we use, we need to have answers. We also want to make sure that systems we use don't go end of life unexpectedly. This came up now because my new SVP has an Operations background, and he asked these questions. Thanks again. |
Help! | Cobranding | Legal | Privacy Policy | About localendar.com | Contact Us |