http://ayende.comAyende @ RahienCopyright (C) Ayende Rahien 2004 - 2021 (c) 202660 Thanks for the book tip. Have you started implementing these "environment validators" yet? I can totally see the value in them, and I'm thinking about how I'd go about it. I've got various programs, written in various languages, across various servers. So, I'm not sure if a single service could do this? I'd probably want to write custom monitors that are scheduled to run frequently on each server, and which do a particular job. Each monitor could report back to one single place, using a well known format. As a metaphor, this is a bit like having various "security sensors" distributed around a building complex. This one place could even warn if reports aren't received when expected (i.e. one of the sensors isn't responding). Results would be aggregated and published via RSS/Email. http://ayende.com/3168/continuous-environment-validation#comment5http://ayende.com/3168/continuous-environment-validation#comment5Tue, 04 Mar 2008 20:58:32 GMTOren, I believe you are mixing together logically different concepts of Environment Validation and Monitoring/Notifications. EV could be a part of the system (it has to be) that just reports the state, while the monitoring/notification/contiuous thing is better to be completely external system that actually decides if the state is invalid, suspicious, good or whatever, and then takes the actions. I've expanded on that a little bit in my blog (http://rabdullin.com), since that is close to the concept of CI/automation engines for business solutions. And we had to implement that recently. Onur, Nagios is a specific software that deals with the "network problems", while Environment Validation is a broader concept that can deal with issues starting from IoC configuration and up to the business validation. Rinat http://ayende.com/3168/continuous-environment-validation#comment4http://ayende.com/3168/continuous-environment-validation#comment4Wed, 27 Feb 2008 20:33:19 GMTWe use ManageEngine® Applications Manager(http://manageengine.adventnet.com/products/applications_manager/), which has worked very well for us. http://ayende.com/3168/continuous-environment-validation#comment3http://ayende.com/3168/continuous-environment-validation#comment3Wed, 27 Feb 2008 19:55:49 GMTAyende , don't re invent the wheel : http://www.nagios.org/ :-) http://ayende.com/3168/continuous-environment-validation#comment2http://ayende.com/3168/continuous-environment-validation#comment2Wed, 27 Feb 2008 19:01:42 GMTI'll add a few items * The alert system itself is functioning * Windows Services are all running (e.g. SharePoint has 3 services running at all times, minimum. Sometimes they're not running.) * Service accounts are functioning * Servers can communicate with each other (this has changed on me, in production! OOPS) * Servers can query AD * Servers can send email via SMTP (again this has burned me in the past, firewall/antivirus updates) I particularly want to emphasize the "make sure alert system works" because I've seen our "emergency pager system" go down, regularly, for hours at a time, without notice from the pager company. http://ayende.com/3168/continuous-environment-validation#comment1http://ayende.com/3168/continuous-environment-validation#comment1Wed, 27 Feb 2008 16:55:14 GMT