I've recently gotten scalr installed and able to launch a farm with the standard roles www, app and mysqllvm. In the farm config I specified 1 www, 1-2 app and 1 mysqllvm (for now).
Every time I launch the farm is spins-up more instances that are necessary - for example, 3 instances of www, two instances app and 1 instance of mysqllvm. I've increased the default timeout and the timeouts for each role to be 600s. Looking at the event logs: 20-05-2009 15:16:02 INFO Cloud PollerProcess Farm Cloud, role app scaling down. i-b7d2a9de will be terminated in 44.1 minutes. Launch time: 2009-05-20T15:10:07.000Z, Current time: 2009-05-20 15:16:02 20-05-2009 15:15:03 INFO Cloud PollerProcess Farm Cloud, role app scaling down. i-b7d2a9de will be terminated in 45.1 minutes. Launch time: 2009-05-20T15:10:07.000Z, Current time: 2009-05-20 15:15:03 20-05-2009 15:14:20 INFO Cloud i-b7d2a9de/trap-hostup.sh 10.252.178.9 UP. Scalr notified me that 10.252.178.9 of role app (Custom role: app) is up. 20-05-2009 15:14:19 INFO Cloud EventObserver Instance i-7dd1aa14. Is DB Master = 0 20-05-2009 15:14:14 INFO Cloud i-7dd1aa14/trap-inithost.sh Received INIT trap from UDP: [174.129.119.97]:32979 (AWS account ID: 273151577831). Finishing host start-up. 20-05-2009 15:14:11 INFO Cloud i-7dd1aa14/instance-init.sh Host 10.252.178.9/app initialized. Awaiting authentication data. 20-05-2009 15:14:03 WARN Cloud PollerProcess Disaster: No instances running in role mysqllvm! 20-05-2009 15:13:28 INFO Cloud EventObserver Instance i-b7d2a9de. Is DB Master = 0 20-05-2009 15:13:28 INFO Cloud EventObserver Adding system A records to zone cloud.apez.biz 20-05-2009 15:13:24 INFO Cloud i-b7d2a9de/trap-inithost.sh Received INIT trap from UDP: [174.129.119.97]:32979 (AWS account ID: 273151577831). Finishing host start-up. 20-05-2009 15:13:20 INFO Cloud i-b7d2a9de/instance-init.sh Host 10.254.91.131/app initialized. Awaiting authentication data. 20-05-2009 15:13:02 WARN Cloud PollerProcess Disaster: No instances running in role www! 20-05-2009 15:12:03 WARN Cloud PollerProcess Disaster: No instances running in role www! 20-05-2009 15:11:02 WARN Cloud PollerProcess Disaster: No instances running in role www! 20-05-2009 15:10:02 WARN Cloud PollerProcess Disaster: No instances running in role app! 20-05-2009 15:09:04 WARN Cloud PollerProcess Disaster: No instances running in role app! Any idea why it does this? It also seems to confuse itself with the multiple www roles and the single elastic IP address, finally leaving the DNS zone records for www.%hostname% pointing to the dynamic external address instead of the elastic one (but reporting the elastic one on the scalr pages). Using the edit zone function and hitting save causes it to update correctly though. Making progress :) (next I plan to attempt to customize the www load balancer role to use Apache with mod_proxy and mod_rewrite as our application requires some mod_rewrite-based traffic routing). Cheers. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "scalr-discuss" group. To post to this group, send email to scalr-discuss@googlegroups.com To unsubscribe from this group, send email to scalr-discuss+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en -~----------~----~----~----~------~----~------~--~---
