3 ms·

Very cool looking service- My nervousness comes from a few points- It looks very new so far.. How does this generate it's statistics? Is it running the checks

by e1ven 18y ago

Very cool looking service- My nervousness comes from a few points- It looks very new so far.. How does this generate it's statistics? Is it running the checks back to a centralized server on their side some place? What's the capacity on that server- If we threw 1200 EC2 nodes to them tomorrow, would it stay up? What sort of service/support plans do they offer? It's free, but It'd be worthwhile to pay to be able to get someone on the phone 24/7. We pay Amazon's gold support, we'd pay yours if it meant you'd take my calls at 3am. We normally use Nagios in-house, which is a great application, but has problems adjusting to dynamic load. This looks very decent, but I worry about the uptime.

rjurney 18y ago
I used Scalr.net for my last startup, and one issue they had twice during the year I paid them $50 a month was... their entire site went down from a DDOS attack. Each time they upgraded their hardware and setup, but overall I think they were running like 98% uptime for the last year. Not quite good enough for the mainstream. Scalr literally handles dynamic scaling of your application, so it failing is a bit worse than this, though.
- numair 18y ago
  Did your instances suffer as a result of this downtime? I've been playing around with Scalr, haven't really seen any problems as of late... Would love more input from others who have used / are using the service.
  - rjurney 18y ago
    No, they just didn't scale during the outage. I do recommend Scalr, its great, but I also tell people that its bleeding edge.
  - sanswork 18y ago
    I use scalr at the moment(paid service) we have just under 20 instances spread among 6 farms. Been using them for about 3 months now with very few problems. The only major issue I've had is that their DB roles aren't great for large db's(>100gig) though I think they are working on an SBS one. As it is I use a modified role to allow SBS but have to muck around too much for comfort on start ups since the SBS partitions don't always mount at startup(sometimes up to 10 minutes later) and when they do that the DB init scripts fail which causes scalr to cycle the instance and start the whole process over again. That said I've only had to do that twice since I've used them.

by e1ven 18y ago

rjurney 18y ago

I used Scalr.net for my last startup, and one issue they had twice during the year I paid them $50 a month was... their entire site went down from a DDOS attack. Each time they upgraded their hardware and setup, but overall I think they were running like 98% uptime for the last year. Not quite good enough for the mainstream. Scalr literally handles dynamic scaling of your application, so it failing is a bit worse than this, though.

numair 18y ago
Did your instances suffer as a result of this downtime? I've been playing around with Scalr, haven't really seen any problems as of late... Would love more input from others who have used / are using the service.
- rjurney 18y ago
  No, they just didn't scale during the outage. I do recommend Scalr, its great, but I also tell people that its bleeding edge.
- sanswork 18y ago
  I use scalr at the moment(paid service) we have just under 20 instances spread among 6 farms. Been using them for about 3 months now with very few problems. The only major issue I've had is that their DB roles aren't great for large db's(>100gig) though I think they are working on an SBS one. As it is I use a modified role to allow SBS but have to muck around too much for comfort on start ups since the SBS partitions don't always mount at startup(sometimes up to 10 minutes later) and when they do that the DB init scripts fail which causes scalr to cycle the instance and start the whole process over again. That said I've only had to do that twice since I've used them.