Questions tagged with AWS Auto Scaling
Content language: English
Sort by most recent
We are using Opensearch cluster with 5 Data node and 3 master Node. Before some days, I have added a new node into my OpenSearch cluster, but still the node upgradation process is running. I have checked in my kibana and found that there is 5 node storing the data and new nodes don't have any data. Is there any way i can distribute the data on all newly added node
Hello, I'm creating a launch template and autoscaling group through cloud formation. The auto-scaling group is currently pointed to the DefaultVersion of the launch template, which is version 1. The latest version of the launch template is version 2. I updated the default version of the launch template to version 3, but I don't see it reflected in the auto-scaling group. I know we update the launch template version through instance refresh, but I would like to know if this is expected behavior to still see version 1 in the autoscaling group even though the default version is now version 2. I want to handle the updates through cloud formation. The problem I'm trying to solve is how do I push changes through cloud formation and handle the updates gracefully in the auto-scaling group. For example - I have two autoscaling groups using the same launch template. Then I decide to update one auto-scaling group to have a different ami. There will be two versions of the launch template, version 1 will have the old ami, and version 2 will have the latest ami. I'd like to gracefully roll out the changes to the autoscaling group 2 through cloud formation. Question 1 - How can I change the auto-scaling group through cloud formation and replace instances gracefully? Question 2 - If I change the version number in the cloud formation template, does it terminate instances first and then redeploy the new ec2 or the other way around?
Hello! I'd like to configure a Auto Scaling Group to use only spot instances and if doesn't have any spot available, the ASG will run on demand instance. How can I do this? Is there some native feature to do that? I was wondering to set the instances distribution to 100% Spot, but I think that will not work. ![Enter image description here](/media/postImages/original/IM74ovTXOrR5WALGYw5nb2XA) thanks for all!
I have an autoscaling group that works with a target-tracking policy and scales out after my instance hits 75% CPU. But I wanna know how much will it take to scale in or terminate the instance made by the autoscaling group right now it's 25 min which is a lot and can I set a custom time for my instance to be deleted during scale in the process? Thanks
Hello, I have a task that I have to do on my work and I would really appreciate if anyone can assist me on this. The assignment that I have to implement, is to create an 'Auto scaling group' using a value that is returned from a 'lambda function'. After an exhausting research, I have found that i have to Create an 'AutoSG', then select 'Predictive policy' and then choosing 'Custom cloudWatch Metric'. From their somehow i have to select the metric that came from lambda function. My problem their is how to construct 'JSON-formatted payload that contains the custom metric definition on Predictive Policy.' Moreover, do I need to add some additional code in my lambda function to send the returned value to 'Auto scaling Group'? Forgot to mention that my lambda function will executing a query on AWS RDS(mySQL) and return that value. Relative resources: https://docs.aws.amazon.com/autoscaling/ec2/userguide/predictive-scaling-customized-metric-specification.html https://aws.amazon.com/blogs/containers/autoscaling-amazon-ecs-services-based-on-custom-cloudwatch-and-prometheus-metrics/
I have to create an auto scaling group in two regions and the only difference between the two are the subnets. us-east-1 has 1 subnet whereas us-east-2 has two. How can I use the condition to call the subnet value from region map. Here's my broken code. Any help is appreciated. ``` Mappings: RegionMap: us-east-1: AMI: "ami-066f487d3b6819b0d" Subnet1: "subnet-0e6f12f64042ea5b1" us-east-2: AMI: "ami-0aef5e0adcbc7cc0f" Subnet1: "subnet-0e6f12f64042ea5b1" Subnet2: "subnet-0bc661bb8d98f3f03" Conditions: region: !Equals [!Ref us-east-2, Subnet2] autoscaling: Type: AWS::AutoScaling::AutoScalingGroup Properties: AutoScalingGroupName: asg1 VPCZoneIdentifier: - !FindInMap [RegionMap, !Ref "AWS::Region", Subnet1] - If regions = us-east-2 then !FindInMap [RegionMap, !Ref "AWS::Region", Subnet2] # This is what I need to figure out ``` I couldn't find any examples of this. Has anyone used a region map and used conditionals with it?
if I changed Ec2 type to Spot instances in autoscaling that is already deployed, does EC2 stop service while it is changing?
I used ECS with EC2 not fargate and also have autoscaling as we used gpu instances, we are considering to change ec2 type as spot instance. I have question that if i do change instance as spot, the ec2 ,which is belong to autoscaling now, could be changed to spot without stop or service stop
Hello, recently, I use Windows 2019 of EC2 instances and using launchconfigurations for AutoScaling. I used own tool for deploy for ec2. Is there anyway I can make AMI and apply to ASG and also exchanging old ec2 to new AMI image automatically? I thought I could use 2 lambda to make it happend, but it didnt go well. My scnerio is deploy app -> API gateway -> lambda 1 ( making AMI) -> when the status chaged to 'available' -> trigger lambda 2 -> changing AMI image of ASG and refresh. anyone have experience of this?
Proceed as follows: step 1: go to Launch configurations console. step 2: choose a Launch configuration. step 3: copy to launch template > Copy selected > Copy. I got an error message: "The security group ID 'awseb-e-yy32fjb***-stack-AWSEBSecurityGroup-RKGBZ9JQ****' is not valid. The expected format is sg-xxxxxxxx or sg-xxxxxxxxxxxxxxxxx." After investigating, I realized that AWS is taking "Security name" data instead of "Security ID" when copying Launch configuration to launch template. as in my case above, 'awseb-e-yy32fjbzmp-stack-AWSEBSecurityGroup-RKGBZ9JQ****' is the security group name, AWS sends this data as security group ID, so get above error.
The value of y-axis in the Predictive scaling policy Graph (under Auto scaling group) is in thousands, but says Load (CPU) utilization. What metric is it? It says aggregated CPU usage across the group, but I have only 1 instance in my ASG (for testing purposes) but the graph shows about 2K of CPU utilization during peak. what does it mean, and what is the UNIT value for the y-axis of the graph?
I have added one TCP port in NLB for existing ASG instance, health check failing, request anyone please guide me what to check.
I deployed stable diffusion v2.0 by aws sagemaker, and create endpoint for real-time inference.(instance type is ml.g4dn.xlarge) Also, i used aws apigateway and aws lambda. my question is that concurrency of prediction process by invoked request. when i check cloudwatch log, i see that request is sequentially process.(one prediction finished, then next prediction proceeded) i expected that the requests are concurrently dealt with, but not. at real-time endpoint, there is no max concurrent options, then invokeendpoint always sequentially procedeed?? no way for making requests are parellely dealt with, except increase instance number?