cannot make more than one spot request using boto3 in one day

Question

Greeting, My code work as long as its the first spot request of the day. If I terminate the instance and make another spot request It just gives me back my old request.

Is there something with my code or with AWS ??? Is there work-around ??

I have tried to clone my AMI and then use the clone AMI change the price or change the number of instance in the spec but it is still not working ???

!/home/makayo/.virtualenvs/boto3/bin/python

"""

http://boto3.readthedocs.io/en/latest/reference/services/ec2.html#EC2.Client.describe_spot_instance_requests

"""

import boto3
import time

myid=

s = boto3.Session()
ec2 = s.resource('ec2')
client = boto3.client('ec2')
images = list(ec2.images.filter(Owners=[myid]))

def getdate(datestr):
    ix=datestr.replace('T',' ')
    ix=ix[0:len(ix)-5]
    idx=time.strptime(ix,'%Y-%m-%d %H:%M:%S')
    return(idx)
zz=sorted(images, key=lambda images: getdate(images.creation_date))

#last_ami
myAmi=zz[len(zz)-1]
#earliest
#myAmi=latestAmi=zz[0]

"""
[{u'DeviceName': '/dev/sda1',  u'Ebs': {u'DeleteOnTermination': True,   u'Encrypted': False,   u'SnapshotId': 'snap-d8de3adb',   u'VolumeSize': 50,   u'VolumeType': 'gp2'}}]
"""



#myimageId='ami-42870a55'
myimageId=myAmi.id
print myimageId
mysubnetId=    myinstanceType='c4.4xlarge'
mykeyName='spot-coursera'
#make sure ajust this but dont do multiple in a loop as it can fail!!!
mycount=2
#make sure ajust this but dont do multiple in a loop as it can fail!!!
myprice='5.0'
mytype='one-time'
myipAddr=
myallocId=''
mysecurityGroups=['']
#mydisksize=70
mygroupId=
#mygroupId=
myzone='us-east-1a'
myvpcId='vpc-503dba37'
#latestAmi.block_device_mappings[0]['Ebs']['VolumeSize']=mydisksize
#diskSpec=latestAmi.block_device_mappings[0]['Ebs']['VolumeSize']
response2 = client.request_spot_instances(
            DryRun=False,
                SpotPrice=myprice,
                    ClientToken='string',
                        InstanceCount=1,
                            Type='one-time',
                                LaunchSpecification={
                                        'ImageId': myimageId,
                                                'KeyName': mykeyName,
                                                'SubnetId':mysubnetId,
                                                        #'SecurityGroups': mysecurityGroups,
                                                                'InstanceType': myinstanceType,
                                                                        'Placement': {
                                                                                        'AvailabilityZone': myzone,
                                                                                                }

                                                     }
                                         )

#print(response2)
myrequestId=response2['SpotInstanceRequests'][0]['SpotInstanceRequestId']

import time
XX=True
while XX:
    response3 = client.describe_spot_instance_requests(
        #DryRun=True,
        SpotInstanceRequestIds=[
        myrequestId,
    ]
    #Filters=[
     #   {
      #      'Name': 'string',
       #     'Values': [
        #        'string',
         #   ]
        #},
    #]
    )
    #print(response3)
    request_status=response3['SpotInstanceRequests'][0]['Status']['Code']
    if(request_status=='fullfilled'):
        print myrequestId,request_status
        XX=False;
    elif ('pending' in request_status):
        print myrequestId,request_status
        time.sleep(5)
    else:
        XX=False
        print myrequestId,request_status


"""
instances = ec2.instances.filter(Filters=[{'Name': 'instance-state-name', 'Values': ['running']}])
while( len(list(instances))==0):
    instances = ec2.instances.filter(Filters=[{'Name': 'instance-state-name', 'Values': ['running']}])

instances = ec2.instances.filter(Filters=[{'Name': 'instance-state-name', 'Values': ['running']}])
for instance in instances:
    print(instance.id, instance.instance_type);
    response = instance.modify_attribute(Groups=[mygroupId]);
    print(response);

Answer 1

This is wrong:

ClientToken='string',

Or, at least, it's wrong most of the time, as you should now realize.

The purpose of the token is to ensure that EC2 does not process the same request twice, due to retries, bugs, or a multitude of other reasons.

It doesn't matter (within reason -- 64 characters, max, ASCII, case-sensitive) what you send here, but you need to send something different with each unique request.

A client token is a unique, case-sensitive string of up to 64 ASCII characters. It is included in the response when you describe the instance. A client token is valid for at least 24 hours after the termination of the instance. You should not reuse a client token in another call later on.

http://docs.aws.amazon.com/AWSEC2/latest/APIReference/Run_Instance_Idempotency.html