OpenStack Compute (nova)

nova returns no status code for state scheduling

Bug #721254 reported by bobya on 2011-02-18

This bug affects 1 person

Affects		Status	Importance	Assigned to	Milestone
	OpenStack Compute (nova)	Fix Released	Medium	Ken Pepple	OpenStack Compute (nova) 2011.2 "cactus"

Bug Description

We are trying to run new instances through AWS JAVA SDK. the instance starts normally, but the response from the server breaks the API . Here is the exact exception:

578 [main] ERROR com.amazonaws.http.HttpClient - Unable to unmarshall response (For input string: ""):
java.lang.NumberFormatException: For input string: ""
at java.lang.NumberFormatException.forInputString(NumberFormatException.java:48)
at java.lang.Integer.parseInt(Integer.java:470)
at java.lang.Integer.parseInt(Integer.java:499)
at com.amazonaws.transform.SimpleTypeStaxUnmarshallers$IntegerStaxUnmarshaller.unmarshall(SimpleTypeStaxUnmarshallers.java:77)
at com.amazonaws.services.ec2.model.transform.InstanceStateStaxUnmarshaller.unmarshall(InstanceStateStaxUnmarshaller.java:51)
at com.amazonaws.services.ec2.model.transform.InstanceStaxUnmarshaller.unmarshall(InstanceStaxUnmarshaller.java:59)
at com.amazonaws.services.ec2.model.transform.ReservationStaxUnmarshaller.unmarshall(ReservationStaxUnmarshaller.java:67)
at com.amazonaws.services.ec2.model.transform.DescribeInstancesResultStaxUnmarshaller.unmarshall(DescribeInstancesResultStaxUnmarshaller.java:51)
at com.amazonaws.services.ec2.model.transform.DescribeInstancesResultStaxUnmarshaller.unmarshall(DescribeInstancesResultStaxUnmarshaller.java:33)
at com.amazonaws.http.StaxResponseHandler.handle(StaxResponseHandler.java:126)
at com.amazonaws.http.StaxResponseHandler.handle(StaxResponseHandler.java:36)
at com.amazonaws.http.HttpClient.handleResponse(HttpClient.java:383)
at com.amazonaws.http.HttpClient.execute(HttpClient.java:157)
at com.amazonaws.services.ec2.AmazonEC2Client.invoke(AmazonEC2Client.java:3639)
at com.amazonaws.services.ec2.AmazonEC2Client.describeInstances(AmazonEC2Client.java:456)
at com.amazonaws.services.ec2.AmazonEC2Client.describeInstances(AmazonEC2Client.java:3061)
at com.sap.qa.Connect.main(Connect.java:112)

xmlns="http://ec2.amazonaws.com/doc/2010-06-15/"><requestId>AZ8DPR5PW3NEDR52MI7W</requestId><ownerId>Main</ownerId><groupSet><item><groupId>default</groupId></item></groupSet><reservationId>r-dw1p8hys</reservationId><instancesSet><item><displayDescription/><displayName>Server 6</displayName><keyName>None (Main, None)</keyName><instanceId>i-00000006</instanceId><instanceState><code/><name>scheduling</name></instanceState><publicDnsName/><imageId>ami-crgu85uu</imageId><productCodesSet/><privateDnsName/><dnsName/><launchTime>2011-02-14 19:31:41.194843</launchTime><placement><availabilityZone>unknown zone</availabilityZone></placement><amiLaunchIndex>0</amiLaunchIndex><instanceType>m1.small</instanceType></item></instancesSet></RunInstancesResponse>

The exact problem in the response that breaks the API is that the return status code does not contain any number it is just <code/> and this doesn't pass the integer parser in the AWS SDK.

could you please change the behavior so that a status code is returned for every status. Thanks

Related branches

lp://staging/~ken-pepple/nova/bad-power-state-ec2

Merged into lp://staging/~hudson-openstack/nova/trunk at revision 785

Rick Harris (community): Approve on 2011-03-10

Jay Pipes (community): Approve on 2011-03-10

Ken Pepple (ken-pepple) on 2011-02-23

Changed in nova:
assignee:	nobody → Ken Pepple (ken-pepple)
status:	New → In Progress

Thierry Carrez (ttx) on 2011-02-23

Changed in nova:
importance:	Undecided → Medium

Revision history for this message

bobya (bobya) wrote on 2011-02-23:

Hi Ken,
I see you've taken this bug also :)
As it took some time we did some investigation and found out that the problem comes from the default options set at the beginning. They are located in "nova/compute/api.py"

We added a line like this:

.................
base_options = {
            'reservation_id': utils.generate_uid('r'),
            'image_id': image_id,
            'kernel_id': kernel_id or '',
            'ramdisk_id': ramdisk_id or '',
+++ 'state': '0',
            'state_description': 'scheduling',
.................

after that every new instance that we started had the code "0" for the scheduling state.

As I am not sure this is the correct way to fix it I am only suggesting here the method we found.

Revision history for this message

Ken Pepple (ken-pepple) wrote on 2011-03-09:

bobya -- sorry for the delay. yes, i believe that is the correct short term fix. pushing for merge into trunk.

OpenStack Infra (hudson-openstack) on 2011-03-10

Changed in nova:
status:	In Progress → Fix Committed

Thierry Carrez (ttx) on 2011-04-21

Changed in nova:
milestone:	none → 2011.2
status:	Fix Committed → Fix Released

Report a bug

This report contains Public information

Everyone can see this information.

You are

Subscribing...

Edit bug mail

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.