Can you please create new AMI image from instance i-000007ff ? - It is running right now and have internal IP 172.16.3.3
This is instance of ami-0000000e, where I populated mock cache for all used chroots. I will use this image then for Copr builders, which should result in faster builds. Once I verify that new image works, I will ask you to delete that old AMI.
That running instance 172.16.3.3 can be stoped. And even terminated after you create AMI. It exist solely for the creation of AMI image.
I'm not sure at all this will be any faster. :)
However we can give it a try...
I asked it to make a snapshot of that instance and it's currently "queued". So, I guess I'll check back in a while and see if it did it or not.
ok, it finally finished.
It's: ami-00000041
Give it a try and see if it's any faster at all...
Reopening: I can not see this image:
and playbook spin up gives me:
... msg: ImageNotFound: Image ami-00000041 could not be found.
I didn't have the image marked public. ;)
I have done so now. If you still can't see it let me know, but hopefully it will be visible now.
It is visible now.
But something is bad. :(
ssh root@copr-be-dev.cloud.fedoraproject.org
su - copr
. /home/copr/cloud/ec2rc.sh
ansible-playbook -v -c ssh /home/copr/provision/builderpb.yml
it will try to spin up VM, but playbook timeout on that task. And: /usr/bin/euca-describe-instances|grep 41
INSTANCE i-00000dfe ami-00000041 209.132.184.218 server-902b20f4-8246-4d8b-a7c9-611ca0262ce1 pending buildsys 0 m1.builder 2014-01-07T07:56:11.000Z nova aki-0000000f ari-00000010 monitoring-disabled 209.132.184.218 172.16.3.5
...
Notice the status: pending. I waited quite long time. And it is still pending.
I am not sure how can I debug it further :(
Additionally those pending VM, can not be terminated:
euca-terminate-instances i-00000dfe i-00000dfd i-00000e11 i-00000e0b i-00000e04
INSTANCE i-00000dfe pending pending
INSTANCE i-00000dfd pending pending
INSTANCE i-00000e11 pending pending
INSTANCE i-00000e0b pending pending
INSTANCE i-00000e04 pending pending
and it still sit there in the same time and Cloud hesitate to terminate those machines :(
I did: cp builderpb.yml builderpb-newami.yml and roll back builderpb.yml to that old AMI id. so I can continue developing with working ansible script.
all instances are terminated now but one. Instance i-00000e04 is still pending.
I can't seem to kill that one either here. :(
I guess we are stuck with it until we next reboot.
I am going to close this now and we can revisit when we have the new cloud.
Login to comment on this ticket.