crewjam / etcd-aws Goto Github PK

So multiple etcd clusters can share the same bucket

Don't panic when provided with an invalid key. Ref this comment:

#2 (comment)

The call to WatchLifecycleEvents in the etcd-aws source is short a parameter (the queueURL) according to ec2cluster source.

As a result the project does not compile.

Hi guys,

I read all the blog post and I loved the solution. But I'm having some problems to get de AWS CloudFormation Template. I'm not a Go developer, and I don't know even start to build this. But I really really need this solution. Is possible to someone send me this etcd.template?

Best regards.

TKS!

Hi!
I were looking at your code as solution for building auto healing etcd cluster on top of ASG.

I started to wonder why you call etcd not etcd2? https://github.com/crewjam/etcd-aws/blob/master/etcd-aws.go#L213
At least when I check in CoreOS alpha (899.11.0) the etcd executable starts the older version and etcd2 the new.

Hello,

i suppose this https://github.com/crewjam/etcd-aws/blob/master/etcd-aws.go#L226
should look like
fmt.Sprintf("ETCD_DATA_DIR=%s", *dataDir)
i order for this to work https://github.com/crewjam/etcd-aws/blob/master/etcd-aws.go#L148-L154

Hi, love your work -- didn't realise it was possible to save/restore the state of etcd and start up without a discovery token.

I have a similar requirement to get a HA CoreOS cluster working in AWS, I had been apprehensive about the cluster getting damaged during a catastrophic reboot.

Having understood what you are doing, have you considered Flocker?

Since you are running the etcd_aws in a docker you will get the load/save into an EBS volume. Don't know if it will make much difference to your design but it removes the need for a named S3 bucket....

Could you please give more details on how to compile this repository with Go? "go install ./..." complains about broken dependencies and doesn't work even if i do "go get %" for all of them...

Since Kubernetes 1.6.0 use etcd3 by default, is there any plan about upgrading the embeded etcd to the lastest version ?

Hi,

When I try to deploy CF via etcd-aws-cfn I get error:
ERROR: deploy: MissingRegion: could not find region configuration

What the way to fix it?

Thank you

Hi,
Seems 5 minutes for
"MasterAutoscale": {
"Type": "AWS::AutoScaling::AutoScalingGroup",
"CreationPolicy": {
"ResourceSignal": {
"Count": 3,
"Timeout": "PT5M"
}
},
not enough. Deployment stops.

When change to 10 - ok.
Here's example:

14:38:32 UTC+0300   CREATE_IN_PROGRESS  AWS::AutoScaling::AutoScalingGroup  MasterAutoscale     Received SUCCESS signal with UniqueId i-df0b7058
Physical ID:CF23-ETCD-PBOGUK-STACK-MasterAutoscale-4N1QTY021CQU
14:32:59 UTC+0300   CREATE_IN_PROGRESS  AWS::AutoScaling::AutoScalingGroup  MasterAutoscale     Received SUCCESS signal with UniqueId i-d3106b54
Physical ID:CF23-ETCD-PBOGUK-STACK-MasterAutoscale-4N1QTY021CQU
14:32:38 UTC+0300   CREATE_IN_PROGRESS  AWS::AutoScaling::AutoScalingGroup  MasterAutoscale     Received SUCCESS signal with UniqueId i-b657a42c

As you can see difference 14:38:32 and 14:32:59 more than 5 minutes.
I made about 10 attempts and results are quite similar.

FYI

Per the documentation, I ran "go install ./..." in the project directory to install the etcd-aws-cfn tool. I got this error:

go install ./...
# github.com/crewjam/etcd-aws/aws
aws/health.go:24: cannot use cloudformation.String("1") (type *cloudformation.St
ringExpr) as type *cloudformation.IntegerExpr in field value
aws/health.go:33: cannot use cloudformation.String("60") (type *cloudformation.S
tringExpr) as type *cloudformation.IntegerExpr in field value
aws/health.go:54: cannot use cloudformation.String("1") (type *cloudformation.St
ringExpr) as type *cloudformation.IntegerExpr in field value
aws/health.go:63: cannot use cloudformation.String("300") (type *cloudformation.StringExpr) as type *cloudformation.IntegerExpr in field value

I replaced cfn.String with cfn.Integer in health.go, and compilation succeeded. I am going to test to make sure the tool works as expected, then I'll submit a pull request.

Hi,

Faced with next situation:
After cluster finished to create - only one node runs etcd-aws service.
On other two I see
Failed Units: 1
etcd-aws.service

journalctl -xe
May 06 12:28:46 ip-10-242-131-220.ec2.internal locksmithd[619]: [etcd.service etcd2.service] are inactive
May 06 12:28:46 ip-10-242-131-220.ec2.internal locksmithd[619]: Unlocking old locks failed: [etcd.service etcd2.service] are inactive. Retrying in 5m0s.

And only if I start service by hand( under root by executing systemctl start etcd-aws) etcd-aws service(and docker container) starts to work.

To recap:
Only one node start etcd-awd after CF deployment, 2 others need to perform to start etcd-aws service by hand.

Any suggestions?