The fence_aws fence agent fails to power on a node in a Pacemaker cluster
Issue
- A node got fenced by
fence_aws
, but it was left in a powered off state instead of getting rebooted.
Feb 24 10:52:00 dn1sh4vm3 fence_aws: Status operation for EC2 instance i-xxxx returned state: RUNNING
Feb 24 10:52:00 dn1sh4vm3 fence_aws: Starting new HTTP connection (1): 169.254.169.254
Feb 24 10:52:01 dn1sh4vm3 fence_aws: Called StopInstance API call for i-xxxx
...
Feb 24 10:52:06 dn1sh4vm3 fence_aws: Status operation for EC2 instance i-xxxx returned state: STOPPING
...
Feb 24 10:52:38 dn1sh4vm3 fence_aws: Status operation for EC2 instance i-xxxx returned state: STOPPING
Feb 24 10:52:39 dn1sh4vm3 fence_aws: Status operation for EC2 instance i-xxxx returned state: STOPPED
Feb 24 10:52:39 dn1sh4vm3 fence_aws: Starting new HTTP connection (1): 169.254.169.254
Environment
- Red Hat Enterprise Linux 7 (with the High Availability Add-on)
- Red Hat Enterprise Linux 8 (with the High Availability Add-on)
- Amazon Web Services (AWS) EC2 Instances as Cluster Members
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.