Resolving Call Home communication errors

Understand Call Home communication errors and how to resolve them.

About this task

Call Home communication errors appear as Ceph health errors so that they are visible in both the Ceph status command-line as well as dashboard alerts.

The following is an example of a Call Home communication error from the command-line interface:

[ceph: root@host01 /]# ceph -s
 cluster:
 id: 171b6a7e-3510-11ee-8966-525400377385
 health: HEALTH_ERR
 IBM Ceph Call Home Agent manager module: error sending <performance> report to 
endpoint http://localhost:8000

Procedure

Resolve Call Home communication errors in one or both of the following ways:

Verify that a connection with the Call Home endpoint is in place.
The IBM Call Home endpoint is configured in the manager option mgr/call_home_agent/target. Ping the URL.
If the network where the Ceph cluster runs is isolated from “outside” using a proxy, setup the mgr/call_home_agent/target properly.
- Use the Ceph configuration syntax:
```
ceph config set mgr mgr/call_home_agent/proxy http://128.64.64.12:8090
```
- Use the environment variable syntax:
```
export CHA_PROXY=http://128.64.64.12:8090
```
For full command and syntax information, see Modifying the Call Home agent configuration.
Manually create a status report and check the output and logs.
1. Log in to the cephadm shell.
  For example,
```
[root@host01 ~]# cephadm shell
```
2. Retrieve the status log.
```
ceph callhome send status
```
  For example,
```
[ceph: root@host01 /]# ceph callhome send inventory
```
3. Check the command output and logs for any errors.