exporter crashing when trying to reach the metrics

[vaibhavjayas]

time="2019-06-13T14:03:46Z" level=info msg="Fetching volumes info" source="cinder.go:54"
time="2019-06-13T14:03:46Z" level=error msg="Get /volumes: Get /volumes: unsupported protocol scheme """ source="cinder.go:57"
time="2019-06-13T14:03:46Z" level=info msg="Fetching services state information" source="cinder.go:60"
time="2019-06-13T14:03:46Z" level=error msg="Get /os-services: Get /os-services: unsupported protocol scheme """ source="cinder.go:63"
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0x82827d]

goroutine 61 [running]:
main.(*CinderExporter).Collect(0xc000180580, 0xc00016d7a0)
/app/cinder.go:66 +0x1dd
github.com/prometheus/client_golang/prometheus.(*Registry).Gather.func1()
/app/vendor/github.com/prometheus/client_golang/prometheus/registry.go:434 +0x193
created by github.com/prometheus/client_golang/prometheus.(*Registry).Gather
/app/vendor/github.com/prometheus/client_golang/prometheus/registry.go:445 +0x590

[mikejoh]

Hi @vaibhavjayas,

Could you somehow verify, in your configuration file passed to openstack-exporter, that you use the correct region_name for your OpenStack environment? Also, if possible, please check if you actually have the volumev3 service available?

The only way i can think of that this could happen is when the region name is wrong or volumev3 (Block Storage API v3) is not available to the user for some reason. Here's the code:

openstack-exporter/cinder.go

Line 24 in 35b2bcb

endpoint := client.EndpointsForRegion(config.Region)["volumev3"]

I managed to reproduce it by setting config.Region (in the above method call) or volumev3 to something different from what we use in our environment and i got back the exact same error messages.