harvest error : failed to unmarshal storage grid err

[wooyoungAhn]

version : 25.01-1
Hello,
I am currently attempting to set up StorageGRID in Harvest, but I encountered the following error when starting Harvest.
Could you please assist me with this issue?

[poller log]

time=2025-08-12T14:11:38.135+09:00 level=WARN source=poller.go:836 msg="init collector-object" Poller=grid-01 error="json: cannot unmarshal string into Go struct field StorageGridError.message of type errs.Message => failed to unmarshal storage grid err" collector=StorageGrid object=Tenant
time=2025-08-12T14:11:38.190+09:00 level=WARN source=poller.go:836 msg="init collector-object" Poller=grid-01 error="json: cannot unmarshal string into Go struct field StorageGridError.message of type errs.Message => failed to unmarshal storage grid err" collector=StorageGrid object=Prometheus

[rahulguptajss]

@wooyoungAhn Could you please share StorageGrid poller configuration? You can run commands as mentioned here https://netapp.github.io/harvest/25.05/help/config-collection/#docker-container

Also please share full logs. You can share them @ https://upload.nabox.org/nufi-gixi-jizy

[wooyoungAhn]

Hi
I uploaded Harvest configuration and log file

Thank you for your help

[rahulguptajss]

Thanks!

[rahulguptajss]

It appears that one of the initial API calls for the StorageGrid connection is failing. Could you please add the following lines to your poller configuration?

recorder:
  path: /path/to/your/local/directory
  mode: record

Please replace /path/to/your/local/directory with an appropriate directory path on your local machine. After making this change, start the Harvest for the StorageGrid poller. Once an error has occurred in the logs, please share the contents of the specified directory by uploading them to the same location as before. You can remove this configuration setting afterward.

For your reference, the full poller configuration should look like this:

sg-tmp:
  datacenter: rtp
  addr: 1.1.1.1
  username: user
  password: pass
  collectors:
    - StorageGrid
  recorder:
    path: /path/to/your/local/directory
    mode: record
  exporters:
    - prometheus1

Thank you!

[wooyoungAhn]

I'm finding it burdensome to have to frequently request file exports from clients. Is there any way I can open the results myself?

[rahulguptajss]

We'll need to analyze the contents of the file with the Harvest code. If you want to check it yourself, then you should look at the _req and their relevant _res files to see if there are any errors in them.

[wooyoungAhn]

Hi
I apologize for not being able to upload the file.
After enabling the record, I noticed something strange in the *res.txt file, as shown below.

cat *.txt

HTTP/1.1 404 Not Found
Accept: application/json


HTTP/1.1 404 Not Found
Content-Length: 85
Cache-Control: must-revalidate,no-cache,no-store
Content-Type: application/json
Strict-Transport-Security: max-age=31536000; includeSubDomains
X-Content-Type-Options: nosniff
X-Frame-Options: SAMEORIGIN
X-Xss-Protection: 1; mode-block

{
  "servlet":"default",
  "message":"Not Found",
  "url":"/api/versions",
  "status":"404"
}

[cgrinds]

Hi @wooyoungAhn what was the matching request that went along with that request? The request will have the same name as the response file. For example, these request and response go together:

8Jm3g7QQ.req.txt
8Jm3g7QQ.res.txt

Is it possible that the hostname or ip address you thought was StorageGRID is something else? That or there is a reverse-proxy or firewall in front of SG that is intercepting the Harvest API calls.

My guess is that the matching request for the response you shared above is

GET /api/versions HTTP/1.1
Accept: application/json

[wooyoungAhn]

HI

They were giving me the wrong IP address.

The problem has been resolved.

Thank you for your help.

[rahulguptajss]

Thanks for the update @wooyoungAhn. You can remove record configuration from the file.