Applications Systems
NetApp polling fails with remote certificate validation errors in Storage Resource Monitor
This article discusses an issue when NetApp polling fails in Storage Resource Monitor because the remote SSL/TLS certificate cannot be validated. This can occur when the certificate name does not match the array address used by SRM, when the certificate chain is not trusted, or when certificate validation checks block polling.
First published date
Last published date
Overview
Storage Resource Montior is unable to poll one or more NetApp storage arrays. Storage data is not collected, and the SRM poller logs show repeated HTTPS/TLS certificate validation failures against the provider endpoints.
SRM Poller Jobs log shows:
WARN SolarWinds.Orion.Security.Validation.CertificateValidatorBase - The remote certificate is not valid due to SslPolicyErrors:RemoteCertificateNameMismatch, RemoteCertificateChainErrors.
Target: x.x.x.x
Chain.Status: UntrustedRoot
Chain.Elements: C=US, CN=selectedcert
ERROR SolarWinds.SRM.NetAPP.NetAppServerBase - Failed to verify remote certificate
NetApp.Manage.NaConnectionException: Failed to verify remote certificate
---> System.Net.WebException: The underlying connection was closed: Could not establish trust relationship for the SSL/TLS secure channel.
---> System.Security.Authentication.AuthenticationException: The remote certificate is invalid according to the validation procedure.
Product section
Cause
SRM is failing remote certificate validation when it connects to the NetApp providers over HTTPS/TLS.
Cause 1: Certificate name mismatch
SRM is connecting to the array by an IP address or hostname that does not match the identity presented in the certificate. This is logged as RemoteCertificateNameMismatch.
Example from the log:
-
SRM connects to x.x.x.x
-
The certificate subject is
CN=selectedcert
This means the certificate does not match the way SRM is addressing the array, so the TLS handshake is rejected.
Cause 2: Untrusted root / certificate chain error
The array certificate chain is not trusted by the SRM poller. This is logged as RemoteCertificateChainErrors with Chain.Status: UntrustedRoot.
This usually means one of the following:
-
The array is using a self-signed certificate
-
The internal CA is not trusted on the SRM poller
-
The certificate chain is incomplete or missing intermediate/root CA certificates on the SRM poller
Resolution
Solution for Cause 1: Fix the certificate name mismatch
If the log shows RemoteCertificateNameMismatch, correct the certificate identity mismatch.
Steps
-
Identify the exact IP address or hostname SRM is using for the provider.
-
Create or replace the array SSL certificate so that:
-
the certificate CN is the array FQDN, and
-
the certificate SAN includes:
-
the FQDN used by SRM
-
the IP address of the array used by SRM
-
-
-
If SRM is connecting by IP address, the certificate must include the IP address in the SAN. This is the key fix for IP-based polling and avoids
RemoteCertificateNameMismatch. -
If possible, update the SRM provider to use the array FQDN instead of raw IP, but only if that FQDN is present in the certificate.
-
Restart the relevant SolarWinds Platform services after replacing the certificate so the pollers pick up the new state.
Solution for Cause 2: Fix the untrusted root / chain error
If the log shows Chain.Status: UntrustedRoot, the SRM poller does not trust the certificate chain.
Steps
-
Determine whether the array certificate is:
-
self-signed, or
-
issued by an internal/private CA
-
-
Export the required certificates:
-
root CA certificate
-
any intermediate CA certificates
-
or the self-signed certificate if that is what the array uses
-
-
Import the required CA certificate(s) into the Trusted Root Certification Authorities store on the SRM polling engine.
-
If intermediate certificates are used, import them into the appropriate intermediate certification store as well.
-
Restart the relevant SolarWinds Platform/SRM services after importing the certificates so the updated trust chain is applied.
-
Re-test polling. Once corrected, the log should no longer report
UntrustedRootfor those arrays.
Temporary workaround: Disable the certificate check
If the monitoring needs to restored before the certificates can be corrected, you can temporarily relax the Remote Certificate Validation settings.
1.Access the Advanced Configuration Page
2. Look for SolarWinds.Orion.RemoteCertificateValidation
-
- Check On Certificate Chain Errors
- Check On Certificate Name Mismatch
- Check On Certificate Revocation
3. Hit Save.