312- CERT_RENEW_3_LNX Failed

Explanation:

The automatic distribution of certificates is implemented in the form of schedules.
Schedules, such as CERT_RENEW_0_LNX to CERT_RENEW_2_LNX, each write parts of the new certificate to a temporary file.
Finally, CERT_RENEW_3_LNX writes the new, now fully available certificate to the ISP's certificate database.

Unfortunately, the above-mentioned schedules cannot detect whether the new certificate already exists.
Therefore, CERT_RENEW_3_LNX will reattempt to write the now existing certificate to the ISP's certificate database the next time.

This attempt fails correctly, and an error message is displayed in dsmsched.log.

This typically looks like this:

  Scheduled Start          Actual Start              Completion Time               Schedule                Status     Return Code                 Reason

2025-07-10 19:50:25     2025-07-10 19:50:36		    2025-07-10 19:50:36    		CERT_RENEW_0_LNX          Completed       0          	All operations completed successfully.
2025-07-10 20:00:25     2025-07-10 20:00:36		    2025-07-10 20:00:36    	    CERT_RENEW_1_LNX          Completed       0          	All operations completed successfully.
2025-07-10 20:10:25     2025-07-10 20:11:37    		2025-07-10 20:11:37   	    CERT_RENEW_2_LNX      	  Completed       0          	All operations completed successfully.
2025-07-10 20:20:25     2025-07-10 20:20:37   		2025-07-10 20:20:40     	CERT_RENEW_3_LNX          Failed          66         	Failed

What can you do?

I. Ignore it

You can ignore the error message because automatic certificate distribution is only temporarily enabled.
After disabling this automatic distribution, no more error messages will appear.
We also independently search for nodes with such problems and then resolve the issue directly on the LRZ side.
However, this process takes some time, which is why it cannot be done immediately.

II. Test functionality

It is best to test this with the CLI client:

dsmc

Or, if multiple servers are configured on the system:

dsmc -se=<the name of the server>

or:

dsmc -optfile=<path to the matching dsm.opt>

If the login works, the new certificate is already present and the error message mentioned above can be ignored.