vRanger tasks fail with the error message - Error: Could not log in to [https://<vcenterFQDN>/sdk/vimService.
In a number of cases this same job will work fine at a different point in time during the day, this error tends to present itself when the vCenter (VC) server itself is either unavailable or is overutilized at the specified time that the backup is suppose to run.
Things to check:
1. Separate the backup of the vCenter server to a different point in time than the rest of your backup jobs. Back up the vCenter VM either before or after the rest of your environment to prevent it trying to open a snapshot of itself and close that snapshot while it is coordinating the processes for the rest of your jobs.
2. Rerun the same job, or create a separate job to back up this VM when available to validate that outside of the 'backup window' this VM does indeed back up, and it is only while during the backup window that this error message is generated.
3. Check the amount of RAM that is allocated to the vCenter server itself. The vCenter server is very RAM dependent, and the processes of opening and closing snapshots is directly tied to the available RAM resources on the VC. The minimum RAM requirement for version 5.0 is 4 GB for just VMware alone (not including any Windows or SQL requirement), conversely the minimum requirement for 5.1 is 10 GB of RAM (if Single Sign On and Inventory Service are on the same VM). If during the day you can log into your VC through the VMware VI and see that 90-95% (or more) of the RAM allocated to the VC is already being consumed then it is more than likely that more RAM will need to be allocated to the VC in order to continue running backups the way that you have them configured.
4. In Ranger you can specify how many jobs are run at a time. You can specify how many jobs can be run per host, per datastore, per repo, per VA ..... (ect). This can be found in Ranger under Tools > Options > My Jobs > Configuration. It could be that perhaps you are doing too many jobs concurrently for your current environment. Perhaps you are doing 10 tasks at a time and moving this from 10 to 6 might rectify this error - remember this may mean that your backup window might increase, however it is variable that might need to be addressed if during the backup window your VC is inaccessible.
The error message is again a loss of communication between Ranger and the VC. This generally is not a network loss, it is more a case where we aren't able to access the information from the VC in order to backup/retrieve the data for the VMs, which tends to be a case of the VC being over utilized at the point in time that the backup was run.
