Description of problem: In provider regions with large numbers of storage accounts, performing a refresh will sometimes result in a Timeout error or ECONNREFUSED exception as it attempts to collect private image information. When this happens the entire refresh fails. These errors are coming from Azure and are beyond our control. Version-Release number of selected component (if applicable): 5.6, 5.7, 5.8 How reproducible: Sporadic. Steps to Reproduce: 1. Add Azure cloud provider in region with many storage accounts. 2. Attempt to refresh multiple times. 3. View logs. Actual results: Refresh fails with TimeoutException or Errno::ECONNREFUSED. Expected results: Refresh skips over such failures and performs the rest of the refresh normally. Additional info: This particular issue was addressed by the latest version of the azure-armrest gem which will skip over these errors while collecting private images.
https://github.com/ManageIQ/manageiq/pull/15239 (Euwe/Darga) https://github.com/ManageIQ/manageiq-providers-azure/pull/74 (Fine)
Azure refreshes working fine with CFME 5.9.0.20.20180206200727_d1873b9 using azure-armrest 0.9.5, moving to VERIFIED on the presumption that the newer armrest gem solved the problem as stated.