CMS monitoring showed a big drop in running slots on evening of the 17th/morning of the 18th, but the SI operator told me he was running tests again (that do not show up in CMS monitoring). Vande continued to show the ~12k slots running continuously. 

CMS job efficiency has dropped off. We also notice some network saturation on LHCONE which is being tentatively blamed on CMS remote reads. There is correlation between the running slots decrease mentioned above and a temporary drop-off in network traffic. 

One gateway had a problem yesterday that caused some red webdav and xrootd SAM tests. 

RobH did a change to the RAL-based redirector: https://helpdesk.gridpp.rl.ac.uk/Ticket/Display.html?id=481341 Nothing was working - needed an auth file adding (explained in the config file), then IPv6 address had not been added (aquilon had this config but somehow did not add it to the host). Now the redirector appears to be working...but Katy is monitoring the cms-aaa-manager01 logs to try to determine if that is properly in contact with the redirector.