> On Jun 21, 2022, at 1:51 PM, Olga Kornievskaia <aglo@xxxxxxxxx> wrote: > > For #2, I have personally ran into that stack trace while > investigating a hung using soft iWarp as the rdma provider. It was an > unpinned request but I think it was due to soft iWarp failure that > cause it not to do a completion to rdma which led to the request never > getting unpinned. Thus I would recommend looking into failures in your > rdma provider for clues on that problem. #2 is a separate problem entirely, and is a known bug. I plan to add some fault injection so we can reproduce and track it down. I think you should prune this issue and focus just on #1 for now. -- Chuck Lever