Michael Toennies wrote: > > This container bug come with the dxclient only. > Because i don't change anything with the move cmd since > month and it had worked well all the time, it must be some > server issue. The fact this does work with the unix client suggests this is not a problem that may be easily characterized. Nothing has changed recently for the server side processing of containers in the server. There are two things I could think of that may be happening: 1) The container is not active, but the dxclient still has some method of trying to move stuff into such containers. The server does enforce the fact the container has to be active/open. 2) The dxclient is sending a bogus count (nrof) value to the server. I do see a bug in the code that it will use the nrof sent by the client to determine if the sack can hold it. Ie, lets say you have 10 food. But you set the count to 1000, and try to put that 10 food into the container. The server will use that 1000 number to determine of the container can hold it, and it probably won't be able to hold it. There should be some sanity checking done in the server such that if nrof requested is > nrof in actual object, reset appropriately.