 |
 |
Hello, Below you can find detailed information regarding FTS channel states: Active: The Channel Allocator is looking for work to assign to a channel, transfer jobs are executed. Drain: The Allocator will not add anything new to a chan- nel but the Trigger will continue to serve jobs that have been assigned to its channel. The effect is to drain all pending transfers from the channel. Inactive: The Allocator will assign work to a channel but the Trigger will not put any more jobs on the wire. This is used by a sysadmin to empty the network. Note that jobs currently active on the wire will complete. Stopped: Neither the Trigger or Allocator will do any work. Nothing will be assigned to the channel and no work will be put on the wire. Existing jobs on the wire will complete. Halted: A serious error has been detected on the channel (e.g. there have been a certain number of ’sequential’ failures on the channel in the last few minutes) and that the channel has been automatically stopped by some monitoring process. This state is designed to prevent the transfer queue draining in case of problems. Regards, Michail ________________________________ From: LHC Computer Grid - Rollout [[log in to unmask]] on behalf of [log in to unmask] Sent: 13 February 2012 06:54 To: [log in to unmask] Subject: [LCG-ROLLOUT] setting FTS channel states Can someone explain the meaning of the FTS channel state options [Active, Inactive, Drain, Halted, Stopped, Archived] in the command: glite-transfer-channel-set The man page for the command is at: http://glite.web.cern.ch/glite/documentation/R3.0/autogen/org.glite.data/org.glite.data.transfer-cli/html/glite-transfer-channel-set.1.html What is the difference between Halted and Stopped? Thanks, Lorne