Minutes from the 7/14/04 LCLS SLC IOC meeting. Next Meeting: 7/21/04 in B5, Rm 211. (1) People availability: Ron - gone next week (2) Prototype: SLC message service on the IOC. Diane is ready to send test messages (that don't require any database) from the SCP to her IOC. Needs Ken's help. ACTION: Diane and KenU DUE: July 23 (3) Ron asks - do we really need a VMS-based timing user interface? For example, do we need to knob timing delays from the Alpha? ACTION: TBD DUE: TBD (4) Ken Underwood gave a general explaination of DBEX-related messages and which micro tasks/jobs are involved in the message transfer. I will add the picture to http://www.slac.stanford.edu/comp/unix/package/epics/lcls/slc_ioc/. He identified some specific work that needs to be done: (a) SLC IOC must set a CSTR timestamp after it receives its database from DBEX. (b) SLC IOC must set some of the status bits in CSTR STAT on IPL (and maybe other times too). (c) VMS PARANOIA will send a micro check function every two minute. The SLC IOC must respond to the message - otherwise, PARANOIA will set it offline and no SCP can talk to the SLC IOC until a restart of the SLC interface of the IOC. (d) The SCP micro IPL function must be modified to NOT send a micro image. In its place, the SCP must send a new SLC-IOC-only message to the IOC telling it to reinitialize its database and reinitialize any SLC tasks to prepare for a new database and other init messages that happen at IPL time. A new bit to CSTR HDSC will be added to allow the SCP to identify a SLC IOC. This will allow us to "IPL" the SLC interface of an IOC without rebooting the IOC. ACTION: TBD DUE: TBD (5) Add a few magnets to XL01 and DBGEN on mccdev. Add another SLC IOC (XL02). We THINK we can run two SLC IOCs on the same node (ie, solaris) for testing. ACTION: KenU DUE: July 21 (6) Functional Requirements: Dayle has started looking at the magnet job functions. Steph and Ron have not looked at their jobs (BPM for Steph, Timing/Misc for Ron). The full list of func reqts are due Aug 27. ACTION: Dayle, Steph, Ron DUEL July 21 (7) Same Device controlled by EPICS or by the SCP: Right now, the idea is that on the SLC IOC, we can control a device (like a magnet) from either the SCP or directly using some EPICS application. But if the magnet is controlled by the EPICS app, then the magnet setpoint in the EPICS database changes but not the one (BDES) in the SLC database. But BACT will still be updated (due to asynch updates) so the SCP will show the difference. We could add in functionality to update the BDES in the IOC's SLC database (I think?) but since it's supertype 2 (meaning that only the Alpha can write to it), it won't make it up to the VMS SLC database. How much work would it be to allow both supertype 2 and 3 updates from the SLC IOC up to DBEX and into the VMS database? Stephanie