This page is http://rsphysse.anu.edu.au/~bdb112/h-1nf/mdsplus/mdsplusnotes.txt

Define a second server
prlas0$ MCR NCP
NCP>define object camac2_server number 0 file mdsserver.com user mdssrv password mdsserver proxy none
NCP>set object camac2_server all

 define/group camac_server PRLAS2::CAMAC2_SERVER


To Do before Tom leaves:
Look at network plan (before 10:30)
See Shiu re startup improvements and network plan at 1:30PM
scope and pads at 2:00, followed by labview talk part III
Take H1SIMOLD, copy to H1DATA, graft in john's subtree
show H-1 electronic log and discuss
discuss clock hardware, any new hdwe that MIT is planning to support
is there a page "MDSPlus port status"
speed of plotting (c.f. stpl)
speed of DAQ - try two simultaneous
support for RS232/Bitbus/PLC communication
serial highway/SCSI alternatives for crate controllers (with JH)
Josh's visit: Say 6-9 months?  set up SQL, MDSPlus update, MDSIP,
Java, netwrking, fileserver/raid
----------
non-critical:
why are the full paths sometimes short, and sometimes long?
COnfusion between mds$tcl and csv
BDB/David to look at Java support - if possible to "downgrade"
Manducci code to 1.1, otherwise use TWF code.  Add main signals menu,
cut and paste (or mouse operation to add new trace, hyperlink access
to core diagnostics

--------------------------------------------------
Protoype H1 tree

H1DATA

points to LABTEST
      and ELECTR_DENS



def h1data$data ECC$USER:[PRL.BDB112]

mds$open,'h1data',31000
dia=mds$value('dia_335')
dtime=mds$value('dim_of(dia_335)')
plot,dtime,dia


PRLAS2$ backup h1data$data:h1data_*.*; [.mdsplustestdata]/igno=interlock

$! Defs for h1 MDSPlus demo, March 99
$! These are somewhat kludged, so need to review before putting onto prlas0,1
def/group IDL_DIR "PRLAS0$DKA600:[IDL.IDL51.]"/trans=(t,c)
def/group h1data$data ECC$USER:[ECC.MDSPLUSTESTDATA]
def/group h1data$ h1data_all

x=mds$value('mdsv1("H1DATA$",*,32500,"I_FAULT")')
mcr traverser -edit -tree h1data -shot -1




IDL> x=mds$decompile('dia_335')
IDL> print,x
Build_With_Units(MDSV1("H1DATA$", $SHOT(), "DIA_335"), "Volts")
IDL> mds$tcl                    
TCL> put/extend dia_335 
PUT> Build_With_Units(MDSV1("H1DATA$",*, $SHOT(), "DIA_335"), "Volts")
PUT> 
TCL> exit
IDL> dia=mds$value('dia_335')           


mds$put, "DIA_1" ,Build_With_Units(MDSV1("H1DATA$",*, $SHOT(), "DIA_335"), "Volts")

IDL> mds$tcl                 
TCL> set tree h1data
TCL> create pulse 32500
; this one won't have the i_fault in it

TCL> edit h1data
TCL> add node i_fault
; whoops, forgot the /usage
TCL> delete node i_fault
TCL> add node i_fault/usage=signal
TCL> write                         
TCL> close
TCL> set tree h1data
; operates on model
TCL> put/extend i_fault
PUT> Build_With_Units(MDSV1("H1DATA$",*, $SHOT(), "I_FAULT"),"Amps")
PUT> <CR> 
TCL> create pulse 32500                                              
TCL> exit                                                            
IDL> mds$open,'h1data',32500
IDL> plot,mds$value('i_fault')



unix: ~twf112/binaries
setup.csh
then set database, data access to prlas2
2$ show log mds$path
et def mds$root:[tdi]
$ help @mdsplus

create/dir [.demo]
set def [.demo]
define h1$data ecc$user:[ecc.demo]
traverser := $traverser
 traverser -tree h1 -shot -1 -edit
cre/dir [.spectroscopy]
2$ define spectroscopy$data ecc$user:[ecc.demo.spectroscopy]


 %LIB-E-ACTIMAGE, error activating image PRLAS2$DKA0:[PRL.MDS.ALPHA.][SYSLIB]HYTECCS
HR.EXE;
-RMS-E-FNF, file not found
define/group camshr remcamshr
 set prot=w:re mds$root:[syslib]hyteccshr.exe
 set prot=w:re mds$root:[syslib]hyteccshr.exe 
 define hyteccshr ecc$dir:hyteccshr.exe/group
 copy hytsysshr mds$root:[syslib]



def/group IDL_DIR "PRLAS0$DKA600:[IDL.IDL51.]"/trans=(t,c)
def/group h1data$ h1data_all
def/group h1data$data ECC$USER:[ECC.MDSPLUSTESTDATA]

$! a script for user xxx112 to copy the h1data database for playing around
$! problem is, the database name must match the file name


$ def test$data scu$prl:[xxx112.test]
$ deass/process h1data$data  ! in case there is an old h1data$data
$ create/dir test$data
$ backup h1data$data test$data
$ define h1data$data test$data

$! the following ONLY on prlas2
def scu$sysroot prlas0$dka0:[SCU1.]/tr=(c)/group
define hyteccshr ecc$dir:hyteccshr.exe/group
define hytsysshr ecc$dir:hytsysshr.exe/group

^C
x=f$lexical('set proc/priv=grpprv)
cont

Version specific files in sys$sysspecific (not many - just one mds so
far)

TreeHooks - allows changing something before another try to open a shot
used to recall disks from store, to change logicals

using(one time access to another database)

backquote - force evaluation of ken variable
 x=f$setprv("ALL")

Problems:
This happens the SECOND time - the first error is OK

IDL> mds$open,'h1data',32500
IDL> mds$put,'a14_1:input_2','build_signal(build_with_units($value*.000134+.12013,"volts"),build_with_units($,"
counts"),build_with_units(build_dim(build_window(0,131000,0.),*:*:1/1E-5),"seconds"))',x
%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=00000000AAAAAAAB, PC=FFFFFFFF8050C8E8, PS=0
000001B
%TRACE-F-TRACEBACK, symbolic stack dump follows
  image    module    routine             line      rel PC           abs PC      
                                            0 FFFFFFFF8050C8E8 FFFFFFFF8050C8E8
                                            0 FFFFFFFF80635B78 FFFFFFFF80635B78
                                            0 FFFFFFFF80636080 FFFFFFFF80636080
                                            0 FFFFFFFF80638CD8 FFFFFFFF80638CD8
 IDL                                        0 000000000018AA4C 000000000018AA4C
 IDL                                        0 000000000018B43C 000000000018B43C
 IDL                                        0 000000000018B54C 000000000018B54C
 IDLTRANSFER                                0 0000000000030600 0000000001D2C600
 MDSLIB_IDLV5                               0 0000000000030D1C 0000000001CDAD1C
 MDSLIB_IDLV5                               0 00000000000305E0 0000000001CDA5E0
 IDL                                        0 00000000001BF3EC 00000000001BF3EC
 IDL                                        0 000000000017016C 000000000017016C
 IDL                                        0 000000000015DF6C 000000000015DF6C
 IDL                                        0 0000000000150D94 0000000000150D94
 IDL                                        0 000000000015006C 000000000015006C
                                            0 FFFFFFFF835EF118 FFFFFFFF835EF118

special variables like  $value $name see tdishare help ($ help @mdsplus)
 mds$value("getdbi('NAME')")

TREES:
Old:
H1SIM
H1DATA
New:
 H1DATANEW1  rf ech magnetics 2mm

$ define/group H1SIMOLD$DATA ecc$user:[ecc.h1simold]
$ create/dir/prot=(g:RE) H1SIMOLD$DATA
$ TCL
 edit/new h1simold/shot=-1

Shows ALL details of data stored in signal, even if there are two
 arrays - one in raw, one in float
TCL>  show data .rf:a14_3:input_3 
 
$ define/group a14_1 ecc18:N1
$ define/group tr612_1 ecc18:N2

/ECC$$USER:/ECC/DISPATCHING/H1SIMPLE.TCL
! This is a relatively simple H1 dispatcher - where possible, literal names are used
! bdb 3/99

close/all			! precaution for incomplete previous cycle
set tree h1data /shot=-1	! use model as template
create pulse "CURRENT_SHOT(""h1data"")+1"/cond		! create the next shot
set tree h1data /shot="CURRENT_SHOT(""h1data"")+1"	! open it
mark							! prune unused branches
dispatch/close   		! precaution against incomplete previous cycle	
dispatch/build/monitor=prlas2::monitor_server  ! search tree, compile action table

dispatch/phase/monitor=prlas2::monitor_server init/sync=10	! dispatch init actions
set current h1data/increment					! update current shot

dispatch/phase/monitor=prlas2::monitor_server store/sync=100	! dispatch store phase actions

dispatch/phase/monitor=prlas2::monitor_server anal		! dispatch anal phase

dispatch/close							! close all trees at all servers
close								! close this tree here

To execute, do:
def camac_server prlas0::camac_server/group
def a14_1 ecc2:n19/group
tcl @[.dispatching]h1psupply
tcl @[.dispatching]h1simple

actmon !(then remove all options) to see the phases

Typical errrors
 13:34:44.71 Action .OPERATIONS:H18212SL:STORE_ACTION failed, %SYSTEM-F-NOSUCH
REASON: camac_server logical not defined

To debug do
TCL
<mouse in lines up to INIT>
dispatch .RF:A14_4:INIT_ACTION
%SYSTEM-F-NOSUCHNODE, remote node is unknown
decompile .RF:A14_4:INIT_ACTION     
Build_Action(Build_Dispatch(2, "CAMAC_SERVER", "INIT", 50, *), Build_Method(*, "INIT", .RF:A14_4), *, *, *)

TCL> 
set tree h1data/shot=34005      
do/method .RF:A14_4 INIT   

tcl
set tree h1data/sho=-1      
creat pulse 37999           
set tree h1data/sho=37999   
do/method .RF:A14_4 INIT    
do/method .RF:A14_4 STORE   
SW32kcam$bytcnt = 64000SW 3kcam$bytcnt 

h1psupply.tcl works on prlas0 now that Tom has done the
MDS$ROOT:[SYSMGR]CAMAC_SERVER_TCLINIT.TCL
containing one line:
ACTIVATE/IMAGE=MIT$DEVICES

 "CAMAC_SERVER" = "PRLAS0::CAMAC_SERVER" (LNM$GROUP_000112)
 "CAMSHR" = "DKA100:[PRL.ECCCAMSHR]ECC_CAMSHR.EXE;108" (LNM$GROUP_000112)

a14 pre/post requires a START
a14 cyclic requires a STOP

Doesn't work - why?

13:48:00.23 Dispatching node .RF:A14_6:STORE_ACTION to 
13:48:00.29 Action LAMWAIT:EXEC_ACTION failed, status = 160333904
13:48:00.29 CAMAC_SERVER is beginning action .OPERATIONS:A14_1:STORE_ACTION
13:48:00.32 Action .OPERATIONS:A14_1:STORE_ACTION failed, %SYSTEM-F-IVLOGNAM
13:48:00.33 CAMAC_SERVER is beginning action .OPERATIONS:H18212SL:STORE_ACTION
13:48:00.35 Action .OPERATIONS:H18212SL:STORE_ACTION failed, %SYSTEM-F-IVLOGNAM
13:48:00.35 CAMAC_SERVER is beginning action .OPERATIONS:A14_2:STORE_ACTION
13:48:00.43 Action .OPERATIONS:A14_2:STORE_ACTION failed, %SYSTEM-F-IVLOGNAM
"I

prlas0$ tcl stop server camac_server

typ/tail mds$root:[syserr]prlas0_camac_server.log
26-JUL-1999 13:48:00.35 Doing action \H1DATA::TOP.OPERATIONS:A14_2:STORE_ACTION 
in H1DATA       shot 34253
%SYSTEM-F-IVLOGNAM, invalid logical name
26-JUL-1999 13:49:12.76 Shutdown request received

Now it listens to a stop!
@prl_mds:showsys
showing processes conatining string CAMAC
2C6000AD CAMAC           CUR      4     5322   0 00:00:06.78      6298     89   

-

PRLAS2$ tcl
TCL> stop server camac_server

Define the reseource SPECTROSCOPY

ADD/IDENTIFIER/ATTRIBUTES=(RESOURCE)/value=identifier:%X10043 spectroscopy
 show/id/attr=res                                    
Grant the "right/resource?" H1DATA  to user MDSSRV

PRLAS2$ set def sys$system: 
PRLAS2$ run authorize
UAF> grant /ident h1data mdssrv /attr=resource
%UAF-I-NOMODS, no modifications made to system authorization file
%UAF-I-RDBDONEMSG, rights database modified

!%UAF-I-GRANTMSG, identifier SPECTROSCOPY granted to FJG112
UAF> grant/ident electr_dens smc112/attr=resource
%UAF-I-GRANTMSG, identifier ELECTR_DENS granted to SMC112
 no need to restart

Whoops - this is how to remove them if you accidentally define a user
of that name                   
UAF> remove spectroscopy                              


showing users who have files open on DKA0 containing  string ECC
                00000000  [VMS$COMMON.SYSLIB]DECC$SHR.EXE;1
                00000000  [VMS$COMMON.SYSMSG]DECC$MSG.EXE;1
                00000000  [SCU1.ECC]HYTSYSSHR.EXE;1
camac_server    23C001B6  [SCU1.ECC]HYTECCSHR.EXE;1



PRLAS0$ tcl
TCL> stop server camac_server
TCL> exit
PRLAS0$ eccop/load
%RUN-S-PROC_ID, identification of created process is 23C0043F
PRLAS0$ show proc ecc_1365
%SYSTEM-W-NONEXPR, nonexistent process
PRLAS0$ show proc ecc1365
%SYSTEM-W-NONEXPR, nonexistent process
PRLAS0$ @prl_mds:showsys ecc
showing processes conatining string ECC
23C0043F ecc_1365        LEF      6       44   0 00:00:00.01       228     33   
PRLAS0$ show proc ecc_1365
%SYSTEM-W-NONEXPR, nonexistent process
PRLAS0$ show proc /id=ecc_1365
%SHOW-E-INVQUAVAL, value 'ECC_1365' invalid for /IDENTIFICATION qualifier
PRLAS0$ show proc id=ecc_1365
%SYSTEM-W-NONEXPR, nonexistent process
PRLAS0$ tcl start server camac_server
11:01:35.92 Action LAMWAIT:EXEC_ACTION failed, %TREE$-W-NOMETHOD
11:01:35.93 CAMAC_SERVER is beginning action .OPERATIONS:A14_1:STORE_ACTION

=========Running multiple CAMAC_SERVERS================
! Need to do on AS0 and AS2
def operations_server prlas2::camac_server/group
Then AS2 runs the OPERATIONS_SERVER and AS0 runs the CAMAC_SERVER.
THis means that fast/small, reliable digitizers under the direct
control of the physics operator can be relied upon here, and other
diagnostics out of control of the physics operator can be put into the
CAMAC_SERVER queue.

Within one shot, the way the H1PSUPPLY.TCL file works, a stall in
either server will not stall the other.  
Before the next shot, the stalled server can be unclogged by
$ tcl abort server CAMAC_SERVER 
which kills the stalled operation but otherwise continues.
DO NOT USE THIS tactic FRIVOLOUSLY

Note also that in the current setup, the servers run on separate
computers with separate licenses, and there is not (at the moment)
significant cross operation between crates - need to test this
tomorrow.

================



PRLAS2$ set file/acl/like=prlas2$dka0:[prl]h1data.dir PRLAS2$DKA0:[PRL.MDS]TEST_trees.dir
PRLAS2$ set file/acl/default PRLAS2$DKA0:[PRL.MDS.test_trees]*.*;*

**** ACL - it seems that files are created with the ACLS of the
directory in which they reside.  If the directory has no ACLS, then
this probably subject to the default protection of the creating
process.
SEE   $ HELP SET SECUR/DEFAULT

set file/acl/like=prlas2$dka0:[prl]h1data.dir spectroscopy$data:[000000]spectroscopy.dir
set file/acl/default  PRLAS0$DKA100:[PRL.JNH112.TREES.SPECTROSCOPY]*.*;*

=============
Tried for h1data on 22/5/00
set file/acl/like=prlas2$dkb0:[prl.mds]test_trees.dir dka100:[prl.trees]*.dir
only worked while there were files of the same shot to superseded

set file/acl/default dka100:[prl.trees.h1data]*model.*
this worked
=============
create/dir dka100:[prl.trees.electr_dens]
set file/acl/like=prlas2$dka100:[prl.trees]h1data.dir dka100:[prl.trees]electr_dens.dir
set file/acl/default dka100:[prl.trees.electr_dens]*model.*
=========
re do after john moves the files??
set file/acl/like=prlas2$dka0:[prl]h1data.dir PRLAS0$DKA600:[TREES]spectroscopy.dir
Why could clive do this (DO...) but the server got an PMDS-PRV result.


========
=====================================================Move wayne's data 9/00=======
create/dir dka100:[prl.trees.fluctuations]
set file/owner=[PRL,WMS112] dka100:[prl.trees]fluctuations.dir
set sec/acl=(IDENTIFIER=FLUCTUATIONS,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) dka100:[prl.trees]fluctuations.dir
set sec/acl=(IDENTIFIER=FLUCTUATIONS,OPTIONS=DEFAULT,ACCESS=READ+WRITE+DELETE+CONTROL) dka100:[prl.trees]fluctuations.dir
! but FLUCTUATIONS rights identifier needs to be defined - WMS112 will
! work, but use H1DATA for now - need to think who this is...should
! not be too wide
! set sec/acl=(IDENTIFIER=WMS112,OPTIONS=DEFAULT,ACCESS=READ+WRITE+DELETE+CONTROL) dka100:[prl.trees]fluctuations.dir
set sec/acl=(IDENTIFIER=H1DATA,OPTIONS=DEFAULT,ACCESS=READ+WRITE+DELETE+CONTROL) dka100:[prl.trees]fluctuations.dir
=======undo wrong ACLS=====
edit /acl DISK$PRLDATA:[PRL.TREES]FLUCTUATIONS.DIR;1 
========then=====
Tricky backup to CHANGE ownerships!
backup/inter GRP_PRL:[WMS112.FLUCTUATIONS]*_35600.*; dka100:[prl.trees.fluctuations]/log/owner=[prl,wms112]
backup/inter GRP_PRL:[WMS112.FLUCTUATIONS]*_356%%.*; dka100:[prl.trees.fluctuations]/log/owner=[prl,wms112]/del

!This one copies them totally unchanged to a temporary place in case of mistakes
backup/nointer GRP_PRL:[WMS112.FLUCTUATIONS]*_356*.*; dka100:[trash.wmsback]/log
This one copies them to their final resting place
$pur/log GRP_PRL:[WMS112.FLUCTUATIONS]*_36%%%.*
$backup/inter GRP_PRL:[WMS112.FLUCTUATIONS]*_360%%.*; dka100:[prl.trees.fluctuations]/log/owner=[prl,wms112]/del
===
$! repeat  for 37, 38 (backup is on DKa200:
$! 38 was not backed up, and I tried to Ctl-Y out, but couldn't near 229 232 233 234 286 290 292 293
$ pur/log GRP_PRL:[WMS112.FLUCTUATIONS]*_38%%%.*
$ backup/nointer GRP_PRL:[WMS112.FLUCTUATIONS]*_38%%%.*; dka200:[trash.wmsback]/log
$ backup/inter GRP_PRL:[WMS112.FLUCTUATIONS]*_38%%%.*; dka100:[prl.trees.fluctuations]/log/owner=[prl,wms112]/del
=== now need to move other trash.wmsback files to dka200 to make room
$ backup/nointer dka100:[trash.wmsback]*.*  dka200:[trash.wmsback]/log/del
=====================================================END OF Move wayne's data 9/00 files=======

$! Clean out deleted variables - always a risk of damage!
TCL CLEAN spectroscopy /shot=-1

$!Save them somewhere for a day or two
$ backup/nointer DISK$PRLDATA2:[TREES.spectroscopy]*_39%%%.*; dka200:[trash.spectroback]/log
Only do the 500 series - these are purged and unlocked
tcl clean spectroscopy /shot=39540/over


Before Clean:---------
SPECTROSCOPY_39540.CHARACTERISTICS;1
                          45/72       20-SEP-2000 14:29:19.12
SPECTROSCOPY_39540.DATAFILE;1
                        6386/6390     20-SEP-2000 14:29:45.28
SPECTROSCOPY_39540.TREE;1
                          39/54       20-SEP-2000 14:25:54.12
After Clean:-----
SPECTROSCOPY_39540.CHARACTERISTICS;2
                          45/54       21-SEP-2000 14:30:18.63
SPECTROSCOPY_39540.CHARACTERISTICS;1
                          45/72       20-SEP-2000 14:29:19.12
SPECTROSCOPY_39540.DATAFILE;2
                          32/36       21-SEP-2000 14:30:18.58
SPECTROSCOPY_39540.DATAFILE;1
                        6386/6390     20-SEP-2000 14:29:45.28
SPECTROSCOPY_39540.TREE;1
                          39/54       20-SEP-2000 14:25:54.12

Total of 5 files, 6547/6606 blocks.

So far only 3954% cleaned in spectroscopy

Wayne's data is not carrying much fat
FLUCTUATIONS_39618.DATAFILE;2  995/1008     21-SEP-2000 14:36:47.70
FLUCTUATIONS_39618.DATAFILE;1 1035/1044     21-SEP-2000 12:06:22.74

Why does ACTMON give <no-node> instead of the node that is being executed?
(middle window of ACTMON)
Answer - the node logical is not defined on this machine?

ACTMON server process has name monitor_server


ne_quick used to call ne_het
 
camac_server seems jammed up?

tcl dispatch/close
PRLAS0$ tcl show server camac_server
28-SEP-1999 16:31:03.48 PRLAS0::camac_server, logging actions, Doing action \H1D
ATA::TOP.SPECTROSCOPY:LAMWAIT:EXEC_ACTION in H1DATA       shot 34547

stop server
$ tcl stop server camac_server
$ typ mds$root:[syserr]prlas0_camac_server.log;-1

n H1DATA       shot 34547
28-SEP-1999 16:17:04.73 Doing action \H1DATA::TOP.SPECTROSCOPY.CAMAC:MOSS_2:INIT
_ACTION in H1DATA       shot 34547
28-SEP-1999 16:17:04.83 Doing action \H1DATA::TOP.SPECTROSCOPY:LAMWAIT:EXEC_ACTI
ON in H1DATA       shot 34547
28-SEP-1999 16:29:54.02 PRLAS0::camac_server, logging actions, Doing action \H1D
ATA::TOP.SPECTROSCOPY:LAMWAIT:EXEC_ACTION in H1DATA       shot 34547
28-SEP-1999 16:31:03.48 PRLAS0::camac_server, logging actions, Doing action \H1D
ATA::TOP.SPECTROSCOPY:LAMWAIT:EXEC_ACTION in H1DATA       shot 34547
28-SEP-1999 16:32:18.08 PRLAS0::camac_server, logging actions, Doing action \H1D
ATA::TOP.SPECTROSCOPY:LAMWAIT:EXEC_ACTION in H1DATA       shot 34547
28-SEP-1999 16:32:28.90 Shutdown request received
%TYPE-F-WRITEERR, error writing SYS$OUTPUT:.;
-RMS-F-SYS, QIO system service request failed



!p.multi=[0,4,4]     
st&for s=33150,33174 do p,'.rf:a14_4:input_1',sho=s&en,loop=25
4.4-4.9 secs to a single X-window, 3.520 seconds, 140800.0us/element to 'Z'
but memory leak of about 3500 blocks per 25 shots!!?? under 5.2

 st&for s=33150,33174 do plot,time,gendata('a14_4_01',time=time,sho=s)&en,loop=25
6.963 seconds, 278520.0us/element 
4.1 secs to 'Z'
print,mds$value("getnci('.rf:a14_4:input_1','length')")
       16732
print,mds$value("getnci('.rf:a14_4:input_1','rlength')")
        3923
datafile is 124/124, contains, a14_6, 4, h18212sl - 18 chans, about 4k ea == 72k
print,mdsvalue('date_time(getnci(".operations:a14_1:input_2","TIME_INSERTED"))')

TCL CLEAN
Note - for recycling a shot, compression is likely to result in
occasional fragmenting of space in the file.  Don't compress, or clean
up after.  Also good for removing replaced data.

Help:  rlength not in MIT hyterptext help for getnci
       can't find info on $shot etc ($constants help is missing or
       refers to non-existant file.



UCX MDSIP services
prlas2$ ucx show serv
FTP                   21  TCP      UCX$FTPD         0.0.0.0             Enabled
MDSIP               8000  TCP      MDSIP            0.0.0.0             Disabled
MOUNT                 10  UDP      UCX$NFS_M        0.0.0.0             Enabled
NFS                 2049  UDP      UCX$NFS          0.0.0.0             Enabled
PORTMAPPER           111  TCP,UDP  UCX$PORTM        0.0.0.0             Enabled
...
prlas2$ ucx enable service mdsip



MDSIP Configuration:
The mappings from remote users to local users is specified in the file:
mds$root:[sysmgr]mdsip.hosts. You put a wild card spec for the incoming
connection using either ip name or ip numbers on the left of the "|" and
a VMS user name on the right. If you get access denied problems then you
should look in mds$root:[syserr]mdsip.log. This will report where the
connection is coming from.

PRLAS2$ sear mds$root:[syserr]mdsip.log conn

Wed May 12 17:15:08 1999 (3) Connection received from bdb112@prlpc3.anu.edu.au [150.203.15.177]
MDSplus GETMSG recv error: connection reset by peer  
Tue Nov 23 05:15:15 1999 (3) Connection received from TWF@cmoda.psfc.mit.edu [198.125.176.13]
Tue Nov 23 05:17:59 1999 (3) Connection received from TWF@cmoda.psfc.mit.edu [198.125.176.13]
Tue Dec 14 17:57:11 1999 (3) Connection received from CAMAC@prlas2 [150.203.15.26]
Tue Dec 14 17:58:09 1999 (3) Connection received from CAMAC@prlas2 [150.203.15.26]
Tue Dec 14 18:09:19 1999 (3) Connection received from bdb112@phypc34.anu.edu.au [150.203.15.123]

==========================================================================================
; Example of remote access from IDL on a PC - need to have the mdsplus/Release dir in path

IDL> mdsconnect,'prlas2::
IDL> mdsopen,'h1data',0
IDL> mdstcl,'show db
% Compiled module: MDSTCL.
% Compiled module: STR_SEP.
 
000  H1DATA                            shot: 35234 [\H1DATA::TOP]   

; No real proof that we are remotely connected - run a timer routine
IDL> .r timer_p
% Compiled module: ST.
% Compiled module: EN.
IDL> st&help,mdsvalue('.operations:i_main')&en
<Expression>    FLOAT     = Array[131071]
15.270 seconds 
; This long delay is typical of TCPIP remote access!!
=======================================================

Procedure to make copies of old data in new files

1: Identify digitizers
csv,'dir s.*_ck
For 8212, find max channel e.g.
csv,'dir s.h18212sl*

2: Identify purpose of each digitizer
csv,'dir/full s.* /output=temp.txt
$sear temp.txt DECLARE
; look at just on digitizer 
$sear temp.txt A14_1,DECLA/mat=AND

e.g. shot 33174
A14_1_01 vern_4 A14 in			OPER
A14_1_02 diamag
A14_1 ch3 ne_het
A14_3 ch1-3:  VERN_1-VERN_3		FIBRE
a14_4 ch1-6 rf a,b,c			RF
a14_5: Moss				MOSS
a14_6: rf d,e,f				RF
H18212SL ch1-9 TFCs,Rog,heme, iong dia  OPER
H18212_1 ch1-4 SLow: RGA, IG, HEME	OPER

ADD NODE/USA=SIGNAL VERN_1
 DECW$SYSTEM_DEFAULTS:L8212_HEADERS.UID - MrmFAILURE

BEWARE of quota of open files - e.g. if add device fails, or setup
device, or save
show proc/quota

:\Program Files\sambar42\docs\mdsplus>java MdsPlusServerDaemon 8000:prlas2:8000

To link into a subtree, logical name ELECTR_DENS

(assuming default is the node where this will live
add node .ELECTR_DENS/usage=structure
set node .ELECTR_DENS/subtree

Notes:
The add/node can be performed from the Edit/Add Child mdenu, the SET NODE
must be performed through TCL.
The leading dot is required.  If left out, the SET NODE/SUBTREE fails
The subtree may be turned off by SET NODE/NOINCLUDE .ELECTR_DENS
but this does not show up in the traverser.  At the moment, John's TC
(h1_manage) does this, but Wayne uses SET NODE/NOSUBTREE to turn his
off, thus making it invisible to John's TC.  Wayne reckons that his
TR612s acquire data even when John's TC turns them off!  However John
is using a TR612 (called, strangely enough, a14_8), which seems to
behave in this mode, though I should double check.





TCL> DECOMPILE .OPERATIONS:H18212_1:INPUT_01
-> all the data as well!

Questions to Josh
Q1/ 

Here is the TCL for adding a subtree (from a separate tree, ELECTR_DENS)

TCL
set default \TOP
add node .ELECTR_DENS/usage=structure
set node .ELECTR_DENS/subtree
Question 1A:  Is this correct?


If I leave out the dot, I get an interesting beast, that has the
subtree icon, but I can't make the SET NODE/SUBTREE work.  
Question 1B: Is this of any use?

add node ELECTR_DENS/usage=structure
: ELECTR_DENS



Question 2:
When I toggle the subtree OFF, it seems to toggle all copies off
(e.g. if John was working on that tree (ELECTR_DENS) he sees the whole
tree off.  This would be annoying if we were taking test shots, say he
was using shot 100000, and I was using 110000 for the whole tree.  Is
there a way around this?
DKA200:[PRL.H1.DATA.PROC.SHOTS]

Windows MDSPLus

TESTTDI
[0:10]
[0,1,2,3,4,5,6,7,8,9,10]

MDSDCL
set command TCL
show db

SET TREE H1data/shot=33174
(error)


Josh Suggests
 TCL('SET TREE name/shot =shot')
but no luck?

http://red.psfc.mit.edu/logbook/signals.htm
 edit DKA0:[SYS0.SYS$STARTUP]ECC_STARTUP.COM


manual check of prlas2 startup
1/ Check that prl_startup and prl_extra are "SUCCESSFUL"
1a/ eccop/show=crates to see if ecc process is running
2/ check that camac_server is defined   def/group camac_server prlas0::camac_server
3/ ucx show serv   then, if disabled,           ucx enable service mdsip
4/ check that a14_1 --> ecc2:n19, h18212SL --> ecc2:n8
$ def/group a14_2 ecc2:n17/group

$ def/group attributes$data  PRLAS0$DKA100:[PRL.JNH112.IDL.DATA_PLOT.MODELS.ATTRIBUTES]



=============Getting prlas0 to work with MDSPLus 4/99, with older VMS===============

$ SET COMMAND MDS$ROOT:[CLD]*.CLD
  Check that logical is defined
$ def camac_server prlas0::camac_server/group
PRLAS0$ show log camac_server
   "CAMAC_SERVER" = "PRLAS0::CAMAC_SERVER" (LNM$GROUP_000112)
  stop the server process
$ tcl stop server camac_server
  <maybe need a bigger hammer, especially of things are amiss  e.g. STOP>
$ STOP ID=c000023
  Load the bridging code
PRLAS0$ tcl
activat/image=mit$devices
dispatch/command/server=prlas0::camac_server activa/image=mit$devices
dispatch/command/server=prlas0::camac_server show db       
exit

  Check if the log is OK - probably is if you got no messages
$ set proc/priv=all
$ typ/tail mds$root:[syserr]prlas0_camac_server.log
 1-MAY-1999 17:49:22.52 doing TCL          command ACTIVATE/IMAGE=MIT$DEVICES
 1-MAY-1999 17:52:11.37 doing TCL          command SHOW DB
Good
Note:  the startup of the activate image can take 20-60 seconds...

 @prl_mds:whohas ecc ecc$dir

Quite often the ECC process (ecc_1365) needs to be stopped
(ECCOP/UNLOAD) AND the camac_server stopped.  Unless camac_server
is stopped and all other processes using the ECC are stoppped, the ECC
can't be restarted.



  Then stop existing scanner, typically on prlas2 (often, Ctl-y doesn't work)
copy tt: h1psupply.tcl
exit
^z
  (trigger data system, so that it finds bogie file)
  delete bodgie file 
  run real h1psupply one time to clear the servers

(Seems that if you put SET CONTROL=(T,Y) in Looptcl.com, it is easier
to control Y out.)


PRLAS2> tcl @h1psupply.tcl
  (trigger, wait for finish)
  Now run a real scanner file on PRLAS0
PRLAS0> tcl @h1psupply.tcl

di/sec mds$root:[sysmgr]CAMAC_SERVER_TCLINIT.TCL 


 

tcl dispatch/command/mon=prlas2::monitor_server show db

prlas0
deass camshr/gr
def/group camac_server prlas0::camac_server


prlas2 timings, 5/5/99
st&x=mdsvalue('.operations:i_main')&en ; --> 0.247 seconds 
st&x=mdsvalue('dim_of(.operations:i_main)')&en ; --> 0.340 seconds 
st&x=mdsvalue('dim_of(.operations:a14_1:input_1)')&en ; --> 0.181 seconds 
st&x=mdsvalue('raw_of(.operations:a14_1:input_1)')&en ; --> 0.055 seconds , .032 seconds 

st&x=mdsvalue('.operations:i_ovf')&en ; --> 0.818 seconds ,0.711 seconds 
st&x=mdsvalue('dim_of(.operations:i_ovf)')&en ; --> 0.830 seconds 




13:00:31.62 Action LAMWAIT:EXEC_ACTION completed
13:00:31.62 CAMAC_SERVER is beginning action .OPERATIONS:A14_2:STORE_ACTION
13:02:36.62 Action .OPERATIONS:A14_2:STORE_ACTION completed
13:02:36.62 CAMAC_SERVER is beginning action .OPERATIONS:H18212SL:STORE_ACTION
13:02:45.08 Action .OPERATIONS:H18212SL:STORE_ACTION completed
13:02:45.09 CAMAC_SERVER is beginning action .OPERATIONS:A14_3:STORE_ACTION
13:03:01.33 Action .OPERATIONS:A14_3:STORE_ACTION completed
13:03:01.33 CAMAC_SERVER is beginning action .OPERATIONS:A14_1:STORE_ACTION
13:03:49.09 Action .OPERATIONS:A14_1:STORE_ACTION failed, status = 134578314
"Incementing shot..."


 status = 134644018       crate not turned on
%ECM-E-UNKNOWN_CRATE, specified crate is unknown on the ether


Latest transforms as of 3/8/99
disecdt shot 34092/83(12T)
v_main 34180, 082 for earlier shots on 5,4
i2t_main - none
i_main 34182



copy_transform, tree="h1data",node='.operations:i_main',sour=34182,from=-1
copy_transform, tree="h1data",node='.operations:v_main',sour=34180,from=-1
copy_transform, tree="h1data",node='.operations:disecdt',sour=34083,from=-1
$ set file/att=(rfm:vfc,fsz=10) h1data$data:h1data_model.datafile;0
$ set file/att=(rfm:fix,lrl:42) h1data$data:h1data_model.characteristics;0
found at node: \H1DATA::TOP.OPERATIONS:A14_2:INPUT_4
No data f
i_main, 34182
Build_Signal(Build_With_Units(-1100 - 4580F0 * .OPERATIONS:A14_1:INPUT_3, "A"), *, DIM_OF(.OPERATIONS:A14_1:INPUT_3))
v_main, 34180
Build_Signal(Build_With_Units(110 * (DATA(.OPERATIONS:A14_2:INPUT_1) - DATA(.OPERATIONS:A14_2:INPUT_2)), "Volts"), *, DIM_OF(.OPERATIONS:A14_1:INPUT_6))
34083, disecdt
Build_Signal(Build_With_Units(-100000F0 * .OPERATIONS:A14_1:INPUT_6, "A/s"), *, DIM_OF(.OPERATIONS:A14_1:INPUT_6))
Build_Signal(Build_With_Units(-1100 - 4580. * A14_2:INPUT_4, "A"), *,
DIM_OF(A14_1:INPUT_3))
isec
Build_Signal(Build_With_Units(-1100 - 4580. * .OPERATIONS:A14_2:INPUT_4, "A"), *, DIM_OF(.OPERATIONS:A14_2:INPUT_4))
wrong

IDL> mdstcl, 'set curr h1data 34429
IDL> mdstcl, 'set curr h1data 34436

Getting new help files from MIT

ftp psfc.mit.edu
cd mdsplus
bin
get tclhelp.hlb-gz


if gunzip won't work, do something like
 
gun:=$sys$login:gunzip-alpha.exe
gun tclhelp.hlb-gz

then copy to MDS$ROOT:

copy/log tclhelp.hlb MDS$ROOT:[SYSHLP]

NODE='.OPERATIONS:I_FAULT'
datesecs=mdsvalue('getnci("\'+node+'","TIME_INSERTED")')
print,mdsvalue('date_time($)',datesecs*2.^16)

Reconciling the Windows and OVMS versions of TREE_VIEW.
To get files form the PC to VMS without losing date/time:
pkzip on the PC, unzip on unix (under camac), then RCP
PRLAS0> rcp -p h-1nf:*.pro *.*

The more common operation is to detect which modules have changed on
the PC, and take new copies to unix ony, where a comparison can be
done with the original files in
/users/prl_h1/prl/jnh112/idl/data_plot/mdsplus
and changes merged over by hand:

(ediff-merge-files-with-ancestor "/users/prl_h1/prl/jnh112/idl/data_plot/mdsplus/mdsplus_data.pro" "/users/prl/bdb112/footemp/mdsplus_data.pro" "/users/prl/bdb112/footemp/old/mdsplus_data.pro")

In the Oct99 update, the PC files originally copied over on
20-May-1999 have changes 
  1/ in the display setup for 24 bit displays
  2/ The IDL command line is improved on the PC,(debug option, more
    info) but separated into its own routine in the VMS version
  3/ Runinfo date fix
  4/ Smart_Table (already incorporated by John)
  5/ TCL: Attempt to write into tranform fails  - do you have write
     permission, pause to debug
  6/ TCL allows to write back into an "undefined" node.
  7/ Add put to wrapper list


backup/delet dka100:[prl.tree_view]cw_tree_view.pro; data_plot:[widgets]*.*;/conf/log

Test crates Oct 99
18,2  (a14)
2,18  (a14)
1,18  (tr612)

r PRLAS2$DKA100:[PRL.ECCCAMSHR]qrep
2,18,10000

r PRLAS0$DKA100:[PRL.ECCCAMSHR]bones
2,18,32000


		Crate 1		Crate 2		Crate 18
BONES(32k)	168kB/s(15/20)	286KB/s(40/40)	168kB/s(30/30)   Mon 11/10@18:39

Also 
@PRL_MDS:showstats all

Magnetic Door Locks:
====================
Station 1 Mod 7
Corridor: 10
Basement:11
RoundHouse:12
EastDoor:9
 
New A14s
636, 637, 638


@prl_mds:everyminute.com

Correct sequence:

edit h1data /shot=34718
put DISECDT IRIPPLE_SEC
close

Some failures (12/10/99)
put .operations:DISECDT_IDL "deriv(mdsvalue('.operations:iripple_sec')"
%TDI$-E-SYNTAX, Bad punctuation or misspelled word or number

put .operations:\H1DATA::TOP.OPERATIONS:DISECDT [0,1] ==> 
(Doesn't like comma)

put .operations:DISECDT .operations:IRIPPLE_SEC
TCL cmd put .operations:\H1DATA::TOP.OPERATIONS:DISECDT  \H1DATA::TOP.OPERATIONS:IRIPPLE_SEC ==> 
put DISECDT IRIPPLE_SEC
TCL cmd put \H1DATA::TOP.OPERATIONS:DISECDT  \H1DATA::TOP.OPERATIONS:IRIPPLE_SEC ==> OK
Command is: close
close

; 3 step slope
plot,mdsvalue("_t=dim_of(.operations:a14_1:input_1); _x=10*_t+115*_t*((_t>-2) and (_t<-1)) + 875*_t*((_t > -1) and (_t<0))")

_t=dim_of(.operations:a14_1:input_1); 
_x=10*_t+115*_t*((_t>-2) and (_t<-1)) + 875*_t*((_t > -1) and (_t<0));
sin(_x)

This appears to do a 125Hz, 3 db BWorth filter
Build_Signal(Build_With_Units(BWFILTER(.0038, 2, .OPERATIONS:TEST_10H_10K), "V [LP 350]"), *, DIM_OF(.OPERATIONS:V_SEC))
atten at 1kHz=0.088/2 = 0.044 - expect 1/64 = 0.0156

ubound(_t) -> 0
lbound(_t) -> -131069
 
plot,mdsvalue("_t=dim_of(.operations:a14_1:input_1);_t1=_t-minval(_t); _t1=_t1/maxval(_t1); _f=100*exp(_t1*9); sin(6.28*_f*_t)")
plot,mdsvalue("_t=dim_of(.operations:a14_1:input_1);_t1=_t-minval(_t); _t1=_t1/maxval(_t1); _f=100*exp(_t1*2.3*2; sin(6.28*_f*_t)")


_t=dim_of(.operations:a14_1:input_1);
_t1=_t-minval(_t); 
_t1=_t1/maxval(_t1); 
_f=100*exp(_t1*2.3*2; 
sin(6.28*_f*_t)

build_signal(build_with_units((_t = DIM_OF(.OPERATIONS:A14_1:INPUT_1);
_t1 = _t - MINVAL(_t);
_t1 = _t1 / MAXVAL(_t1);
_f = 100 * EXP(_t1 * 2.3 * 2);
SIN(6.28 * _f * _t);)
),"foo"),*,dim_of(.operations:a14_1:input_1))



Build_With_Units((_v = .OPERATIONS:V240_RAW;
smooth1d(abs(_v), 100);
), "Volts Mean")

Double smoothed - get artefacts at ends (even with one smooth)
Build_With_Units((_v = .OPERATIONS:V240_RAW;
SQRT(smooth1d(smooth1d(_v ^ 2, 99), 99));
), "V rms")

copy_transform, so=34786,from=34786,to=34788,node='.operations:v240_avg'
copy_transform, so=34786,from=34786,to=34788,node='.operations:v240_rms'
copy_transform, so=34786,from=-1,to=-1,node='.operations:v240_raw'
copy_transform, so=34786,from=34774,node='.operations:v_sec_lpbw'
copy_transform, so=34785,from=-1,node='.operations:v_sec_l'
copy_transform, so=34786,from=-1,node='.operations:test_10h_10k'
copy_transform,sou=34801,from=34787,to=34804,node='.operations:v240_100hp

notes on tree_view
Put a flag in virtual memory warning so you only get it once. (or
later, change the colour of something, that way we don;t need to
dismiss it).

here's a clue:  Your display of freemem doesn;t work if < 100K - that
is weird.

1/ Does plot inc still work?
2/ add title to tree_view widget
3/ Add ylabel to mdsplus_data
4/ IDL error checking in mdsplus_data?

Problems with IDL_NODE
Needs two entries in the tree, hard to update and view without lots of
button pressing - invisible
Maybe use a conglomerate of a trnasform and a string?


John's new viewer

IDL data_plot.ini (in ecc$user)
copy_transform,sour=34836,from=34838,/mode,node='.operations:disecdtl_idl'
copy_transform,sour=34835,from=34822,to=34839,/mode,node='.operations:i_sec'
copy_transform,sour=34838,from=34822,to=34839,/mode,node='.operations:V_SEC_l'
copy_transform,sour=34838,from=34822,to=34849,/mode,node='.operations:V_SEC_lpbw
Gain of Secondary Convertor stepdown
print,(105.1/100)*((1+12.498)/(1/(1/0.498+1/105.1)))
fh=0.503*(1-0.503/105) & print, 105.6/100*(12.27+0.99+fh)/fh  ; sec -ve
fh=0.502*(1-0.502/105) & print, 105.6/100*(12.26+0.982+fh)/fh ; sec+ve (29.05 USE)
copy_transform,sour=-1,from=34910,to=34939,node='.operations:disecdt3'

main convertor gain
fh=0.497*(1-0.497/105) & print, 105.6/100*(99.+0.99+fh)/fh  ; main -ve
      214.520
fh=0.499*(1-0.499/105) & print, 105.6/100*(99.3+0.99+fh)/fh  ; main +ve
      214.306 USE

main convertor gain with extra a14 in parallel
a14=50&fh=0.499*(1-0.499/(a14+5)) & print, (a14+5.6)/a14*(99.3+0.99+fh)/fh  ; main +ve
      226.5 USE

DVM tests of main and secondary voltage dividers 28/10/99

Input Voltage (NOT including 1k) = 14.93V       
        With 1A14  2A14	     0 A14 
Sec  +  556mv	   529	     587
Sec  -  556mV	   
Main +  71.3mv	   67.8
Main -  71.3	   67.8

So the + Main Ratio is 101.5/100.5*(14.93/0.0678) 222.4 (2 A14s)
211.5 (1A14)

and +Sec ratio (12.5+1)/12.5*(14.93/0.556)
29.0

copy_transform,sou=34996,/model,from=34989,to=34999,node='.operations:r_main
copy_transform,sou=34998,/model,from=34989,to=34999,node='.operations:i_main
copy_transform,sou=34996,/model,from=34989,to=34999,node='.operations:v_main_l
copy_transform,sou=34996,/model,from=34989,to=34999,node='.operations:v_main   

Changed over to two a14s (and new digitizer) at shot 35014 - use two
a14s until 34057, so gain=222.4

copy_transform,sou=35057,/model,from=35014,to=35057,node='.operations:v_main'

4t, auto time scale
copy_transform,sou=34955,/model,from=34954,to=35057,node='.operations:I_OVF'

copy_transform,sou=35065,/model,from=35014,to=35063,node='.operations:r_main'

investigate decreases in memory when MDSPLus is accessed
help,mdsvalue('_x=(.operations:a14_1:input_2);_x[-5:-1]')& default, om, freemem()&print,freemem()-om&om=freemem()

Seems like tree view gobbles about 2mbyte for each display of a 131k
sample signal - the room used shold be only 500k, and shold start to
get recycled.

This doesn't help much!
IDL/ARRAY_MEMORY=(INITIAL=2048, EXTEND=1024)

Data_plot bugs:
=================
need to manually quote label when editing attributes


2/11/99 copied vax versions of *.com; under prl_mds to [ecc.prl_mds]
This needed to be updated in scu$prl:[prl112]prl_startup.com

Where is the proper copy of this directory??

Find all I_fault trips
shotscan,from=33482,to=35102,diag='.operations:i_fault',int=[-2,-1],sx=[-10,10],op='MAX',xdi=''
SHOTSCAN_OPERATIONS_I_FAULT33482-3000.DAT;1
 
34168       34237       34239       34240       34243       34244       34245       34258       34278       34379
34399       34408       34438       34502       34540       34563       34635       35058       35095       35096
35102

shotscan,from=33200,to=35135,diag='.operations:i_fault',int=[-2,-1],sx=[-10,10],op= 'v=max(smoothn(time=t,d,100))',xdi='',ps=10,/replot,yr=[1e-3,10],xr=[200,1200]+1000

copy_transform,sou=35129,from=3 5107,to=35114,node='.operations:dimaidtl_idl'
copy_transform,sou=35129,from=35107,to=35114,node='.operations:disecdtl_idl'

Very nice ramp of 20A in 200ms

shot, 35169
irip=gendata('.operations:iripple_main',time=time)
plot,time,irip+3.3*integr(time,irip+6,-127),yr=[-20,30],xr=[-4,-1]
plot,time,4014+irip+3.3*integr(time,irip+6,-127),yr=[-10,25]+4000,xr=[-3.5,-1.5],tit=runinfo()+'Main supply ramp, 4000->4020 Amps'

Shot, 35168

Old ECCDIR
"ECCDIR" [super] = "PHYVS2$DKA100:[ECC.]" (LNM$GROUP_000112)
New EccDIR
   "ECCDIR" = "PRLAS0$DKA600:[ECC.PRL_MDS.]" (LNM$GROUP_000112)

edit DKB0:[SCU1.SYSMGR]ECC_STARTUP.COM this appears to be the active ecc file?


need to update stpl and stplnew in 
/users/prl_h1/prl/idl_new.dir/since20jul.dir/stplnew.pro
to remove eccdir - not clear how we want to replace it - perhaps
findfile or findpath e.g.
 print,filepath(root=getenv('prl_mds'),sub='stpl','foo')
PRLAS0$DKA600:[ECC.PRL_MDS.stpl]foo
gET this file (it came from h1d.com or h1data.com?
  "IDL_STARTUP" = "ECC$DIR:QUICKXSTPL.PRO"
Put a comment in the stpl default file?  will print nothing if no comment?

looptcl looks at
def/group MDSPLUSMODE CONTINUOUS

****** see **************
HELP HINTS sec assign prot

Boyd has to undo his hints!!!!

set file/acl/def/like=h1data$data:h1data_model.datafile disk$prldata2:[trees]*.dir/log
set file/acl=(IDENTIFIER=H1DATA,ACCESS=READ+WRITE+EXECUTE+CONTROL)/def disk$prldata3:[trees]*.dir/log

set sec/acl=(IDENTIFIER=H1DATA,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) disk$prldata2:[trees]*.dir/log


doesn't seem to add acls?

set file/acl= (IDENTIFIER=H1DATA,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) electr_dens$data:electr_dens_35639.*/log

no

set file/acl/like=h1data$data:h1data_model.tree  electr_dens$data:electr_dens_35639.tree/log

set sec/acl=(IDENTIFIER=H1DATA,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) disk$prldata2:[trees]*.dir/log

(no default - works?)

set sec/acl=(IDENTIFIER=H1DATA,options=default,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) disk$prldata2:[trees]*.dir/log

This works!!!!!
set sec/acl=(IDENTIFIER=H1DATA,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) electr_dens$data:electr_dens_model.*
========

Rationalize on Monday!!!
********THIS ONE ACTUALLY WORKS 2/2001********** 
set sec/acl=(IDENTIFIER=H1DATA,options=default,ACCESS=READ+WRITE+EXECUTE+CONTROL) disk$prldata3:[prl.trees]*.dir/log
*****************************************

create/dir dkb100:[ptrl.trees.h1data]
set sec/acl=(IDENTIFIER=H1DATA,options=default,ACCESS=READ+WRITE+EXECUTE+CONTROL) disk$prldata4:[prl.trees]*.dir/log
create/dir dkb100:[prl.trees.spectroscopy]
set sec/acl=(IDENTIFIER=SPECTROSCOPY,options=default,ACCESS=READ+WRITE+EXECUTE+CONTROL) disk$prldata4:[prl.trees]spectroscopy.dir/log
set sec/acl=(IDENTIFIER=FLUCTUATIONS,options=default,ACCESS=READ+WRITE+EXECUTE+CONTROL)  disk$prldata3:[prl.trees]fluctuations.dir/log
*****************************************

The MDSPlus system passes on acls by copying from the model files.
 

backup/nointer DISK$PRLDATA3:[PRL.TREES.SPECTROSCOPY]s*model.*;* DISK$PRLDATA4:[PRL.TREES.SPECTROSCOPY]/del/conf


Good rsync line ofrm Tom
rsync -avz lost.pfc.mit.edu::MDSplusDoc d:/mirrors/MDSplusDoc
first time 3.7Meg to create 16Meg
second time 6351 bytes to check up to date


Problems with MDSPLus Crates - link-partner-exited


                Host Version 7.3

                Crate   Firmware        Remote Address
                1       7.8             008003130058
                2       7.8             00800313026F
                18      7.6             008003130056
PRLAS0$  tcl
TCL> activat/image=mit$devices
TCL> dispatch/command/server=prlas0::camac_server activa/image=mit$devices
%SYSTEM-F-LINKEXIT, network partner exited
TCL> dispatch/command/server=prlas0::camac_server show db       



copy_transform,sour=-1,from=35757,to=35765,node='.log.machine:comment'
SPECTROSCOPY.CAMAC:MOSS_3:INPUT_1

set sec/acl=(IDENTIFIER=bdb112,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) idl_local.dir


Problems when rationalizing operating systems

mds server account not valid on prlas0
 [PRL,MDSSRV]

PRLAS2$ eccop/sho=cra
(Works on prlas0: get teo different error messages on PRLAS2??)

%DCL-W-ACTIMAGE, error activating image HYTSYSSHR
-CLI-E-IMGNAME, image file PRLAS0$DKA0:[SCU1.ECC]HYTSYSSHR.EXE;1
-SYSTEM-W-SYSVERDIF, system version mismatch; please relink

%DCL-W-ACTIMAGE, error activating image HYTSYSSHR
-CLI-E-IMGNAME, image file PRLAS0$DKA0:[SCU1.ECC]HYTSYSSHR.EXE;1
-SYSTEM-F-PROTINSTALL, protected images must be installed

$!========= Simple test sequence for PRLAS machines===========
$!May need group privilege

$! Simplest MDS - no response is correct
ccl set mod h1crate

$! Test TCL - shows the current H1 shot
tcl show current h1data

$! Test MDS servers - should show the server
tcl show server prlas0::camac_server

$! Test ECC server - show list a few crates (only PRLAS0 and 2)
eccop/sho=crates


58:19 begin trigger
59:00 begin sec phase
59:14 begin new shot
59:24 begin int
   41 end
Total cycle of 82 seconds
8.6MB compressed to 1.4MB in 42 seconds total --> 203kBsec
maybe 72 seconds with one server? also 47 secs another time

Test TR612 - works
set tree fluctuations/sho=38219
do/method .camac:tr612_4 init
spawn ccl
set mod tr612_4
pio/f=25/ad=1   
exit
do/method .camac:tr612_4 store
SW 32kSW 32kSW 32kSW 32kSW 0k
%TREE$-E-NOOVERWRITE, write-once node already contains data, no overwrite permi

Still doing 128kW fetches per channel

Fixed 25/5/00!!!!!!!!!!!!
SW 26k
%TREE$-E-NOOVERWRITE, write-once node already contains data, no overwrite permit
ted

FTP site for new mit$devices

alpha1.psfc.mit.edu/MDSPLUS/AXPVMS/SYSLIB