| ID | Date | Author  | Topic | Subject | Text |  | 
| 112 | 01 Nov 2003 | Stefan Ritt |  | Do not frob | > I found where we tickle the race condition in db_create_record().
 >
 |  | 
| 108 | 01 Nov 2003 | Stefan Ritt |  | more odb | > I added error checking to the places where we read "/runinfo/run number". In
 > general, I do this:
 |  | 
| 110 | 14 Nov 2003 | Stefan Ritt |  | more odb | Ok, I apologize. It's all ok. Thanks for clearifying. Concerning the assert's, it
 would be nice to be able to disable them
 |  | 
| 104 | 16 Nov 2003 | Stefan Ritt |  | Phantom | I have seen the same behaviour and it annoys me, too. What I did in the past
 is a "cleanup" in ODBEdit which removes these
 |  | 
| 98 | 17 Nov 2003 | Stefan Ritt |  | Revised MVMESTD | Let me propose a revised scheme for midas standard VME calls (mvmestd.h).
 
 |  | 
| 82 | 20 Nov 2003 | Stefan Ritt |  | Implementation of db_check_record() | As Konstantin pointed out correctly, the db_create_record() call is pretty
 heavy since it copies whole structures around
 |  | 
| 105 | 20 Nov 2003 | Stefan Ritt |  | Phantom | I tried to reproduce the problem, but without success. So in case this happens
 again, one should debug the code im cm_watchdog()
 |  | 
| 93 | 20 Nov 2003 | Stefan Ritt |  | cannot shutdown defunct clients | > 1) shutdown from mhttpd "programs" page -> "cannot shutdown client"
 > 2) "sh mhttpd" from odbedit ->
 |  | 
| 102 | 20 Nov 2003 | Stefan Ritt |  | set-uid-root midas programs | > dio- is required to be setuid-root to gain I/O permissions. I looked at it a
 > few times, and it is probably safe, but
 |  | 
| 95 | 20 Nov 2003 | Stefan Ritt |  | cannot shutdown defunct clients | > INT == "int", wraparound in 1 month > DWORD == "unsigned int", wraparound in
 2 months
 |  | 
| 100 | 21 Nov 2003 | Stefan Ritt |  | Revised MVMESTD | Thanks for your contribution. Let me try to map your functionality to mvmestd calls:
 
 |  | 
| 97 | 24 Nov 2003 | Stefan Ritt |  | cannot shutdown defunct clients | > But there is one problem with "cleanup". It has a hardwired timeout of
 > 2 seconds.  This is a problem for tasks
 |  | 
| 90 | 30 Nov 2003 | Stefan Ritt |  | bad call to cm_cleanup() in fal.c | > fal.c does not compile: it calls cm_cleanup() with one argument when there
 > should be two arguments. K.O.
 |  | 
| 85 | 30 Nov 2003 | Stefan Ritt |  | Implementation of db_check_record() | Fixed and committed. Can you check if it's working?
 |  | 
| 88 | 01 Dec 2003 | Stefan Ritt |  | delete key followed by create record leads to empty structure in experim.h | > I have noticed a problem with deleting a key to an array in odb, then
 > recreating the record as in the code below.
 |  | 
| 81 | 12 Dec 2003 | Stefan Ritt |  | db_close_record  non-local/non-return | Hi Paul, 
 sorry my late reply, I had to find some time
 |  | 
| 79 | 12 Dec 2003 | Stefan Ritt |  | Several small fixes and changes | I committed several small fixes and changes: 
 - install.txt which mentions explicitly ROOT
 |  | 
| 74 | 15 Dec 2003 | Stefan Ritt |  | Poll about default indent style | Dear all, 
 there are continuing requests about the C
 |  | 
| 4 | 18 Dec 2003 | Stefan Ritt |  | Alarm on no ping? | > I want midas alarms to go off when I cannot ping arbitrary remote hosts. Is
 > there is easy/preferred way to do this?
 |  | 
| 76 | 18 Dec 2003 | Stefan Ritt |  | Poll about default indent style | Hi Paul, 
 I agree with you that a nesting level of
 |  |