Apologies if this isnt the correct forum.
Running Pick Universe 10.1.6.
We're getting a message like this:
cbarnum Unable to execute trigger as trigger creator "hvyuser".
Following this is a write failure on the particular file involved.
Its happening only on 2 files, of the 60 or so that are t
i get the same error, unless i use 'FOR EACH ROW CALLING' rather than 'CALLING'.
Also, the 'called' trigger needs to be globally cataloged, ie '*UPCASE.IT'
j
On Thu, Nov 20, 2008 at 10:31 AM, Tom Dodds <[EMAIL PROTECTED]> wrote:
> Does anyone know if Triggers are supported under the PICK flavor o
A nice multivalue subroutine for interactive screens, that takes row
and column start points and headings for associated attributes,
conversion and formatting of same, number of data rows displayed, and
(A)dd, (I)nsert, (D)elete, (P)age, (E)nter (for next or field), and
(S)ave prompts would be a ni
A cruel trick using a no-op and VOCLIB?
On 4/30/08, Baruch Salamander <[EMAIL PROTECTED]> wrote:
> Greetings,
> Did anyone ever experience the following on Universe?
> >BASIC BP TEST
> >
> The program didn't compile in TCL and no error messages appeared after
> pressing return. The sam
Since we're talking OOB (out of box) why not try the user semaphore
locking that is built into the system. Check the BASIC LOCK and UNLOCK
, and LIST.LOCKS in tcl. Your process cant write until it can lock. It
writes, then unlocks. and so on.
j
On 4/25/08, Marco Manyevere <[EMAIL PROTECTED]> wrote
this is a test only. sorry, please ignore.
On Sat, Apr 19, 2008 at 12:08 PM, Dave Taylor <[EMAIL PROTECTED]> wrote:
> Well, here it is Saturday morning and my long-anticipated weekend project of
> installing Universe on Linux has just come to a screeching halt.
>
> I downloaded RHEL 5.1 Desktop an
It appears that ours is equally slow or fast depending on one's frame
of reference, unless I am missing something...
This is test purposes only... ten million iterations.
CRT OCONV(TIME(),'MTS')
A = "ABCDEFGHI"
B=""
FOR X = 1 TO 1000
B<-1>=A
NEXT X
CRT DCOUNT(B,@FM)
CRT OCONV(TIME(),'MTS')
B
For number one, you could also open the file in BASIC and do a
VAR=FILEINFO(,#) where number is any number between 1 and
about 15. File type is returned with a var of 4 in our system.
On 2/6/08, john reid <[EMAIL PROTECTED]> wrote:
> 1. we use ANALYZE.FILE .. it gives the scoop on
1. we use ANALYZE.FILE .. it gives the scoop on anything
not a directory and a little bit on dynamic files.
2. That's an indication of some type of memory leak or stack overflow
I think. I'm pretty sure its an abnormal condition. You could
probably duplicate it in a sub calling itself until it g
what would be an example of a windows batch file, that one would want
to run? Maybe a small example, like what the batch file would look
like, would be great. It sounds interesting.
john
On 1/24/08, Amy Raisanen <[EMAIL PROTECTED]> wrote:
> My goal is to run a Universe program that creates a Windo
uot;:FILENAME:@(-4)
> END
> END
> RETURN
>
> In use, you CALL OPEN.FILE.SUB("MYFILE",FILEVAR,ERR) ; IF ERR THEN
>
> Interestingly, on D3 this does not seem to have much benefit, where the OPEN
> function seems faster than the LOCATE.
>
> /Scott Ballinger
> Pare
I just saw this today, and although i am certain that everyone must
know it already, i am including it.
For itypes that call subroutines, where the subroutines open lots of
files, put a named common in the subroutine that contains what amounts
to dimensioned arrays that hold the file vars. Put a sw
> >>To: u2-users@listserver.u2ug.org
> > >>Subject: RE: [U2] Universe 10.1 Itype possible parsing problem
> > >>
> > >>What it is doing is evaluating as:
> > >>If @record is less than 3 is greater-than or equal to @record<6>
> then
> >
I appreciate the feedback. Thanks all.
john
On 1/9/08, Herve Balestrieri <[EMAIL PROTECTED]> wrote:
> This interesting "feature" is reproduced also at release 9.5.1 of UniVerse
> (and likely from the beginning), but this one was not yet referenced at
> IBM nor its predecessors.
> Note that compili
In the following query, a file is listed with two identical itypes,
save for field 2 spacing, as shown below. Fields 3 and 6 are all equal
in the query shown.
The difference is that there is no space between the > and = . This
appears to yield wrong results, the bad one being shown as field 2
(all
I read a number of questions where I cant determine if I have more
experience or less experience than the questioner. This is one of
those. I'm not aware of an oconv that extracts a value, so I have to
go with CRT VAR1<1,1> rather than using the OCONV, knowing that there
is probably some OCONV that
We had the same problem with a scheduling module. We found that if the
module CHAINed itself after a given period of time (like an hour) the
zombies went south, following the deceased parent.
On 11/5/07, Gunasegeran Pasupathy <[EMAIL PROTECTED]> wrote:
> Dear David,
>
> Try setting the NOTIFY ON
Group,
I think that I have sent this before, so apologies. We have a number
of RedBack processes which seem to be unable to properly do an EXECUTE
' XXX ' CAPTURING OUTCOME
The error message is :
Wed Oct 31 14:46:53 0 RedBack Program "XPRINT.REMIT":Line 95,
Message[010291] which translates to
It appears that 'EXPLAIN' is somewhat documented via an internet
search, implying that it might be standard equipment for at least some
SQL's in order to show query plans. I did find a file on our system
named SQL.HELP. A search revealed 6 records containing 'EXPLAIN' but
it also contained a lot o
Universe 10.1.6
Universe has a 'revolving' unix errlog file...
Occasionally a PHANTOM-executed ITYPE will cause a message to be written
to the errlog, stating:
Tue Oct 16 07:14:53 0 Program ".ITYPE.": pc = 38, Message[040025]
with a 'real' user name of course.
which translated means 'nonnumer
The previous responders certainly know alot more about this stuff than
I and I they mentioned all of the things to do. However, if I had to
do just ONE thing, I would certainly pick the separation expansion to
4. The way I read it, the average group contains 1/3 of a logical
record, so expanding t
It MAY be the case that there are 'unexpected' data types, in one,
possibly even two of our files. Occasionally (maybe 100 times per
second), a PHANTOM executed ITYPE will cause a message to be written
to the errlog, stating:
Tue Oct 16 07:14:53 0 Program ".ITYPE.": pc = 38, Message[040025]
I e
LIST.ITEM and SELECT * FROM can also be useful for q&dirty.
On 10/16/07, roy <[EMAIL PROTECTED]> wrote:
> Is F11 in the VOC ?
>
> You may need to create it!
>
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Brutzman, Bill
> Sent: Tuesday, October 16
and the FILE.STAT?
On 10/16/07, roy <[EMAIL PROTECTED]> wrote:
> Random reads and updates on a file with ~2 million records. I separated the
> reads and writes to a separate program that only does this processing to no
> avail.
>
> Topas shows 100% disk usage during this process and all other use
s".
>
> It's a wonderful little tool. You can create your own
> application-specific
> utility verbs that create select lists to be acted on by the next
> command.
>
> cds
>
> -Original Message-
> From: john reid
> Sent: Wednesday, August 08, 2007
Once again at risk of stating the already known...
The 'k' causes the end of the execution to retain the 'KEYS' from a select list?
j
On 8/8/07, Ray Wurlod <[EMAIL PROTECTED]> wrote:
> K mode means that the Verb (the Catalog entry) has "K" in field #4.
> It may be the answer you need.
> Reference
maybe just OPEN "","" TO .FILE
On 7/27/07, john reid <[EMAIL PROTECTED]> wrote:
> Slick...
> j
>
> On 7/26/07, Charles Barouch <[EMAIL PROTECTED]> wrote:
> > All,
> > I just put up a new (free) tool on my site. It
> > takes very littl
Slick...
j
On 7/26/07, Charles Barouch <[EMAIL PROTECTED]> wrote:
> All,
> I just put up a new (free) tool on my site. It
> takes very little time to use and I'd appreciate it if some of you would
> beat on it and let me know what you think.
>
>
> http://www.keyally.com/fasterskel/Main.php5
>
>
OUCH!
Probably not without looking at the manual though..
:-)
On 6/29/07, MAJ Programming <[EMAIL PROTECTED]> wrote:
Aside from the historical rambling, your question on LOCATE can be easily
answered tenfold by the persons on this forum.
Mark Johnson
- Original Message -
From: <[EMAIL P
'Must be able to mutitask'
Is that like having the capability to alter the biocharacteristics of
a task, sort've on the fly?'
Just kidding
john
On 6/28/07, Barry Rogen <[EMAIL PROTECTED]> wrote:
Position: Full time, permanent U2 programmer
Environment: UniVerse, resi
Great minds
Although I find that if I have it CHAIN itself every hour, it doesnt
leave its messy orphan children around.
j
On 6/26/07, David A. Green <[EMAIL PROTECTED]> wrote:
David,
What I did was to write a Master Scheduler Program that runs as a phantom
and phantoms off the background p
program has been cataloged, though, you can do
PHANTOM HI
And the first "_" field of the id would then be HI instead of RUN.
You can then use the program name to find the ID in the &PH& file.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf
Universe 10.x
PHANTOM SQUAWK RUN JMRBP HI
RUN_36022_14417 record has been created in the '&PH&' file.
Phantom process started with process ID 7408.
On 6/21/07, Kevin King <[EMAIL PROTECTED]> wrote:
Is there a way for a phantom to know the _PH_ item that it's output is
being redirected to?
-
I didnt attempt to create on the 'master' in universe 10. I wrote an
application to find every pathname, created my own VOC from that
location, and created the trigger on that name. My approach would also
not have a trigger in place for an addition to the df. To the good,
the application creates C
ng
everything in the IS department because he/she was the only one in the
department.
- Original Message -----
From: "john reid" <[EMAIL PROTECTED]>
To:
Sent: Thursday, June 07, 2007 8:54 AM
Subject: Re: [U2][EMPLOYMENT]
> Not talking just one thing though. Foremost is
this former IT manager.
j
On 6/7/07, Jerry <[EMAIL PROTECTED]> wrote:
Sounds normal for this business. Every job I've ever had was similar, never
had a job where I did only one thing all day every day.
- Original Message -----
From: "john reid" <[EMAIL PROTECTED]>
To:
Maybe I read the job description wrong, but it seems that you are
searching for one person to do just about everything that is involved
with keeping a system running. Administration, design, DEVELOPMENT,
and the ability to work all hours, not to mention having time to
'fluidly' communicate with pr
I have a feeling that this one could go on a bit. Far be it for me to
break the chain...
We have separate accounts that exist solely for the purpose of users
who have tcl access. Within those accounts, is a minimum of
potentially damage causing verbs.
But, if security is a BIG priority, the most
Here 'Trigger! Sounds like it may be time for some type (really
really simple) of update audit.
j
On 5/21/07, Pamela J Robbins <[EMAIL PROTECTED]> wrote:
Has anyone seen a problem like this? The user claims they changed a
record, then checked to make sure it actually changed. They find
every
Not sure if sql is an option for you... but there is a union construct
in it. Both files have a '1' id. The ids are not shown in second
example.
j
SELECT @ID,DT FROM DAILY UNION SELECT @ID,DT FROM MONTHLY;
DAILY.DT1...
1 12-30-2001
1 08-12-2010
2
cation.
--Bill
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of john reid
Sent: Friday, May 11, 2007 3:04 PM
To: u2-users@listserver.u2ug.org
Subject: Re: [U2] Locks
I didnt see any response to this. Did I miss something?
john
On 5/10/07, Brutzman, Bill &
I didnt see any response to this. Did I miss something?
john
On 5/10/07, Brutzman, Bill <[EMAIL PROTECTED]> wrote:
How do I rollback a U2 transaction?
Right now, all I know to do is to restore the entire data file from "tape".
--Bill
---
u2-users mailing list
u2-users@listserver.u2ug.org
The order I use is to build the key by using a routine, then cut the
KEY record lock loose, followed by a RECORDLOCK on the transaction to
be written.
We do a similar function, and have not seen signficant wait times.
If you're already doing that, I'm not sure what could be occurring.
john
On 5
re:REMOVE ...
being careful of course to remember that values and subvalues get
fetched as a part of the REMOVE. That is to say, you may not always
get a 'whole' attribute.
john
On 5/1/07, Dave Davis <[EMAIL PROTECTED]> wrote:
Look at the "REMOVE" command.
It doesn't actually delete anything,
Jef,
can you describe what you mean by 'process', ie do you mean to 'make
available to perform operation(s) upon; to see individually ;etc?
john
On 5/1/07, Jef Lee <[EMAIL PROTECTED]> wrote:
Some time ago, someone posted code showing a very fast way to process
all the attributes in a record.
Universe 10.1.6 Pick
The below 'does nothing' program yields the cross reference beneath it
using the
BASIC TEST.XREF -X
command.
Would anyone know of a way to cause the whole variable and label names
to show on the print file located in the .L directory,
rather than the 'truncated to 15 charac
That looks really cool, eh?
On 4/17/07, john reid <[EMAIL PROTECTED]> wrote:
I could be missing something, but to display the multi's, build an
i-descriptor that looks like these column heads for S1 and SS2.
LIST BIG.INDEX TOTAL F2 TOTAL S1 TOTAL SS2 02:02:03pm 17 Apr 2007
I could be missing something, but to display the multi's, build an
i-descriptor that looks like these column heads for S1 and SS2.
LIST BIG.INDEX TOTAL F2 TOTAL S1 TOTAL SS2 02:02:03pm 17 Apr 2007 PAGE1
BIG.INDEX. F2 SUM(F2)... SUM(SUM(F2))
TEST13317 13
Did you compile the program?
I get this when write a program, then catalog it without compiling:
Unable to open "JMRBP.O/NOT.COMPILED" for cataloging.
On 3/7/07, Brutzman, Bill <[EMAIL PROTECTED]> wrote:
I have had better luck with '*' option below... as the '*' uses the global
catalog.
--Bill
We have some items that make life easier, but they all rely upon a
periodic build of a so-called 'INODE.XREF' file. If you can live with
that, we can help. Maybe someone else has something better though.
j
On 1/3/07, Mike Pflugfelder <[EMAIL PROTECTED]> wrote:
Does anyone have a program writte
Tried it... works.
j
On 12/1/06, Clifton Oliver <[EMAIL PROTECTED]> wrote:
I've not tried it, but triggers should work on a dictionary. I am
assuming that is what you meant, not work on a directory level that has
a type 18 dictionary.
Gordon J Glorfield wrote:
> But would triggers work on a d
Alive and well at 10.?.
On 11/28/06, chuchi <[EMAIL PROTECTED]> wrote:
Hi all!!
Someone know if exists triggers in Universe??
Thanks all!!
Pulse para ir al sitio web
JesC:s Barbero RodrC-guez
Departamento de AnC!lisis y ProgramaciC3n - Desarrollo tecnolC3gico
Zoco Gran Santander, 1B* Planta
I know that I sometimes have problems if after building the index, I
dont issue an
ENABLE.INDEX to handle subsequent updates to the file. - universe 10.1
j
On 11/16/06, John Jenkins <[EMAIL PROTECTED]> wrote:
If someone has the file handle already open then you will be compromising
the file as
Its not a proc but...
VOC item:
PA
TERM 100,1000
DIVERT.OUT ON &HOLD& <> TTY.OFF
LIST VOC SAMPLE 10 WITH *A1 LIKE V...
DIVERT.OUT OFF TTY.ON
CT &HOLD& <>
TERM 80,24
If you enter the unique name on the command line, its referenceable thruout.
For example "GSJ littleolme" yields:
LIST VOC SAMPLE 1
Original message --
From: "john reid" <[EMAIL PROTECTED]>
> Sorry for the misunderstand about the hash file, and dir.
> j
>
> On 10/11/06, [EMAIL PROTECTED] wrote:
> > The trigger is not ON a directory it is on a hash file where the trigger
is
> a
This is the first DIR
type that the outside has access to so..
It DOES work successfully provided the person is not coming in from the
outside. Hence my questioning buffers, effects of ODBC connections
--
Debster
-- Original message --
From: "john reid" &
Do you have triggers on other directories?
I get the following when attempting to create a trigger on a
directory, in Universe:
CREATE TRIGGER TESTJ AFTER INSERT OR UPDATE OR DELETE ON JMRBP FOR EACH ROW CALL
ING '*MAIN.TRIGGER';
UniVerse/SQL: Invalid file type specified.
Warning: SICA information
re
all sent out @ the same time or before the message was fully sent to all
users, how long should one wait before deciding that no one else
responded (minutes, hours, days)?
dougc
john reid wrote:
> No offense intended, and probably none of my business, but we could
> probably cut down on
No offense intended, and probably none of my business, but we could
probably cut down on activity if responders would read any previously
sent comments. In this particular case, several responses were very
similar in content.
On 10/5/06, Ray Wurlod <[EMAIL PROTECTED]> wrote:
To be a comment in
In Universe...
You could:
SELECT FILE WITH NAME = "<>]"
or include it in what the user enters for example "[JONES]"
Quitting when no records: I uses GO but I think that there is a gosub.
TEST.SELECT
0001 PA
0002 SELECT PD WITH @ID = "XX"
0003 IF @SELECTED > 1 THEN GO BEYOND
0004 GO END
0005 BEY
This has always confused me. This is the some output of the program
shown. Is that correct, or am I completely missing the barge?
At the risk of a rule violation, I am attaching the picture. Please forgive.
FLD:2
0 33 76 101 119 105 115 47 32 77 99 77 117 114 114 121 47 87 105 108
108 109 97 110
btw, why would you want to go back to the repeat when you get a null?
You've used up the SELECT list, right? Also, the NEXT might be
misleading to the next lucky person. NEXT Part.Count might be easier
to read.
On 7/27/06, john reid <[EMAIL PROTECTED]> wrote:
Gotcha
TESTL: 28:
Gotcha
TESTL: 28: clear
:: F.Item/
DBFILE: A IM Mod=1 Sep=4 Type=18
:: s
TESTL: 33: done = false
:: F.Item/
INTEGER: 0
::
:: F.Item/
DBFILE: A IM Mod=1 Sep=4 Type=18
:: s
TESTL: 33: done = false
:: F.Item/
INTEGER: 0
::
On 7/27/06, Brutzman, Bill <[EMAIL PROTECTED]> wrote:
What's a 'cracker'?
On 7/18/06, Jerry Banker <[EMAIL PROTECTED]> wrote:
Exactly.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Dave Walker
Sent: Tuesday, July 18, 2006 1:48 PM
To: 'u2-users@listserver.u2ug.org'
Subject: RE: [U2] SOX question (United S
Must have something to do with the conversion... 'cause this seems to work.
j
SELECT DTTEST FROM VOC WHERE VOC.DTTEST EQ (SELECT TODAY FROM TEST ) EXPLAIN
SQL+;
Optimizing subquery block 1
Driver source: TEST
Access method: file scan
UniVerse/SQL: Executing non-correlated subquery block 1
Opti
LONGNAMES ON in the account seemed to help us.
j
On 6/13/06, gerry-u2ug <[EMAIL PROTECTED]> wrote:
Stupid brainfart question (?) :
we had things configured so that CREATE.BFILE created type 19 files.
sometime recently the command has reverted to creating type 1 files.
I can't remember how we g
any includes?
what was entered when the message rendered?
On 6/2/06, Kevin King <[EMAIL PROTECTED]> wrote:
That would be my guess, that the line number isn't right.
-Kevin
[EMAIL PROTECTED]
http://www.PrecisOnline.com
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED
Many of those problems were fixed when card readers went to a more
elaborate belt feed, as they were 're' flattened prior to the read
station. The same technique was tried in early belt-fed weapons, to
the chagrin of the dear departed testers.
They're gonna throw us out of this list...
On 5/28/
You guys had punch cards? We had to use Sears catalog pages, cut to
the proper size
Maybe we both need a good nights sleep
j
On 5/28/06, HENDERSON MIKE, MR <[EMAIL PROTECTED]> wrote:
Groan!
A, punched cards, it takes me back
Mind you, in my day we couldn't afford ... Etc
:-)
> -
Not too familiar with sb, but if it is re-opening the file, then any
record locks are lost. A workaround MIGHT be to issue a RECORDLOCKU in
the subroutine prior to the write.
j
On 5/26/06, Charles Barouch <[EMAIL PROTECTED]> wrote:
I'm getting this error and I don't know how to fix it. A full re
compare for any character less than char(31).
Found a bunch, too!
j
On 5/26/06, john reid <[EMAIL PROTECTED]> wrote:
Not sure how spaces enter into it, although I see that it yields the
same in the following with or without spaces.
j
0001 EXECUTE 'CT JMRBP TEST.FOLD'
0002
ginal Message-
> From: Brad Davis
> Sent: Thursday, May 25, 2006 11:40 AM
> To: u2-users@listserver.u2ug.org
> Subject: RE: [U2] FOLD can mutilate
>
>
> If fold cannot find a space to break on, what does it do? I think it
> breaks the fold length, so it can break a string.
>
Anyone notice the length problem on the fold of spaces?
TEST.FOLD
0001 EXECUTE 'CT JMRBP TEST.FOLD'
0002 A = SPACE(10)
0003 CRT LEN(A)
0004 CRT 'A:':A
0005 B=FOLD(A,1)
0006 CRT 'LEN B:':LEN(B)
0007 CRT 'B:':B
0008 STOP
10
A:
LEN B:-1
B:
--
john
--
Failing a proper log file, is there the equivalent of the -TRAP run
time break to debug when an error occurs? It might not be do-able in
production environment but maybe for selected users.
On 5/2/06, Dave Walker <[EMAIL PROTECTED]> wrote:
I'm sure we've all seen the user that ignores umpteen "
I know that you didnt ask about Universe... but... I do know that the
concept of LONG.NAMES ON or the equivalent Unidata directive, may be
related if you are having a problem with name truncation. I believe
that our global catalog keeps generating subdirectories until the name
can be accomodated in
It appears that not much very exciting happens after feeding SYSTEM 1
thru 9 or 10 thousand.
On 2/8/06, Ray Wurlod <[EMAIL PROTECTED]> wrote:
> The two secrets of success:
> 1. Don't tell everything you know.
> ---
> u2-users mailing list
> u2-users@listserver.u2ug.org
> To unsubscribe please
Can you use the USING /tmp with concurrent? I have used concurrent
but it ran for 6 or 7 hours.
On 2/8/06, Brian Whitehorn <[EMAIL PROTECTED]> wrote:
> Hi,
>
> just wanted to check with the group if anyone has resized large file/s
> using the concurrent option. Good / Bad / any "gotcha's" to loo
Me too, 'cause TEST.2 isnt empty I dont think. Unless blanks are not 'something'
I always use EQS or NES functions for evaluation. Just make an array
using the same delimiters, and say
TEST.3 = @VM:@AM:' ':@VM:'zzz':@VM:@AM:'abc':@AM
TEST.1 = @VM:@AM:@VM:@VM:@AM:@AM
TEST.2 = @VM:@AM:' ':@VM:' ':@
test only... please ignore... having sending problems AGAIN.
On 1/31/06, Martin Phillips <[EMAIL PROTECTED]> wrote:
> > Based on a recent thread about UniVerse 'internals' class availability - I
> > researched. The class is now named 'UniVerse Theory and Practice'.
> > It has recently been added t
Universe 10.1.6 running PICK.FORMAT.
We still get this type of entry in the errlog:
Jan 30 10:22:55 SPOOL.NUM: 17 Unable to create RedBack
capture file.
The EXECUTE CAPTURING CAP.FIELD SETTING SETFIELD
does not seem t
Universe 10.1.6 running PICK.FORMAT.
We still get
Jan 30 10:22:55 SPOOL.NUM: 17 Unable to create RedBack
capture file.
entries in the errlog.
We learned that the EXECUTE BLAH CAPTURING BLAHH SETTING SETFIELD
does not
If you run it as a phantom, be sure to do a periodic 'CHAIN' of itself
in order to clean up the leftover orphans.
j
On 1/20/06, Don Kibbey <[EMAIL PROTECTED]> wrote:
> Don't overevaluate "free" code! don't like it, fix it. (then send
> the results to those of us too lazy to do either...)
>
>
>
>
i did do a test, and it appears that i am mistaken. I can create all
types of errors but not that particular one. Would it be possible to
put out more data on the error without sacrificing confidentiality?
john
On 12/27/05, john reid <[EMAIL PROTECTED]> wrote:
> This is only a guess,
This is only a guess, as I have no first hand experience with that
error. Is it possible that the item that was locked was re-read after
being updated in storage and with no COMMIT? It ought to be fairly
easy to test that as a possible means of replicating the problem.
john
On 12/23/05, Serguei
> I have also had personal conversations about it with several
> Vmark/Ardent/Informix/IBM people who were in a position to care or take
> action. I remember asking about it in a question/answer panel during the
> Ft. Lauderdale, 1998 national conference. So it has been a conscious
>
Universe 10.1, sunOS 5.8
I notice that an ls -lt in the u1 /uv /catdir directory indicates that
the *PROGRAM.NAME is updated apparently each time an execution
happens, at least that is what it looks like to me. Anyone know if or
why that is happening?
--
john
---
u2-users mailing list
u2-use
I recall occasions where I had to issue the ENABLE.INDEX command in
order to kick start the automatic indexing process. Sometimes I had
to issue it more than once. I know not what caused the problem or
what I was doing wrong.
j
On 12/13/05, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> Hi All,
To spot the bad guys, build an idesc named GARBAGE, then
SELECT FILE WITH GARBAGE NE ""
You might need a subvalue count as well.
GARBAGE:
I
COUNT(@RECORD,@FM);COUNT(@RECORD,@VM);COUNT(OCONV(@RECORD,"MCP"),".");COUNT(@RECORD,".");IF
@[EMAIL PROTECTED]@4 = @3 THEN "" ELSE @1:"-":@2:"-":@3:"-":@4
Did this a long time ago. It wont cost much to see if it works.
j
** SUBROUTINE MOD10CD(NUM1, CDNUM)
M2STR= ""
SUM.STRING=""
SUM.OPERANDS=""
NEXT.HIGH.10=0
INPUT NUM1
LNUM1=LEN(NUM1)
FOR I = LNUM1 TO 1 STEP -2
IF MOD(I,2) = 0 THEN
SUM.STRING<-1>=NUM1[I,1]
END ELSE
M2STR:=NUM1[I,1]
END
That'd be
.L RELLEVEL
read "DOT ELLE SPACE RELLEVEL"
On 11/7/05, Jacques G. <[EMAIL PROTECTED]> wrote:
> Is there a way to determine which Universe version is
> running (besides the RELLEVEL entry in the VOC ?)
> Perhaps a @ variable or with an undocumented SYSTEM
> command ?
>
> The length of th
Does NFA allow ANY trigger to run across machines? I just tested with
Universe and I get a write failure with trigger code installed where
file lives, and I get an SQL error if I attempt to install the trigger
locally. Kind've the worst of both worlds.
john
On 11/1/05, Wendy Smoak <[EMAIL PROTEC
I think the <<2>> needs to say just 2. The <<2>> becomes the prompt
message. ...
On 11/1/05, Barry Brevik <[EMAIL PROTECTED]> wrote:
> This is UV v9.6.1.3 on Windows.
>
> I have a paragraph which in part looks like this:
>
> RUN BPCUST B$SOD16AE
> DATA <<2>>
>
> Inside the program B$SOD16AE, I ha
Havent I read about a WRITEF, that forces flushing?
On 11/1/05, David Jordan <[EMAIL PROTECTED]> wrote:
> If you are talking about using the sleep statement, I assume you are using
> the same program to write and read. I believe that the write process does
> not return to the program until it has
This is interesting because an inactive file is only so until its
updated. One could probably use BEGIN and END TRANSACTION to obtain a
consistent view. Two programs, in two sessions, one doing an update,
the other doing a simple select and save list yields this:
RUN JMRBP TEST.SELECT.SAVE
0 reco
This is only a stab...
if you make field 7 the name of a phrase, say ACT.ACT, for the dict
item, with the phrase being in the same dict as
PH
ACTION ACT.DTE
then do the same query?
On 10/11/05, Jeff Powell <[EMAIL PROTECTED]> wrote:
> I am querying a file with multivalued fields but I am not getti
Try not to use dim arrays to hold file variables. difficult to diag.
Keep label and variable names to 14 chars and under for universe's xrefer
Avoid ever 'falling' into a label
Never branch around stuff
Sometimes its nice to comment CASE statements
CASE LEN(ID) LE 5 ;* its a phone number
..
On 9
emory overall size never exceeds 13
megabytes, and indeed decreases in size in some instances, and the
performance never varies in the indexed search speed.
I'd really appreciate your comments or questions.
Thanks,
john reid
This e-mail, including attachments, may include confid
I don't suppose that the type of problem being solved could be delved into a
bit more? There are a ton of string routines that I have found useful, but
they seem to be solutions waiting for a problem to solve, in this case.
j
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROT
Is @NOTHING sacred?
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of u2ug
Sent: Friday, September 10, 2004 3:34 PM
To: [EMAIL PROTECTED]
Subject: [U2] major (?) @var security hole
this may be common knowledge , but I stumbled across this yesterday at a
cli
I'm truly dismayed that this triggering business is giving me sorta fast
results, when others have seen a significant slowdown. We constantly strive
for productivity slowdowns, and what do we get but something that seems to
work OK.
I will be opening a ticket with blue to get to the bottom of th
1 - 100 of 109 matches
Mail list logo
