Re: [U2] OPENSEQ and Abnormal termination of UV
In message 321088.49154...@web36901.mail.mud.yahoo.com, Jacques G. jacque...@yahoo.com writes If your file is opened, then comparing it to an empty string is what will cause an invalid data type error. WHY!!! If the datatypes don't match, then the result of the comparison is FALSE, not INVALID. It is logically correct to do such a comparison. The result I am looking for is the logically correct response. It SHOULD NOT crash the program! And it worked fine on INFORMATION, why not UV? It is a bad idea to allow it, it encourages bad programming practices and it can hide a bug in your program for example if you do: READ CAR FROM CARS ELSE CAR = '' IF CARS # '' THEN Here what was intended was to compare CAR to '' not CARS but having the S there is a typo. If we don't allow datatypes of different types being compared this bug has a greater chance of getting caught during development because the program will fail on comparing the file to a null string. If you use FILEINFO it shows that you know that the variable you are testing is a file. The big problem with your attitude is that you are assuming a typed language - and even then it's invalid. In BASIC, *EVERYTHING* is a string (apart from file variables). Therefore any comparison should be valid. And, using your own words against you - it shows that you know that the variable you are testing is a file. What if I DON'T know (which is why I was doing the test!). At the end of the day, an IF test for equality should return FALSE on a type mismatch, not CRASH. Because FALSE is the *correct* answer. That's what I absolutely HATE about artificial stupidity. It thinks it knows better than me, and gives me what it thinks I should have, not what I asked for. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Code 128 Soft Font
In message 4bf49b9a.5070...@comcast.net, Charlie Noah cwn...@comcast.net writes [Excuse me for a minute, my mouse cursor just took off by itself across the screen while I was typing, and I had to see how far it would go. It crawled slowly to the left of the screen and just sat there. Spooky!!!] I think this is a well-known software quirk. Doesn't often happen but nothing to worry about. Indeed, I am typing this email on a piece of software that exhibits exactly this sort of quirk - when I mouse-click on a scroll-bar the bar moves, but so does the mouse! Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] OPENSEQ and Abnormal termination of UV
In message 536051.22442...@web36905.mail.mud.yahoo.com, Jacques G. jacque...@yahoo.com writes Oh - that reminds me of something else I'd call a bug. It might well have been fixed by now (I met it in 9.6) but you couldn't safely use a file variable in an IF statement. Can't remember the details, but it was something like FVAR = some conditional code OPEN FILE TO FVAR more code IF FVAR EQ THEN oh the file isn't open, so do whatever If your file is opened, then comparing it to an empty string is what will cause an invalid data type error. WHY!!! If the datatypes don't match, then the result of the comparison is FALSE, not INVALID. It is logically correct to do such a comparison. The result I am looking for is the logically correct response. It SHOULD NOT crash the program! And it worked fine on INFORMATION, why not UV? Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Print Wizard on Linux (was Re: Code 128 Soft Font)
In message 0baf3814a3ac4766bc06290e17ab8...@glen, Glen Batchelor webmas...@all-spec.com writes Let's break a few things down and compare: PCL input and output is covered under Ghostscript, though some complex PCL6 stuff may be buggy. Find a bug and report it at this point. It will get fixed. Generic PCL3/4 stuff should be solid at this point, but I would verify with the devs. The XPS/GPDL code has been out for a while now. Note that PCL5 is a superset of HPGL2, so you've got vector graphics available there ... BT,DT,GTS - written a DataBasic program to dump HPGL to a laserjet ... Windows printing drivers? While some obscure printers are problematic in CUPS configuration, I've never had an issue making programmatic paper/drawer/slot/etc selections when a PPD was available and installed properly. I'm willing to bet the *nix admin(s) running the equipment selected printers better suited to CUPS than Windows, so I wouldn't worry too much about that. Standard advice, if there are problems locating a printer driver, is to look for the Windows PPD, as they are supposed to be identical. It's just that the Windows setup programs hide as much as they can (including the PPD) from the user. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Code 128 Soft Font
In message pine.lnx.4.64ras.1005190850490.5...@nimbus.anzio.com, Bob Rasmussen r...@anzio.com writes On Wed, 19 May 2010, charles_shaf...@ntn-bower.com wrote: You refer to our Print Wizard product. No, we don't have a Linux version. Any plans to develop a Unix version? All of our servers run on Unix. Some of our workstations too. The same issue exists on Unix, so no. It is quite easy to set up one or more Windows machines as print servers, running Print Wizard, and feed them from Unix or Linux. This is what we recommend, and we have many customers doing this. The only problem with *that* - is that your $100 solution has just become a $1000 solution. Find a machine, find somewhere to put it, pay for Windows (and you probably need the expensive pro version), and then pay all the running costs (electricity, air-con, etc etc aren't cheap). A lot harder to justify for occasional use. Don't forget - nix houses aren't used to the one server per service concept - they've probably got one old box running twenty, thirty occasionally used services - $100 to add another service on it is a very different proposition to finding a complete new machine just to run one thing. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] OPENSEQ and Abnormal termination of UV
In message e6179e13392ec14aabcd5272c3aedd61124ed...@exchangesvr.momtex.com, John Hester jhes...@momtex.com writes Yes, I see your point. I wonder if the integer gets treated like a string in the first instance. I wonder what the result with FILEVARS1 would be. Illegal assignment. You can't store a file descriptor in a dynamic array, because it's not a character string (it MIGHT be implemented as such, but it doesn't make sense to treat it as such, so it's a logical stupidity. I don't know of any MV implementation that permits this). Cheers, Wol -John -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Anthony W. Youngman Sent: Tuesday, May 18, 2010 7:45 AM To: u2-users@listserver.u2ug.org Subject: Re: [U2] OPENSEQ and Abnormal termination of UV In message e6179e13392ec14aabcd5272c3aedd61124ec...@exchangesvr.momtex.com, John Hester jhes...@momtex.com writes I think it's something along those lines, but I don't think it's trying to stick the entire contents of the file into a variable. What I think OPENSEQ is doing is keeping track of the position where the EOF mark is so it will know when the end of the file is reached. For a file greater than 2GB in size, this position is an integer that takes more than 32 bits to store. UV, being a 32-bit application, is not going to be able to handle it. The maximum positive integer value a 32-bit application can reference is 2147483647. The problem is, FV.FILE and FILEVARS(1) are *allegedly* *functionally* *identical*. An element of a dimensioned array, is supposed to be a normal variable in every way shape or form. The problem is that, in this instance, it clearly isn't because the variable works while the element (allegedly identical) causes a crash. I'd agree with Perry. It's a bug. ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] fnuxi problem
In message 9ea3c4a038d14a2fa37a2ea3c801a...@lbs8, Martin Phillips martinphill...@ladybridge.com writes Hi all, A follow up to my own query. This site uses entirely dynamic files. The recommended way to do fnuxi recursively over a whole application is to use find as in my previous email. This will end up doing the fnuxi against the DATA.30 and OVER.30 files. The OVER.30 file reports that it is not a UV file and is skipped. If I do the fnuxi against the directory that represents the dynamic file, it appears to work. If I do it against the DATA.30 file it sometimes hangs. So, it now sounds as though I need to construct a find that will produce all the files that don't end .30 (so I get the dictionaries and indices) and a second pass to do all directories. Once again, anyone had experience of this? not being on my linux system I can't do a quick man, but has find got an option to say how deep to search? Off the top of my head I think it has. If you can say only search the current directory, and run it in the account directory, it sounds like that'll do exactly what you want. That's assuming you're only converting one account. If you're converting multiple accounts, Brian's search for VOCs and process each in turn in a script sounds like the route you want. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] -1 and Null values
In message ddc8e880ea0e4f438407b6fa651ca12f07eae...@es02.imb.net, Dan McGrath dmc...@imb.com.au writes Hi all, Hopefully I'm not going crazy, but I could have sworn in UD 7.1 that doing a -1 to append an empty string would not actually appended anything. For example: As I understood it, the problem was always appending TO an empty string, not appending an empty string (though I was always used to INFORMATION :-) Because an empty string could have no values, or 1 empty value, PI always assumed it had no values. So ... VAR = VAR-1 = VAR-1 = A would leave VAR equal to A But a further VAR-1 = would then append a null to give A:@FM All very logical (as I would expect from Prime :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Universe and stripped drives
In message 6d57ef06d84b5541af530227504bb68f5c4f4ef...@34093-mbx-c06.mex07a.mlsrvr.c om, George Gallen ggal...@wyanokegroup.com writes This I will have to look into. Right now I'm deciding if I want to setup the partitions in an LVM or not. To follow up a little further, look at log-based file systems. TuxFS for one is an example I believe (snag is, I don't know of any production grade FS's). Basically, the way they work is copy on write. Let's say, I update the file /home/wol/directory/file. It will write my changes to a NEW block, and update the directory inode info in ram. Then it will write the changes to directory to a NEW block and update wol similarly. Likewise home, and finally /. So, should the system crash at any point, the ONLY danger to the filesystem is if it falls over while updating / itself. But the thing about a file-system like this, if you want to take a backup you just lock / and backup a single instance of it. No matter what changes are going on around you, you have a disk instance that is frozen in time for safe backup. Finished? Unlock your / and let the FS throw it away. I gather some filesystems will lock / long term so you can get to stuff you've long deleted/corrupted, only clearing and freeing the instances as disk space is required. And it's pretty efficient, only saving changes to files, not a copy of the entire file every change. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] AccuTerm File Transfer
In message 491936aa72824093a41621174a396...@glen, Glen Batchelor webmas...@all-spec.com writes Why not? I do that with PDF files just fine. Just make sure the MIME content is correct in the response header, for the file type you're serving. And if the client and server are on the same network, why not have a specific type 1 file that is shared, so the client can reference it with an url and accuterm would just fire up Word directly (don't know accuterm, but if Wintegrate can do it surely accuterm can too). If it's a Windows server no probs, *nix server just run Samba. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Universe and stripped drives
In message 6d57ef06d84b5541af530227504bb68f5c4f4ef...@34093-mbx-c06.mex07a.mlsrvr.c om, George Gallen ggal...@wyanokegroup.com writes I planned on doing this as well. There will be 3 system drives which I'm going to put some of our busy temp files. How much ram have you got? How big are your temp files? Have you got tmpfs? Okay, for me it's just a home pc, but with 4Gb ram it makes sense to have a large swap and mount /tmp to a tmpfs partition, with the result that all my temp files are stored in ram. If it gets too overloaded they'd get dumped to swap, but on my machine that doesn't happen :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] UV JDBC SQL syntax
In message 9c300472e764f645b1bab4571ac8d1260147cd3ee...@bc-comm.fusionware.net, Robert Houben robert.hou...@fwic.net writes Depending on what tool you are using, the dots will get you into big trouble. For most compliant SQL engines the dot has special meaning, separating qualifiers. iirc, the UV SQL engine can (MOST) of the time, cope with this. But the heuristics are fragile. Best avoided and, if it can cope, I've broken it on various occasions (I haven't used SQL on UV much, I just sucked it and saw, and don't remember much of it now. It wasn't difficult). Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Quick poll - how many use 3-tier or N-tier Architecture
In message 4bc5e280.5070...@advantos.net, Bill Haskett wphask...@advantos.net writes Wol: What do you consider properly normalized and what example would you give for designing a new mv FILE as a set of nested tables? Basically, use relational theory. Okay, mv design isn't as clear as for a RDBMS (they force everything into two dimensions - imho a very stupid idea), but the underlying maths provides a solid foundation. The way I'd do it is to do an EAR (Entity/Attribute/Relationship). Each entity I would aim to put into an individual FILE. All the associated attributes and relationships, I'd analyse using relational theory, then recombine them in NFNF using values and sub-values as required. This is where the fact that MV is not fully multidimensional can be an advantage, or can be a pain if you need more levels than provided by sub-values. But my rule-of-thumb is that if your FILE primary keys are equivalent to real-world keys, then your design is about right. If you need what looks like a SQL complex key, chances are you're falling into Einstein's too simple trap. That said, I have had some exposure to accounts systems, and doing an EAR on them is likely to be interesting, let's say ... most of my experience has been in other stuff like bill-of-materials, and statistical analysis of prices. Thanks, Bill Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Quick poll - how many use 3-tier or N-tier Architecture
In message 012d01cad820$f3492290$d9db67...@com, Symeon Breen syme...@gmail.com writes Absolutely Brian, i agree wholeheartedly - However try having that conversation with a computer science guro... but then again they don't live in the real world. As soon as they mention the word proof (quite important in software and relational database design, I believe :-) that places them *firmly* in the realms of maths. And, to quote Einstein, As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality. :-) Indeed, as I've often said, it's pretty easy to prove that relational databases, in order to qualify for the moniker relational, *must* be inefficient energy hogs. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Quick poll - how many use 3-tier or N-tier Architecture
In message 0f0fc5f04b472746b01fa2c4897cf972014314ee6...@excl01.mouser.lan, Baker Hughes baker.hug...@mouser.com writes For anyone - what Multi-Value aware / friendly middleware products are there? (That don't require data normalization before sending to the middleware.) Why don't you want to normalise? Legacy code? ALL new databases should be properly normalised (but no, I do *not* consider first normal form as properly normalised :-) Retrofitting normalisation on an old database isn't easy, but failing to design a new FILE as a set of nested tables is imho very bad practice. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] VSG Creates Incompatible ODBC View
In message j2j8c1cabce1004051146xf7980b4dzcfbaf08ab99de...@mail.gmail.com, Kevin King precisonl...@gmail.com writes I have a customer who has a view that appears like this in SQL: SELECT ID,TYPE,BEGIN_DATE,END_DATE,FISCAL_PERIOD,PERIOD_OPEN FROM T_CRP_FISCAL_PERIODS UNNEST NL_ALL BEGIN_DATE,END_DATE,FISCAL_PERIOD,PERIOD_OPEN WHERE ID = 001FISCAL.PERIODS; This view, as far as I have been informed, was created via the VSG. And this is valid SQL when run from the sql command line in Unidata. However, when this view is used over ODBC, it returns 001FISCAL.PERIODS is not a valid field name. For this particular situation, changing the double quotes to single quotes seems to do the trick, but that leads me to two questions for this esteemed group: a) Is VSG creating incompatible SQL, or does this appear to have been created by a human? b) Will we encounter other difficulties by this change from double quotes to single quotes in the WHERE clause? This sounds similar to a problem I had. Different, in that it was table names, and I can't remember the details, but - I think it was the MS-Access query generator - would select from a file but would assume that file001 existed! And it worked fine provided stuff was all MS (like reading/writing to excel), but broke when it was trying to read from UV. Not much help, but this might be another clue... Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] FTP Users in HP-UX
In message 4bb4975f.4030...@tpgcrafters.com, Tom tech...@tpgcrafters.com writes I can FTP as root so I am thinking the FTP daemon is running. It just any other user that cannot FTP. That's weird ... or is it you're ftp'ing from the server to itself? When you and they do an ftp, do you all get the ftp login prompt? If the behaviour is different for them and you, then it's presumably a firewall thing - their ftp request isn't even getting as far as the server. Once you've got as far as the ftp login, what happens there? Can they log in as root but not as themselves? You've now got a permissions problem. Oh - and your comment about the box being rebooted says immediately that this is a soft problem. Have you got any records as to what you did when you configured ftp? Because what seems obviously to have happened is that you made changes to the running configuration, but you didn't then push them back to the config files. So when the system all the changes got lost. You need to look at what permission changes you made, especially ones that affected the running ftp daemon. Because obviously you needed to make the changes, but that's as far as they got... If this doesn't help, come back with a LOT more detail about how it works for root, but not for ordinary users, and somebody should have the experience to chip in and identify the problem. If you can, posting a successful and an unsuccessful session trace would be almost perfect... Tom Cheers, Wol -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Tom Sent: 01 April 2010 13:13 To: u2-users@listserver.u2ug.org Subject: [U2] FTP Users in HP-UX We have something strange happening with FTP on our UNIX box. We have some users setup to FTP files from our windows server to the UNIX box. They were working fine until we rebooted the UNIX server. Now none of them have FTP access into the UNIX box. This was the first time the box had been rebooted since setting them up. These logins can telnet in fine but cannot ftp. I have reviewed all the settings and believe they are all correct. I am not a UNIX guru but know enough to be dangerous;) Wish there was a setting up FTP for Dummies instruction manual. Any clues as to what may be part of the puzzle I am missing? Hardware : HP Operating system : HP-UX O.S. version : B.11.00 UniData version : 6.1.16 Tom ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users __ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email __ Confidentiality Notice: This e-mail and any attachments are intended solely for the addressee and may contain confidential or privileged information. If you are not the named addressee, or the responsible for delivering the message to the named addressee, please notify the sender as soon as possible and delete the material from your computer. This message will be protected by copyright. If it has come to you in error, you must not take any action based on its contents nor must you copy or show the message to any person other than the intended recipient. _ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email __ ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Subject: 2 servers - how do I point to a VOC on another server?
In message snt126-ds139c3098dde64b42057b6aa0...@phx.gbl, Doug dave...@hotmail.com writes Hi Chris: Without buying UVNet and or PRC which cost many thousands of dollars. You have a couple of choices: 1) You can use the smoke and mirrors of VOC pointers 2) You can get a real version control system. Most of the version control systems CVS or Subversion are free. They require scripts to code to hook them up. And if you're going to learn a new version control system, GET A DISTRIBUTED ONE! Okay, you might be the sole developer ... but a distributed system still helps. I learnt some sourcesafe in my last job, and one of the biggest pains was another developer working on a module I was using, or needing to bugfix a module that was being developed on. You can add your own niggles. I've started learning git, and while I haven't really scratched the surface of what it can do, it's easy to have multiple developments going on against the same stream of code. (You just curse a bit when you need to merge those streams, although git does try to make it easier for you :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] CPU Spikes to 100%
In message 10998-1269884106-801...@sneakemail.com, Tony Gravagno 3xk547...@sneakemail.com writes From: Curt Stewart The environment is a virtualized Windows 2003 Server and Universe 10.2.10 On occasion, I've been notified twice, it may have occurred more often, a Universe process (tl_server.exe) will consume 100% of the CPU. Curt, I wasn't watching this tread earlier, sorry. This is the Exact same problem that Mike Roosa at Tolt is having, as documented in another recent thread here. Rocket has been unable to thoroughly diagnose the issue but tl_server does seem to be either the culprit or a telltale victim. I highly recommend you contact Rocket and tell them this may be related to the Tolt situation. Note that their environment is also virtualized in VMware. Perhaps there is something in VMware, like flushing, or some abnormal condition, which causes tl_server to start spinning. Ahhh! This sounds like a long-standing problem that I thought had been fixed some while back. We used to suffer this with UV9.6 on NT but I don't remember it with UV10/Win2K. We didn't have a virtualised environment at the time, so if it's the problem we had, it's nothing to do with VMware. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] nt to unix
In message 6d57ef06d84b5541af530227504bb68f5b69aa6...@34093-mbx-c06.mex07a.mlsrvr.c om, George Gallen ggal...@wyanokegroup.com writes Both are intel based. NT-Fedora I wanted to copy an account, instead of exporting the data as bunch of flat files, then importing and rebuilding the files. I was hoping to just be able to move the whole directory structure from the NT to Linux, then add that account into the list of usable accounts under the Linux UV. I'm more interested in viewing the data, (they are type 30 files, so I don't think that dos2unix utility would be a healthy thing to do - I would think they should have the native @fm,vm,svm's vs lf's). I'm not concerned about any programs, just data. Or is there a better way to move an account that would work. Both boxes are based on 32 bit O/S's Ages ago I did that no problem (except nix to NT). Running UV9.6 on both machines - NT4 and SCO Unix - I ran Samba on the SCO and copied accounts in both directions without any problems. The other thing is, I think Unix line endings work fine on Windows for program files BUT if you go the other way the carriage return in the line ending mucks things up. But it's visible in ED so you can just loop through all the source programs removing them. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Unidata Silly Gripe
In message 031801cacc3b$feb964a0$fc2c2d...@com, Symeon Breen syme...@gmail.com writes Just a different way of doing it i suppose - i do have separate sub directories but they are under the dev account and not under say BP Just a little point - bearing in mind I've never used genuine Pick but I think Larry et al are describing *TYPICAL* Pick usage. You couldn't do that on Pr1me so anybody (like me) only used to the Pr1me approach this would seem strange. We had three (actually four) main program directories on our system, called CBP, GBP and RBP. But on a Pick system they would typically have been defined as subfiles of BP, eg BP,COL BP,GEN and BP,REM. Cheers, Wol -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Larry Hiscock Sent: 25 March 2010 16:23 To: 'U2 Users List' Subject: Re: [U2] Unidata Silly Gripe Really? We have more than 20 subdirectories in our program directory. Each of them is defined as a DIR in Unidata, but at the Unix level (and convcode is a system-level command, not a Unidata verb) each is simply a sub-directory of the source directory. Larry Hiscock Western Computer Services -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, March 25, 2010 5:35 AM To: 'U2 Users List' Subject: Re: [U2] Unidata Silly Gripe Probably because you would never have subdirectories in your program directory. snip ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Not Consistent Failure
In message snt141-w2040511e4ab89e0e961de1c2...@phx.gbl, Dan Fitzgerald dangf...@hotmail.com writes One thing you could do is to make that sleep like 30 seconds, find that pid, then use the AIX truss command against it while it's sleeping, and see what bombs. When they drop to AIX, are they in the directory specified in /etc/passwd, or somewhere else? And how are you invoking UV? From the local .profile? And are you just running UV, then logging out from inside the .profile? What we did when running on nix (Pr1me RISCoS) was to 'exec' UV (or PI) from the .profile. That way, UV overwrites .profile in memory - a sort of chain - and when it quits the login session quits. At least that will (or should) stop your users dropping to nix - a login failure should simply dump them straight out of the system again. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Unidata Silly Gripe
In message 00c201cacc68$af62e700$0e28b5...@com, Larry Hiscock lar...@wcs-corp.com writes Actually, you COULD do this on the Pr1me. Our application was migrated to Unidata from Pr1me Information. We always kept program sub-directories segregated by application (e.g. AR, AP, GL, etc). But I'm guessing you had *separate* VOC entries for AR, AP etc. At the OS level they were under one sub-directory but inside of PI they were separate FILEs. What I'm talking about - what I think Pick has - is where you have - at the *PICK* level, one BP FILE, and then loads of subfiles in it. Which is stored (by default) as one directory with sub-directories at the OS level. In other words, in PI you're talking about the OS level - nothing to with PI. In Pick you're actually IN PICK, it's nothing to do with the OS. --Larry Cheers, Wol -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Anthony W. Youngman Sent: Thursday, March 25, 2010 3:08 PM To: u2-users@listserver.u2ug.org Subject: Re: [U2] Unidata Silly Gripe In message 031801cacc3b$feb964a0$fc2c2d...@com, Symeon Breen syme...@gmail.com writes Just a different way of doing it i suppose - i do have separate sub directories but they are under the dev account and not under say BP Just a little point - bearing in mind I've never used genuine Pick but I think Larry et al are describing *TYPICAL* Pick usage. You couldn't do that on Pr1me so anybody (like me) only used to the Pr1me approach this would seem strange. We had three (actually four) main program directories on our system, called CBP, GBP and RBP. But on a Pick system they would typically have been defined as subfiles of BP, eg BP,COL BP,GEN and BP,REM. Cheers, Wol -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Larry Hiscock Sent: 25 March 2010 16:23 To: 'U2 Users List' Subject: Re: [U2] Unidata Silly Gripe Really? We have more than 20 subdirectories in our program directory. Each of them is defined as a DIR in Unidata, but at the Unix level (and convcode is a system-level command, not a Unidata verb) each is simply a sub-directory of the source directory. Larry Hiscock Western Computer Services -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, March 25, 2010 5:35 AM To: 'U2 Users List' Subject: Re: [U2] Unidata Silly Gripe Probably because you would never have subdirectories in your program directory. snip ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Pick Pocket Guide
In message col107-w66ce4bea66ba98420166ce2...@phx.gbl, Jo Lester jp.les...@hotmail.com writes I haven't bought an old pick book by Jonathan Sisk for weeks. If Jon printed, or released existing books, however old to Kindle, I'd buy. This goes for you too, Tony. You write well. I don't know if Jonathan still owns his own copyrights. However, I would have thought, if he does, publishing them via a print on demand publisher like Baen might be a good idea. Dunno if O'Reilly would do something like that. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Terminal Emulator and Telnet Client for a MacBook Pro
In message 005901cac75f$0852b8e0$18f82a...@com, Kevin Gusler kgus...@mortgagebuilder.com writes Has anyone tried running SBClient 5.3.4 - 5.4.4 in Parallels on a Mac? Are there any issues? We are running (windows) SB+ 5.3.4 and Universe 10.3.3 and have a few users that wish to use Macs for their client pc's. Just wondered if there was a downside. How new are the Macs? Does parallels only run on the new Intel Macs? (Sorry, I don't use a Mac, don't know when the switchover was. But there are probably still a fair few Power Macs still in general circulation.) Thanks in advance. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] How do you determine terminal emulator?
In message 19552-1268350334-97...@sneakemail.com, Tony Gravagno 3xk547...@sneakemail.com writes It does. ViaDuct was my favorite emulator in the early 90's until I found wIntegrate, and that was my favorite until the later 90's when I found AccuTerm. :) All support scripting. The problem is that they're all different. So if you send ESC:FOO to any one of the emulators, it might do something nice or it might lock it up. That's the chance you take and fine-tuning the scripts to play nice if they hit the wrong emulator is what I call the hard way. You may get it right with some experimenting but it could be time consuming to get it just right. So you configure all the emulators to respond to the same sequence. The problem is if you've got different real terminals that you can't program. But it looks like some variant of ctrlE might well be a common factor. Below is the code I used to tell whether the user was using a pt250, a wyse85, or wintegrate/pt250. As you can tell, it was written long, long ago. * Select Terminal Type Here * * * Use the answerback mechanism so the terminal is selected correctly. PRINT Requesting answerback ; * This message is necessary to clear the buffer CLEAR INPUT PRINT CHAR(27):CHAR(5) SLEEP 1 TTY = HUSH ON LOOP INPUT KEYSTAT,-1 WHILE KEYSTAT INPUT JUNK,1: ; TTY := JUNK REPEAT HUSH OFF * Now try again - the above code doesn't work with the new winterm. awy 21/09/01 IF TTY ELSE CLEAR INPUT PRINT XX:CHAR(5) SLEEP 1 TTY = HUSH ON LOOP INPUT KEYSTAT,-1 WHILE KEYSTAT INPUT JUNK,1: ; TTY := JUNK REPEAT HUSH OFF END SKIP = 0 IF TTY ELSE LOOP CRT ; CRT @SYS.BELL : Which terminal type 1 (VT100) 2 (PT200/wIntegrate) 3 (ignore) ? : : INPUT TTY : BEGIN CASE CASE TTY = 1 ; TTY = VT100 CASE TTY = 2 ; TTY = PT200 CASE TTY = 3 ; SKIP = 1 CASE 1 ; CONTINUE END CASE EXIT REPEAT END EXECUTE SET.TERM.TYPE : TTY : WIDTH 80 LENGTH 24 HUSH If I remember correctly, the wyse/vt answerback sequence was ctrlE, while for the pt250 it was escctrlE. The winterm thingy was a wyse thin terminal running WinCE or something like that. We played with them but ended up not using them. The requesting answerback tipped the user off that something was happening, and the 1-second sleep gave the terminal/emulator time to respond and fill the buffer. The program then checked what response it got and carried on ... as you can see, we'd programmed it to respond with the terminal type. So occasionally, when we got a new real terminal, things fell over until we remembered to program the terminal response :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Scrub or Combine Distributed Part Files?
In message 4e79ea8c2ba6694eb6a87686c3a9a76908a8a...@afs_exchange.afs.afsi.com, Mark Eastwood ma...@afsi.com writes I suspect the Distributed file is not the problem, but rather bad dictionaries. Do all the dictionaries have a data-type defined (AS types attr 6, DI types attr 8)? A common issue I've encountered is when a dictionary is right justified but attribute contains non-numeric data. Following up from this, go to PickWiki, to the Universe page, and then look at the page there that talks about ODBC. It will point you at a bunch of utilities, written specifically to go through warning/cleaning your data. I learnt the hard way that HS.SCRUB had a habit of crashing and it was difficult to debug why - programs should NEVER crash when fed duff data and HS.SCRUB is (or was) an egregious example of a program that does exactly that :-( Oh - yes I don't think you can run HS.SCRUB on the master distributed file, you have to run it on the parts. But read the docu. And it could easily NOT be a duff dictionary, but a duff VOC :-( Read that wiki page! Lastly, are you using ODBC? If not, and you're planning to, don't bother. ODBC is now a layer on top of something else (OleDB?), and you should use that instead. It'll be quicker, easier, nicer, generally better all round. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] How do you determine terminal emulator?
In message 8363-1268176878-166...@sneakemail.com, Tony Gravagno 3xk547...@sneakemail.com writes You can try to do this the hard way and face lack of success, or you can do this the easy way, which is perhaps not as elegant but it's a fast and effectice solution. The hard way: try to come up with scripts that will return useful information from specific emulators without messing up any other emulators. You can probably create a single script for any two emulators, but that's pushing it, and three may be too much to ask. IF you've got decent emulators (like Wintegrate) this way is easy. It's when you've got real terminals it's a problem. As I said, luckily I had a sequence that would provoke a response from both *real* wyse85s and *real* pt2x0s. I then just added the correct answerback responses to Wintegrate's pt250 emulation so that worked, too. Wintegrate has a definition that says when the host sends sequence X, eat it and reply with sequence Y. The user *can* muck things up, but we never had any trouble. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] How do you determine terminal emulator?
In message 6.0.1.1.1.20100309151717.05cc2...@12.167.143.145, Beverly Wilson bwil...@daedalus-books.com writes Hi folks, I'm wondering if anyone on the list can tell me what I can do to determine whether a user is using the Viaduct terminal emulator via UniBasic? Can you configure Viaduct? It should have some sort of answerback function. We have some code that can tell if the emulator is Accuterm or Wintegrate, but I haven't been able to come up with a command string that would cause Viaduct to return a value. I've looked through the docs, but haven't found anything. (I may be suffering situational blindness.) To give you an idea, these lines when run in through Wintegrate puts an X into the variable C: PRINT CHAR(27):CHAR(1):ENTER :DQUOTE(X):CHAR(13): INPUT C WAITING 1 ELSE C = Been there done that, but what you're doing looks slightly odd to me. We had a mix of Prime PT200s, PT250s, Wyse85s, and Wintegrate configured as a PT250. All of the real terminals would respond to a sequence something like esc? (can't remember what it actually was). So I wrote the login program to send this sequence to the user, and the terminal would (as it was meant to) respond with some string without any user intervention. And I programmed Wintegrate to behave just like a pt250 including this answerback. If you look at the pt250.wis (or is it .wit?) file distributed with Wintegrate you might well find my code in there to do exactly that. You'll need to read the Viaduct docu, but if it's at all user-configurable, you should be able to program it to do an answerback. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] UniVerse on Windows 2003 Server - Environment
In message 4b8d4c38.3010...@advantos.net, Bill Haskett wphask...@advantos.net writes When you experience the slowdown, check the Task Manager. It is likely you'll see something bad there. If not, you should get the sysinternals code procmon and diskmon. They'll help too, if the problem isn't with the CPU usage. Actually, it's pretty interesting what one finds running on servers these days. :-) And see how much cpu telnetd is chewing up :-) Admittedly I'm talking about UV 9.6, so this particular issue is probably long fixed, but we had telnetd's get detached from the terminal and they would just sit there in a cpu-munching loop. Is it possible that - although they now clean up after themselves - they could still temporarily be getting themselves in this state? Until we realised what was going on, we had to reboot the server because they didn't clean up after themselves, but we never noticed the problem once we were on 10. Bill Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] [UV] On Linux
In message 734517.36695...@web36904.mail.mud.yahoo.com, Jacques G. jacque...@yahoo.com writes Hello, I've it mentioned a few times that Universe is available on Linux. On what flavor of Linux is it available ? Has anyone here been using it and are there any issues versus how it works on UNIX ? Look on the compatibility matrix. I'm pretty certain that RedHat/Intel is supported. I *think* SuSE/Intel is *probably* supported. I certainly have the impression that SuSE is supported but it might be SuSE/390 for mainframes. I've also seen reports of it running on other linuxen, but installing it can be a pig. The RedHat options on the install don't necessarily work on other versions, and you might find yourself having to edit the install script (in particular the cpio options), or messing about with the runlevels to get the daemon to start by default. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Resize Question
In message 000501caa516$38b303c0$aa190b...@com, Andrew E. Tegenkamp and...@g3.com writes Thanks again for taking the time to help me out. I have been researching this all day as I did not want to bother the list anymore. I am still looking up and reading commands to try to find the right commands to issue to complete these tasks and I appreciate the time it takes to help me find the right road to take. There's an article on PickWiki about ODBC (written by yours truly :-) Go into the Universe page and it should be visible from there iirc. It was written for 9.6 so things have changed a bit but it might have some useful information. In particular, it contains links to some programs that will go through your files looking for known gotchas with the data. Oh - and ODBC now sits on top of OleDB, so you're better off using that if you can. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] The opposite of Export is.. sometimes not Import...
In message f07cf5651001211306h41458f3bkf3a2452d48f1d...@mail.gmail.com, Richard Lewis rbl...@gmail.com writes I would probably start by making an index on the Part Number, based on an I-descriptor with an expression like this: CONVERT(OCONV(OCONV(PartNumber,'MC/N'),'MC/A'),'',PartNumber) Slightly longer, but I'd do it as follows CONVERT( ABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890, , UPCASE(PartNumber)); CONVERT( @1, , UPCASE(PartNumber)) where PartNumber is a D-pointer to the raw data field that is the Part Number. This will strip out all the non-alphanumeric characters. You could then easily access by Part Number no matter how many dashes, dots, etc. were entered on either system, via the index. Gets the same result as Richard (but uppercase), but it's generic - change the char string to whatever characters you want to keep, and everything else will disappear... Oh - and even when you're doing the exact match select, it's probably a good idea to select on the index at the same time - it'll speed things up even further unless PartNumber is your @ID. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Credit Card info
In message 32605-1263604391-637...@sneakemail.com, Tony Gravagno 3xk547...@sneakemail.com writes Final note: I recommend breaking up any secure data you have and storing it in different files. A compromised credit card number is no good without other data including name, address, zipcode, phone number, etc. If you store the card ID in pieces, and encrypted, and separate from this other data, then even if the environment is compromised, the only person who could make use of the data would be someone who is intimate with your code and file structures. That was something I was thinking of. I saw on Risks where somebody discussed this print only the last four digits of the card number. I *think* actually, that's NOT what you should do for credit cards. The reason is strange, but makes sense ... Certainly with Barclaycard/Visa, the *first* four digits are pretty much constant per the issuer. It's the last digits that vary most. So if you only display the *first* four digits, you will give enough info to the card owner for him to identify his card, but any attacker will only be able to identify the bank that issued the card. All Barclaycards, for example, begin with 4929 iirc (or they did, I think there are a couple of other variants around now). Other cards are, I gather, the other way round. That article on Risks was how people who didn't understand WHY a particular 4-digit group had been chosen, arbitrarily changed it and thereby actually undermined the entire security behind the idea. The danger is if different people print different bits of the number. An attacker can then put the whole number together from different printouts. Either way, if you're going to print 4 digits, DON'T pick which four at random or because someone else says this is the four. Ask yourself WHY pick that four, and there's a damn good argument which tells you which set to pick, and it isn't just because they're the first, or the last. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Encryption errors - don't care
In message 0f0fc5f04b472746b01fa2c4897cf972011e10a44...@excl01.mouser.lan, Baker Hughes baker.hug...@mouser.com writes The native encryption in UV 10.2 is very nice. But does anyone know the secret to 'turn off' the encryption errors on our development system? We have the development machine setup identical to the live so it serves as a hot backup. We don't have an encryption key loaded though since all sensitive data has been cleansed from the dev db. Is there a way around this: Unable to get encryption info (19690) in ENCINFO. and others like it. Delete or wipe the error message in the error messages file? Do be careful if you try this - I haven't! By the way, did you post this completely new topic by replying to an unrelated message and wiping the subject and message? Please don't! Please create a new thread by starting a clean new email for a clean new subject. I know some (broken) mail clients group messages by subject, but there's a perfectly good, long-standing and *standard* header code in-reply-to which is meant explicitly for grouping related messages together. My email client uses this which means the above reply tactic mixes your message in with loads of unrelated messages. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] ITYPE BASIC use?
In message f1eed293-d08d-4d31-87c3-de5dd8acc...@lynden.com, Rick Nuckolls r...@lynden.com writes I doubt that I will be the only one to reply, but The variables @id and @record must be set prior to the call, as in: Note that both of these are *optional* *if* they aren't used. But it's extremely unlikely you won't use at least one of them. I don't think I've ever bothered with @id, just so happens that's the way the i-types I've used have panned out. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Count question
In message d85db451e1913e4d9889cfb25543b24701975...@auexci01.ad.internal, Boydell, Stuart stuart.boyd...@spotless.com.au writes Whoops - forgot the reuse function: LIST MY.FILE EVAL SUM(GTS(MY.MV.FIELD,REUSE(0))) In this instance you might need SUM() twice ... Note that when the example was provided it had two values each with four sub-values 0011: 4|0|0|0}4|23|4|0 iirc your eval will return 1}3. So the eval actually needs to be LIST MY.FILE EVAL SUM(SUM(GTS(MY.MV.FIELD,REUSE(0 Cheers, Wol eg LIST GRTRANS EVAL SUM(GTS(REC.QTY,REUSE(0))) FMT 10R REC.QTY G38781801 1 G46781921 2 18675004 1 4 6 1 G40781832 3 3 ... Stuart Boydell -Original Message- Hi all, Using UniVerse 9.6 on AIX4.1, I'm trying to write a routine to count the number of occurrences of values than 0, for example, against this attribute: 0011: 4|0|0|0}4|23|4|0 I'd like to return a value of '4' for the 4 values 0. Not having much luck with COUNT or DCOUNT. Any help is greatly appreciated. Ed Hess ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users ** This email message and any files transmitted with it are confidential and intended solely for the use of addressed recipient(s). If you have received this communication in error, please reply to this e-mail to notify the sender of its incorrect delivery and then delete it and your reply. It is your responsibility to check this email and any attachments for viruses and defects before opening or sending them on. Spotless collects information about you to provide and market our services. For information about use, disclosure and access, see our privacy policy at http://www.spotless.com.au Please consider our environment before printing this email. ** ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Unidata 6.1 Replication
In message e6179e13392ec14aabcd5272c3aedd61112ce...@exchangesvr.momtex.com, John Hester jhes...@momtex.com writes There are also issues with UV dynamic files (don't know about UD) because some of the file structure information is cached in memory and not immediately flushed to disk. This is no longer true. Can't remember which IBM'er it was, but it was confirmed that UV flushes everything at once, and it's been that way for quite a while now. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] UVPE linux setup
In message 47c1b98393896f41a7bfe435442394e5268...@gmsdc.gerzio.ca, u2ug simpson-u...@gerzio.ca writes I have just setup uvpe 10.3.3 on opensuse 11.1 Locally universe seems to be running fine. However I can't access it via telnet or UO.NET either local or remote. I must have missed something in the installation guide - how do we configure uvpe on linux to accept connections via telnet uo.net ? I don't think you do! Okay, going back to my days with PI/Open on 7330s, but you logged into Unix then invoked UV. You need to set up telnetd on SuSE (it's disabled or, more likely, not even installed on pretty much every linux nowadays). You can then just log in remotely over telnet, and fire up UV either from the command line or from .profile. General advice nowadays would be (I've never done it myself) to fire up a terminal session over something rather more secure than telnet - something like SSH for example. UO.net, someone else will have to guide you with, but I'd check the firewall rules first ... Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] errlog question
In message ba3c8714a4e42a4fae7e0e551a40b96f03d82...@seisvr-11.sportsendeavors.com, Doug Chanco dcha...@sportsendeavors.com writes Can anyone shed any light on what the below message means? Sun Nov 29 18:08:54 -12729 yavegrt Program CONVERT.BASENUM.SKU.INFO.SORT: Line 221, Message[040037] Sun Nov 29 18:13:12 -8275 carbas 11949-KARS is not present on file. Sun Nov 29 18:13:56 -4529 brewbr 1962583 is not present on file. I am mostly curious about : the negative number (-12729 ) and the Message[040037] the negative number looks a bit like a user number to me. It's the classic signed/unsigned problem. It's displaying a user number between 32000 and 64000 (whatever the exact values are) as a 16-bit SIGNED value. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] uucp
In message 4b06d726.9020...@powellclan.com, Jeff Powell j...@powellclan.com writes Have you tried rsync? You can copy whole directories omitting files that have not changed. rsync -auv --compress source_dir destination_server:/dest_dir In our case we do this every two hours as a backup to a hot standby server. It works great. I can log in to the standby server and see sales that were made that day. iirc it's even better than that. I don't have the source immediately to hand, but (a) rsync was, I think, written by one of the SAMBA guys, so they knew what they were doing, and (b) not only does it only copy files that are changed, but if they are big it only copies the bits that changed! So if you're only updating the odd record in a several-megabyte file it won't understand the U2 structure, but it'll spot that just a couple of blocks have changed and only copy them. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] INPUT -1 not returning the correct value
In message 26304456.p...@talk.nabble.com, T Stokes t.sto...@monolith.com writes I have a situtation where the INPUT -1 is not returning the correct value. The program 'freezes' due to this problem. Has anyone seen this situtation. Do you mean INPUT var,-1 ? And what do you mean by not returning the correct value? What do you think is the correct value? My immediate guess is that your program is freezing because it's getting a null input, which is exactly what INPUT x,-1 is *meant* to do! Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] sending text messages to a regular cell phone
In message 5c6143395b81b6f741a069337c6f444e.squir...@webmail.ourldsfamily.com, Karl Pearson ka...@ourldsfamily.com writes and a copy is sent to the users's regular email inbox, which is useful if you use Verizon, which chops text messages down to 160 characters (I don't know why, they just do). That's the SMS standard. SMS stands for *short* message service, which is defined (iirc) to be a maximum of 160 characters. Typically it's actually less, about 130. Some systems will break up the message and send it as 130/160 character chunks. Others won't. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Command logging in Universe 10.2.1
In message of599fdc36.969bd559-onc1257656.0049829c-c1257658.00308...@nemian.lu, bpar...@nemianlife.lu writes Hi all, We are using Universe 10.2.1 to run a Life Insurance application and we use the UV command line to carry out various database administration tasks. All other access is via the application, where all transactions are recorded to an audit trail. We have the auditors in at the moment and they are very interested in having a compulsory tracing of the command line sessions logged as if a COMO ON was issued as the first command however I don't know of any way to make this mandatory, i.e. not allow COMO OFF. Does any one know any better. May be a bit more hassle, might not be ... See what the auditors think about indexes/triggers logging all changes to files. It might create a heck of a lot of disk usage, but it would achieve the same thing and probably be a lot easier to enforce. Or ask what commands they're worried about - obviously ED for a start - and put a wrapper round that ... I don't know how to force comos. To make it worse, you can do a COMO ON and it will kill the first como and start a new one - which is a damn useful feature on occasion! The other thing is, if you use a terminal emulator, can you force that to log all sessions? Many thanks for any help. Brian Parker Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Sequentially Hashed Files
In message 6e8c6c3ab287df45837a5816ce37531d68d6544...@daconosbserver.daconosbs.loca l, David Jordan da...@dacono.com.au writes Hi Ray Are you using UniVerse or UniData. I use dynamic files and never have to resize them in UniVerse. The OP said that the files keep on growing and shrinking - and that is not good for dynamic files. Although I would ask why you have to keep resizing and generally maintaining the files? You shouldn't have to do that with dynamic files (unless, of course, you're short of disk space and overflow wastes huge amounts). I'd go with Adrian's suggestion and use MINIMUM.MODULUS. *However* you did say the files regularly get emptied. Do you have an app that processes the file and empties it? If it's disk space you're worried about, just do a CLEAR.FILE inside your app once you've emptied the file - I think you can do it as a basic statement rather than an execute. You'll need to look at how to do it (it should be very simple) but I'd be inclined to do the below on every file after it's processed lock file clearselect select file if count = 0 then clearfile release That's all your maintenance and disk freeing automated for you... Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] OPOS printing to wireless receipt printer
In message ba2e471c9317804eb8470750f458bd983da42a4...@public, Tom Whitmore tewhitm...@ratex.com writes Hi, We have implemented OPOS for IBM registers. We are using HostAccess as our terminal emulator (BlueZone may be able to work as well) and the program communicates to the devices through a HostAccess hook. HostAccess does have routines that ease the implementation. You need to get the free software from http://monroecs.com/oposccos_current.htm, and install it on the PC. There is some configuration work you need to do, so you and OPOS reference the same device name. Once done, you can tell OPOS to print bold and it will know how to instruct the printer to print bold. What PC? I hope this makes sense! It is difficult to get started, but once you have the pieces together then things fall into place. If you read the OP, it sounds like he might not have a Windows PC. Sounds as if things are more likely to fall apart, rather than together! :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Sequentially Hashed Files
In message 5e5ef002669716488cf5f69288ba6b093bf...@msgl-lon01-es01.msgluk.local, Raymond P. de Bourbon rdebour...@msgl.com writes Does anyone have any experience using Sequentially hashed files? The documentation on Sequentially hashed files is not very explicit about what kinds of benefits or problems can be expected / experienced.. It also only hints towards one use case, but IMO doesn't do enough to clearly define when these types of files are actually beneficial and why... I've never had any need to use sequentially hashed files, but I'm quite happy to try and explain it ... Two questions to start with: Firstly, is your app generating sequential numeric keys? And do you only create records, never delete them? Secondly (less importantly), are your records of a roughly consistent size? The point of sequential files is that, if your answer to both questions is yes, a sequential file is trivially and perfectly balanced. And such files are common enough to make creating a special file type worth doing. Rather than trying to find some hash algorithm that does a good job, sequential files use the key value as the hash value. So if each record fills a group, record 1 goes in group 1, record 2 goes in group 2, etc etc. If two records fill a group, then it goes 1, 1:2, 1:3,2, 1:3,2:4, etc etc. I would have thought that if the typical record ended up in overflow, then sequential would be a good match for sequential keys too regardless of whether the size is consistent. However, DON'T take my GUESS on that. And bear in mind the advice is, if the answer to the first question is no, then DON'T go near sequential files. They're a special case, for a special-case situation. It's just that that special case is actually quite common. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Unibasic: Sample program - to extract data from Table
In message 02fb01ca4c38$ab3ad6c0$01b084...@com, Symeon Breen syme...@gmail.com writes Actually for many it is mass produced. Specification is being done to the absolute minutia, for example in the Unified Rational process, when generating use cases these get transmitted down to the architectural specifications and become the actual classes in the code (there is even software to build such skeleton code), with each method and property defined, again this is then used by the test plan to test each class/method/property etc. The coder kind of just fills in the gaps. In my previous employment we had whole teams of BA's, DBA's, System Architects, UI Designers, UX analysts etc, Then a bunch of people in the Philipines to do exactly as was written down in front of them. This kind of scenario happens a lot in larger projects. Which is probably why so many larger projects are either a disaster from the start, or rapidly turn into legacy (often before they're deployed) as the business needs change. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] I need your eyes
In message fba5ea46361c5c4fb61c02795188f6627b5a644...@vaunsw141.au.cbainet.com, Hona, David david.h...@cba.com.au writes It's interesting to see IBM's claims about they didn't buy U2 specifically (which we know they purchased Informix DB technology and U2 was a bonus) and it doesn't fit into their future goals. I find that amusing when they continue to sell databases like IMS and Informix C-ISAM (etc) and the like... which also don't fit... I guess the PR gurus had to come up with something!! Maybe IMS and C-ISAM are classic legacy, they're not selling and IBM can try to move them across to DB2. The trouble with U2 is it won't lie down and die ... :-) At the end of the day...it whatever works, I guess and whomever gives you the best value proposition. :) Well, if they're into selling DB2, and maintaining legacy stuff as it shuffles off to the graveyard, then yes, I don't think IBM is getting the value proposition they want from U2 :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] UniObject .NET Session and Common Block Bug
In message 13e4ae055d203648a4121114f2458c8c7a4adb8...@tefnut.perth.itvision.com.au , Adrian Halid adrian.ha...@itvision.com.au writes Thanks for the help. It seems once you name your common block you do not get the error anymore. I don't think that's the problem ... Just having a read of the documentation about the common statement in the documentation. A common area can be either named or unnamed. An unnamed common area is lost when the program completes its execution and control returns to the UniVerse command level. A named common area remains available for as long as the user remains in the UniVerse environment. From what it says I should be able to have them unnamed and loose it between subroutine calls. That may well depend on your emulation. AIUI, unnamed common persists for as long as a program runs. It's returning to TCL that clears unnamed common (as I say, AIUI, I've never used unnamed common). Named common persists until you log out. And I've had several obscure bugs where something tries to changed the size of common - that's not allowed. So if subroutine A effectively contains COMMON A,B,C while subroutine B effectively contains COMMON A,B,C,D,E,F then running them in the same execution environment *will* cause a crash because B has tried to redefine common. Regards Adrian Halid Senior Analyst/Programmer Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Migrating both OS and Universe Database
In message 663c35400910071056r188989b1x40a2e0b34083c...@mail.gmail.com, Kyle Moore kmo...@mooreimages.com writes Get the new hardware, plan the steps for install/conversion, create test cases, run a test conversion, adjust conversion procedures, run your test cases and, once all issues are resolved and documented, schedule downtime, blow away the target system so you have a clean platform again and pull the trigger. You run in to the same problems (maybe more because you have time to find them) but you aren't under the pressure of a downtime window to fix them. You can also get your users to do acceptance on the system before the real conversion (spread the responsibility). That makes the actual conversion much less stressful and more successful. Seeing as it's a migration FROM Solaris/UV TO Solaris/UV it shouldn't be a problem. Though the advice usually is upgrade one component at a time, so you should really look at upgrading to UV 10 on your old machine, bedding it in, then moving to your new Solaris and machine. That said, I wouldn't expect any problems doing it your way. Make sure your new machine is configured as closely as possible to the old machine, in particular all the paths for the UV system directory and accounts directories. Just copy all the accounts across. Make sure they have entries in the UV.ACCOUNT file. Make sure that anything in the global catalog is catalog'd on the new system - I think the command is MAKE.MAP.FILE - this will create an MAP file that lists all the catalog'd routines. Run that on the old system and go through it - make sure there's nothing there that will surprise you. Then hopefully you're all fit to go! There are some tricks - if you've lost the source/binary to a catalog'd routine I'm pretty sure you can extract it from the catalog - I'm not quite sure how, though. Cheers, Wol On Wed, Oct 7, 2009 at 4:48 AM, Louise Blair (DHL GB) louise.bl...@dhl.comwrote: Morning All, Looking for some pointers/information/guidance from anyone that may have already been through what we're going to be doing over the next few months. We will be migrating both our OS and Universe Database to a new physical server as detailed below: Current Future Sun Solaris 2.6 (SunOS 5.6) Sun Solaris 10 Universe 9.5.2.1 Universe 10 (at least) I remember many, many moons ago migrating from PiOpen to Universe and it was a weekend of hell, if my memory serves me correctly, and I want to cut down the level of hell during this migration if at all possible. We have many, many files, programs subroutines that make up our application, built up over the last 20 years or so which will mostly be moving. I also have a colleague that will be attending the U2 University in Liverpool next week and will gladly listen to any helpful information any of you, also attending, may have. Thanks in Advance, Louise Blair ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] hpux to linux
In message , Ken Hall k...@old-scholls.com writes Harold - I have done migrations from AIX to Linux and from Windows to Linux, and the nix to Linux migrations are quite easy. The biggest issue is that your backup method might need to be modified or rescripted to run correctly. Otherwise, moving from one nix server to another is mostly installing Universe properly. The application generally migrates easily with a uvbackup on the old system and a uvrestore on the new system (most easily done with UniAdmin tools). Why don't you setup a linux box on a 5 user UV development box and try the migration. Let me know if you run into problems. The other point to watch out for is endianness. Basically, if you backup your AIX system using uvbackup, then use uvrestore on linux, everything should be hunky-dory. If you use global catalogs then you'll need to recatalog, and then everything should just run, famous last words ... If you copy your accounts using cp, or rsync, or whatever at the nix level, then you'll probably need fnuxi. I'm guessing you're talking linux on Intel (big-endian), while AIX is probably little-endian. So all the binary info in the datafiles will be the wrong way round, which is why you'll need fnuxi to fix it. Once you've done that, and again recatalog'd any global catalog stuff, again it should just work ... Admittedly I've been running all on Intel, but I've had no trouble copying accounts between SCO Unix, linux, and Windows. It works fine with minimal hassle as long as you are aware of the pitfalls. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Start Phantom in Another Account?
In message 8c1cabce0909220808y7f888d42t245da8abd2e9e...@mail.gmail.com, Kevin King precisonl...@gmail.com writes Unidata 7.1.22. Is it possible to start a phantom job in another account? If so, how does one make this happen? This is a UV technique, but it's all TCL ... ED VOC TEMP I 0001: PA 0002: LOGTO OTHER.ACCOUNT 0003: PH NEW.PHANTOM FI : TEMP I know it's a bit of indirection, but if you can create and execute an entry in the local account, you can create a throw-away intermediary that does a logto and fires off the phantom you want. I've done on several occasions in UV, and I'd be surprised if it didn't work the same way in UD. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Worst Case/Best Case
In message f22433e8895313419b41ba386c0fdeea02125...@planomail.corp.tylertechnologie s.com, Bessel, Karen karen.bes...@tylertech.com writes You seem to have completely missed my point. Referential integrity is only ONE reason why SQL is a real RDBMS as opposed to U2. ONE example. There are MANY. It's ridiculous to consider U2's approach with referential integrity to SQL comparable - enforcing referential integrity at an applications programming level is much more time-intensive and error-prone than setting it up at the database level. And you're missing MY point. Actually, you're talking rubbish - do you really think Structured Query LANGUAGE is a database? But let's ignore that ... In a properly designed (and I stress the word properly), MultiValue application, the application does NOT enforce RI - the database does it! Do you understand the distinction I make between natural and statute integrity? In a relational database, probably some 90% of the relations are natural relations. In an MV database, those relations don't even exist - all the data is stored in a single record. The only integrity that is of any concern to MV is statute integrity - where some business guy has decided that a relationship (not enforced by the laws of nature) exists. That leads to those nasty situations where the data in the database does not co-incide with reality ... and because relational theory makes no distinction between natural and statute relationships, and because it has so many more EXPLICIT relationships with which to make a mistake than MV, then things are more likely (and they do) to go wrong. It's okay to wear your rose-colored glasses on if you want toIt seems that I'm the only naysayer here, and I'm sorry to be playing the doom gloom role, but IMHO it's reality - the writing is very clear on the wall to me that U2 cannot compete with the big boys. It's even less well equipped to compete now that IBM has left the room and a relatively unknown company has acquired it. I pity the companies that buy into this nonsense. There's a nice quote I see regularly - For a successful technology, reality must take precedence over public relations, for nature cannot be fooled. --Richard Feynman Unfortunately for us, at the moment, public relations is winning out in favour of relational. But if you want me to describe things in scientific terms... Relational theory is MATHS (and very good maths, too). Quantum Mechanics is MATHS (likewise, very good). Special Relativity is MATHS (also good). But relational theory (maths) is to managing information (real world) what quantum mechanics (maths) is to black holes (real world). In other words, it's just the WRONG TOOL for the job. And the poor business analyst is the sap who has to try and drive screws with a hammer. Did you get my point about storing lists in a relational database? About how it FORCES you to mix up data and metadata? And don't you understand why that is a BAD THING? Us humans have this religious thing. We *want* to believe in things. So we believe in mathematics. And even when the real world tells us the maths is irrelevant, we persist in trying to force the world to do what we believe the maths tells us it should do. (That's my point about quantum mechanics and black holes - QM just *doesn't* *work* for describing black holes - just as relativity doesn't work at the atomic level. As a scientist, I know both theories are wrong, but the man in the street says the scientist believes in them, so I will. The businessman says the mathematicians believe in relational, so I will too. And unfortunately, mathematicians tend to be even worse at believing in their theories than the man in the street!) (Note I said the maths is irrelevant. Maths is never wrong - it either adds up or it doesn't. But even if it adds up, that says ABSOLUTELY NOTHING about whether it works (or not) when describing a real-world phenomena.) Just my $0.02, for what it's worth. I've lived in the Pick world since 1988 and I don't want to leave it behind, but that time has definitely come. As soon as my education enables me to join the .NET/SQL world, I'm going to bid a fond farewell to this era of my career, that has enabled me to raise my daughters, straighten their teeth, buy a house, and all of the other good things that have come my way over the last 20 years. Sadly, I think you're making a sensible decision :-( From a science perspective I'd say you're embracing a false religion, but from a business pov it makes sense :-( Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users
Re: [U2] Worst Case/Best Case
In message of13b08d19.b2465ce3-on86257638.005863de-86257638.0059b...@ntn-bower.com , charles_shaf...@ntn-bower.com writes From a practical POV Karen is also right. It is a flat world for better or worse. It is nearly impossible to NOT interact with flat relational databases and hence the problem. So you're a member of the flat-earth society, are you? :-) Actually, that's probably a perfect analogy :-) All the evidence is there to the contrary, but you just can't persuade the members of the flat-database society that reality and their beliefs just don't, actually, coincide. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Worst Case/Best Case
In message 4ab7b2ef.6030...@sun.com, Lance J. Andersen lance.ander...@sun.com writes You cannot put Access in the same category as Oracle, DB2, Sybase, MySQL, SQL Anywhere, PostgreSQL... etc... Not even close. Oh yes you can! Note that Charles said designed! The database is irrelevant if the *design* is poor (that's the point of MV - it makes that whole CLASS of DESIGN errors a lot harder to make). charles_shaf...@ntn-bower.com wrote: By the way. To anybody that thinks a flat relational database enforces relational integrity. Have you ever seen an MS Access database designed by an end user? Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Worst Case/Best Case
In message d1be4de74846bd499929b632eaec736a0b7c9...@nt102.clark.root.local, Oaks, Harold harold.o...@clark.wa.gov writes Dawn: That wasn't Dawn, it was me, Wol. Where might we find the proof you speak of? Is there a nice, tidy paper somewhere? The Mathematician in me is quite interested. I would be overjoyed to show this to management. Let's start with sets. To a relational database, EVERYTHING is a set. So if your data happens to be a list, or a bag, you can't STORE it in a RDBMS, you have to MODEL it instead. So, instantly, you are mixing data and metadata. BAD MOVE. Now your applications need to know what stuff IN the database is data, and what is metadata. Relational theory says data is two-dimensional. We know it isn't :-) An RDBMS stores data in two-dimensional structures. An MV database stores data in n-dimensional structures. In Maths, the general always trumps the specific - n-dimensional trumps 2-dimensional. But being n-dimensional, we can pretend to be 2-dimensional. Relational can't pretend to be n-dimensional. Look on Pickwiki - http://www.pickwiki.com/cgi-bin/wiki.pl?MVDefinition - that was me - it was an attempt to demolish CD's twelve rules. Have a look and see what you think. I'm quite happy to carry on discussing this - either here, on community, or private email. And I suspect Dawn would like to discuss this to try and get a paper together. The basic problem is that relational *theory* is both *sound*, and *good*. So people assume that RDBMSs are sound and good too. But if you read my article on Pickwiki, it's obvious nothing could be further from the truth. Good maths does not necessarily make good engineering - and RDBMSs are an example of crap engineering. Thanks- Harold Oaks Clark County,WA Cheers, Wol -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Anthony W. Youngman Sent: Friday, September 18, 2009 2:19 PM To: u2-users@listserver.u2ug.org Subject: Re: [U2] Worst Case/Best Case In message 002a01ca37d4$03bc58b0$0b350a...@com, Symeon Breen syme...@gmail.com writes Just to pick up on one point - I am a .net developer but we use u2 as the data store in a growing business with many new customers every year. It is more a problem with peoples mindset than a problem with u2 technology itself, lets home rocket can tackle this head on. As I KEEP banging on, it's EASY to prove that U2 (and MV in general) is a far better database engine than an RDBMS, we just need to show that to management. Efficiency and relational are NOT compatible, and the maths to prove it is simple. Einstein's corollary to Occam - make it as simple as possible, but no simpler - relational is OVER simplified. Cheers, Wol -- -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Worst Case/Best Case
In message f22433e8895313419b41ba386c0fdeea02125...@planomail.corp.tylertechnologie s.com, Bessel, Karen karen.bes...@tylertech.com writes I'm with you there, Harold. I don't agree that U2 is a match for any real RDBMS like SQL. It just doesn't have the same capabilities. For example, referential integrity. In SQL, it's there, it's built in, it's an integral part of the system and you can't create orphan records if you set up your relationships properly. In U2, you can do that too, with application programming - lots and lots of programming, which can be (and often is) flawed. This is just one example of WHY you can't compare U2 to other RDBMS on the market - you're comparing pomegranates and pineapples, and when you are done with your comparison, U2 comes up lacking. Please - define referential integrity. In a MV database, for the most part, orphan records don't happen because MV *does* enforce integrity. It just does it completely differently - it stores related attributes together so they can't get orphaned, rather than scattering them across multiple tables. As has been pointed out before, it's easy to make a mistake when defining your relational database. Those sort of mistakes just don't tend to happen with an MV database. Don't get me wrong --- I love U2, and my career STARTed in a Pick shop. I'm sorry to see where it's gone over the years. Plus, to all the people out there who keep saying that it's us to list members to convince our companies to stay with U2 instead of migrating away, I say Hogwash. In anything other than a small mom pop shop, programmers do not have ANY input into these kinds of decisions. Management looks at a new and pretty system with lots of bells whistles, and they don't really care what platform is running on - We have to have it. I'm inclined to agree with you about PHBs ... but as for MV not having things like referential integrity, why not? It's easy enough to enforce it. Just like all the alleged advantages relational has, it's EASY to do in MV. You just have to think a little, and it's a shame it's not part of the basic system. But that gives us flexibility, which relational doesn't. How often do you hear of relational shops mired in concrete. And the MV shops run rings around them - projects ON time, ON budget, and ON target! Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Worst Case/Best Case
In message 12010-18...@sneakemail.com, Tony Gravagno 3xk547...@sneakemail.com writes From: Anthony W. Youngman Please - define referential integrity. In a MV database, for the most part, orphan records don't happen because MV *does* enforce integrity. It just does it completely differently - it stores related attributes together so they can't get orphaned, rather than scattering them across multiple tables. As has been pointed out before, it's easy to make a mistake when defining your relational database. Those sort of mistakes just don't tend to happen with an MV database. Sure they do. A fundamental difference between MV and relational is as follows: - In a relational DBMS, a DBA defines RI at the database level. The DBMS enforces the rules. - In MV, a programmer defines RI at the application level. BASIC code enforces the rules. This is why MV shops generally don't have DBA's, they just have programmers. But a well-defined MV database uses one record per entity. Looking at real-world INFORMATION (as opposed to mathematical DATA), a MV database stores all related attributes in a single record. That's my point in that wiki article about natural and statute law integrity. Properly designed, natural integrity is enforced by an MV database - delete the entity and all the associated attributes vanish with it. Create an entity and all the attributes appear (even if they are null :-) Yes, it's difficult to enforce statute integrity with an MV database, but then what happens when the real world decides to ignore your statute, as it will do ... Then the data in your relational database is bad because the database FORCES you to enter bad data. That's the point here - an MV database enforces integrity similarly to the way the real world enforces integrity. And when designing an MV database an entire class of integrity problems (natural integrity) just seems to happen right. That relates to a second fundamental difference: - In the RDBMS, the schema definitions contain the primary view of the data for both updates and reports. - In MV, the schema (dict items) is generally only used for reporting views, and application code is used much more extensively than in a relational app to create new views. In fact, an MV app doesn't really need schema at all - unless you build around it with something like SB+, but then it's still up to the application to use the defs properly for updates, not the DBMS. Touche. And I would agree that MV databases should be designed with far more rigour than they often are. But I would argue that most of the failings here are poor MV design, not a fault of the MV database itself. I'd like to see a synonym dictionary entry. So you are allowed only ONE D-type entry per field, which can be enforced, and as many synonyms as you like. Unfortunately, that's a change that will be hard to implement in a mature system ... The MV programmer is just as capable of missing an update as the RDBMS DBA. RI errors in an RDBMS are less likely because schemas don't change as often as they do in MV, and most/all updates run through a consistent schema definition. Unfortunately the way most MV people code, there are many points in the application where data is read and written, and the only rules applied to updates are the rules that the programmer remembers when he/she is in code. So it is more likely in an MV application that there will be RI errors. However this is also a source of versatility which we enjoy. I suspect that in relational databases, there are many points in the application where the only rules applied are the ones the programmer remembers when they are in code ... And as I said, natural RI tends to just happen in MV designs... I think you'll probably find *more* RI errors in the typical relational app than in an MV app. All that said, in MV we can add triggers. The BASIC code used for triggers is equivalent to the RI added to schema, and the MV DBMS should process trigger updates as faithfully as the RDBMS. I could be wrong but I don't think most MV developers use triggers too much, so we can't rely on this as the argument to say MV really does RI at the DBMS level. You can cheat and use i-descriptors as well :-) No, three cheers for the Pick model and all that, but it doesn't enforce integrity. That's the job of the application developer, and after that the DBMS just follows the rules it's given. Nor does the relational model enforce integrity. It's just that the MV model means that a large class of integrity constraints just happen. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick
Re: [U2] Problem with Multi-value market in general
In message 8b8d8a92acaa5b45986c590ff22600ce023ea...@bai-sbs.baisbs.local, Frank Eperjesi frank_eperj...@baipro.com writes Not to knock Rocket (Especially now that I guess I am be a dealer of there). Also I have no history with them A lot of what all you say as to U2 not evolving enough is true, but part of the reason for that and problems in the D3, U2 or mult-value world in general is that almost all the current multi-value providers are doing one or more of these things .: 1) Either living off the bones of their installed base and/or swooping in to feed off of whoever is still running D3 and its flavors, but doing nothing to attract new users. IBM falls into the 2nd category. 2) Jbase ( I think Cache) are successful application software companies that basically created their own version of Multi-value to migrate to save their investment. I have not looked a Jbase in a while, but Cache has some nice new features. Just to niggle (a BIG niggle), what you say of Cache might be true, but it's NOT true of jBase. jBase was written - like INFORMATION - to be a MV done right, for whatever value the people involved thought was right. Cache is the old MUMPS, and actually hired one of the people who wrote jBase (Jim Idle) to write their MV personality. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Problem with Multi-value market in general
In message 8b8d8a92acaa5b45986c590ff22600ce023ea...@bai-sbs.baisbs.local, Frank Eperjesi frank_eperj...@baipro.com writes Not sure what the purpose of your email was, but please stick to facts I don't mean to be rude, but it would help if you did a little investigation about your facts too. 1) I did not say I knew Cache was a MV application. I indicated I was not sure (see below) However, I did go to their web site and they say they are/were a high level Application provider for various medical record applications. It appears that is their major source of income. It isn't. It has an MV compatibility layer. Not sure why they developed Cache but it appears to be related heavily to te support their core application products (that make the big bucks). If their old core product was not MV they are just another company who has decided to feed off the old MV installed base. That is not necessary a bad thing, they may be a very good alternative, but it is what it is Cache == MUMPS (which is contemporaneous with Pick, ie it first appeared in the 60s). Just as Pick Systems has morphed into Tiger Logic, but the basic Pick product still remains, so has MUMPS morphed into Cache, but it's still basically the same thing. http://www.intersystems.com/healthshare/index.html P.S. In support of an older Pick client or ours that wanted to migrate to it, we are currently doing a conversion for that client to it. The conversion seems fairly easy and trouble free from an OLD version of Pick. 2) Jbase representatives when they originally approached me many, many, many years ago pretty much described the relationship between Temenos and Jbase just as I described it. The conglomerate that bought Jbase did so to insure that they had a platform that could run Temenos applications. This included IBM platforms not supported by Pick or whoever. Jbase did exist before they were bought out and does still exist. However it is a tool for corporate to save the value of their core Temenos applications that produces most of their income and allow their core application to run on huge IBM and other platforms that their Bank clients were insisting on. As a data base it is lunch money to corporate. P.S. It appears to be nice product with a lot of great features (I have no conversion experience) However, it is what it is. jBase presumably is a nice product (I have no experience of it). And as I said, the Cache MV layer was written by Jim Idle, co-author of jBase... Frank E. Eperjesi Vice President Business Automation, Inc. (714) 998-6600 phone (714) 998-6170 Fax Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Problem with Multi-value market in general
In message 25013-46...@sneakemail.com, Tony Gravagno 3xk547...@sneakemail.com writes I'm not sure why you're emphasizing OLD. The MV implementation in Caché supports many versions of Pick plus the Prime nuances of U2. :-) I believe one would find the level of thoroughness corresponds to the size and value of the prospect audience - in other words, Universe apps might migrate faster than any other. jBase followed the Prime philosophy INFORMATION is Pick done right. I *think* Jim came from Prime. And like I tried to do with MaVerick, Jim took the done right attitude. I'm not surprised he appears to have taken a lot from Prime. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Worst Case/Best Case
In message 002a01ca37d4$03bc58b0$0b350a...@com, Symeon Breen syme...@gmail.com writes Just to pick up on one point - I am a .net developer but we use u2 as the data store in a growing business with many new customers every year. It is more a problem with peoples mindset than a problem with u2 technology itself, lets home rocket can tackle this head on. As I KEEP banging on, it's EASY to prove that U2 (and MV in general) is a far better database engine than an RDBMS, we just need to show that to management. Efficiency and relational are NOT compatible, and the maths to prove it is simple. Einstein's corollary to Occam - make it as simple as possible, but no simpler - relational is OVER simplified. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] just a suggestion but ......
In message 007001ca385a$857a7e50$906f7a...@com, doug chanco d...@chancofamily.com writes does anyone besides me think that a new list rocketU2 or something could be added to talk about rocket, so that this list can go back to being technical? I certainly think that people need to vent/express their opinion about rocket/where u2 is going, if its dying (or not), etc .. but maybe its time to move thoes discussuions over to the community list or create a new one specific to rocket/ the sale of U2 I don't think there's any point in a new list ... I know we want (some of us at any rate) for this to be a technical list, but all this Rocket stuff is going to blow over fairly quickly. And imho it belongs on this list because it's stuff we NEED to know. Give it till the deal is formally approved and Rocket start talking, and all these threads will be gone. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Change is a coming! [not-secure]
In message be82285f86b8c24aaa426d8793d4fdd50b7d3...@doit-ex301.exec.ds.state.ct.us , Hennessey, Mark F. mark.hennes...@ct.gov writes snip Huh, does this make me a Rocket Scientist? /snip I'll go out on a limb and guess No ;) Who wants to be a rocket scientist? The science is dead easy, it's the technology that's the tricky bit! :-) (The science is just hydrazine + fluorine = lots of hot gas (or similar). The technology is making sure that the gas goes one way and the rocket the other, rather than everything going in all ways simultaneously ...) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] UV and triggers
In message 4a63297f.2030...@comcast.net, Charlie Noah cwn...@comcast.net writes John, I can't believe you're recommending this. The only reason Universe treats a RETURN as a STOP in a top-level program is that Universe tries to be everything to everyone, forgiving mistakes and blunders, and trying to determine what the user really wanted, no matter how badly the code is written. In 1998, we converted from Universe to Jbase, and you can't imagine all the problems we had fixing bad Universe code that worked (sort of) so that it would work in a system that requires you to do things correctly. When coding and breaking rules, you have to consider that the code may be migrated someday. This is a case of just because you can, doesn't mean you should. I doubt that's the case ... STOP may mean stop, but RETURN means go back one level. So RETURN makes perfect logical sense even in an outer program... I'm not talking about using the strengths of your particular implementation even though they are unique to your implementation, just doing dumb things that will bite you (or your predecessor) down the road. I put program in quotes in my last paragraph, because I'm not even sure all variants of BASIC accept the PROGRAM keyword! Certainly in my early days programming PI I never used it, and I *think* the reason I didn't was because it didn't exist in INFOBASIC. iirc it was added at the same time as the FUNCTION keyword. So I wouldn't say John is recommending you use return, I'd say RETURN *always* works because subroutines should always end in a return, and at one time everything BASIC was a subroutine. Oh - and I think PROGRAM and SUBROUTINE are actually synonyms as far as the compiler is concerned :-) Just my 2 cents worth. And my tuppence. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] COMMON misMatch
In message 2b1985075953c947a7a53a53987adab22b5...@arnold.hkmetalcraft.lan, Brutzman, Bill bi...@hkmetalcraft.com writes A user here obtained an error message... COMMON size mismatch in subroutine HOP.R87. Upon logging out and then logging back in again... The HOP program was able to launch without this error. Help with a diagnosis and more reliable cure would be appreciated. This user is WinXP / Dynamic Connect / UniVerse 10.1 / HP-Ux 11i v2. Has this program been created by an auto-generator program? Did the user go into the program by the *same* *route* both times after logging in? If you've got a program generator auto-generating named common names you have to be very careful - I fell foul of that a couple of times with PACE. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] More questions on indexing
In message of44cc39a8.134b5a9d-on882575f5.006f2b76-862575f5.006fa...@usbank.com, bradley.sch...@usbank.com writes Thanks, Rod. One more thing. I've written a test program to simplify index creation and building. When I run it under type U or type P using BUILD.INDEX or BUILD-INDEX, it gives me the Enter New line to continue... prompt after each screen of *'s. Definitely not desirable for an automated process. It doesn't do this at TCL. How can I work around this? I'd rather not throw a bunch of DATA statements in there. EXECUTE 'DELETE.INDEX LS.INV.NUM N.CONTRACT.KEY' DATA 20 EXECUTE 'CREATE.INDEX LS.INV.NUM N.CONTRACT.KEY' EXECUTE 'BUILD.INDEX LS.INV.NUM N.CONTRACT.KEY' iirc, UniVerse has the NO.PAGE keyword. Does UniData have it? And if so, does it work for BUILD.INDEX? Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] General guidelines on indexing
In message ba2f7087c5e55f4cb14ec12d9bee354501a20...@svr-email-1.civica.root.local, Edward Brown ebr...@civica.co.uk writes After indexing, we made a lot more use of the SETINDEX and READFWD logic in our programs. I find this curious / disappointing - is it really the case that unidata can't take the mix of indexed / unindexed dictionary items and do just as efficient a job as the code you're writing? Also, the performance of dynamic arrays need not be as much an issue as you've found. If they're built up with -1 rather than a counter then the speed penalty of adding items to a very large list is much the same as a tiny one. Is it? Depends on how the memory is managed, but I understood that adding items required a memcpy every now and then. The bigger your dynamic array, the more expensive is the memcpy when it's required. The only real issue with dynamic arrays is if the machine does not have the physical memory available to hold the variable. Been there, done that ... 32 users on a 16meg EXL 7330. It spent most of its time in swap ... (Finally talked the company into upgrading the ram when a 1meg stick went bad. Bought 32meg for that machine, and used its 16 to upgrade the other two machines to 24 each - Prime replaced the 1meg free as part of the upgrade deal :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Calculation on multi-valued elements
In message 4a53ea56.5050...@gmail.com, Charles Stevenson stevenson.c...@gmail.com writes The semi-colon way of writing is more readable, but UV does not let you reference such an I-descriptor (as reusable code) from some future one. For example, next year you want to know how fast/slow certain programs run during certain times of the day. You'll want to use the RUN_TIME result within a second I-descriptor. UV won't let you if it uses multiple buffers as written below. I think this is a serious defect dating back to Prime but since it has never been changed, I'm guessing it would require tremendous rewrite. I could be wrong (it's ages since I used INFORMATION), but I didn't think Prime suffered from this problem ... I don't remember Prime i-descriptors pulling in the source of i-descriptors they referenced, and this problem came as a bit of a shock to me when I started using UniVerse. There are a bunch of discrepancies between PI and UV, and I always thought this was one of them. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] FW: Valid Backup on Windows Universe
In message 4a4122b1@gmail.com, Steve Romanow slestak...@gmail.com writes Kevin Gusler wrote: Steve - Why do you think this? The failure I see of this system is that even though the db is paused, it is not in a safe state for backup. Once you broke the mirror would it not be safe to backup that array?? Kevin Gusler Applications Developer Mortgage Builder Software, Inc. Main:800.850.8060 ext. 106 Fax: 248.304.0601 www.mortgagebuilder.com The db is basically in mid sentence. The commits are not atomic. If you restore to the tape, you may in my case have half an order or an invoice half processed. the unidata dbpause does not make people get out of maitnenance screens or release locks, so records can be in an uncommitted state. Risk is small, but it is present. ___ But that's a programming issue. It's easy to do exactly the thing with a SQL database. It's just that because pretty much EVERY write to a SQL database will corrupt integrity until the transaction is complete, SQL programmers use transactions as a matter of necessity. Because in MV database writes have traditionally often been complete transactions, we don't have the same mindset. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Mixed Case UV Basic Programming Standards.
In message 6806801183013245804b6a7c6b34750715a...@hera.gnosys.local, phil walker p...@gnosys.co.nz writes As anyone written or thought about writing a code refactorer for U2-BASIC, this would/could/should standardize code semantics and also appearance? If this could be made configurable to the style you want to use then that would be good...and any program with a GOTO in would be deleted...without warning ;-). So any proper well-written program would get deleted without warning? :-) About the ONLY place where GOTO really is the best way of handling things is dealing with error conditions - and this is the code most programs (not only in MV) lack. Most well-written programs will contain GOTOs, to clean up the mess caused when the program trips over something the computer or user should have done and hasn't (and it's not always the programmer's fault - it could well be hardware failure, network dropped, or something stupid like that). Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Program Comments/Documentation/Notes/Revision History
In message 1fdb619c0906290746n2d553c0ao867cb5d64327b...@mail.gmail.com, Dawn Wolthuis dawnwolth...@gmail.com writes On Sun, Jun 28, 2009 at 4:32 PM, Steve Romanowslestak...@gmail.com wrote: Mecki Foerthmann wrote: It really boils down to this. What can you do with Python in a U2 database that I can't do with Pick Basic? Umm, real unit testing ala nosetest I realize this is the U2 list, but I thought I would note that at least one implementation of mvbasic is enhanced for OO and does have a unit test framework (like junit). I will state up front that I'm using Cache' then just refer to it with other terms such as our MV toolset as this is not intended to be a sales pitch. Many of you will know this, I expect, but Jim Idle (of jBase fame) worked for Cache for a while. I meet him every now and then on a compiler-compiler mailing list. So I would guess he was rather involved in writing a lot of this stuff for Cache :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] [UV] Currently preferred Linux distro
In message d85db451e1913e4d9889cfb25543b247d07...@auexci01.ad.internal, Boydell, Stuart stuart.boyd...@spotless.com.au writes Just calling on the wisdom of the group to ask what is the currently preferred Linux distro for running UV? iirc, if you want a *supported* environment, you want RHEL or SLES. Bearing in mind CentOS is the same as RHEL, and OpenSUSE is close to SLES, those would be options. Fedora of course is RedHat, so you're probably okay there. But as far as I'm aware, it should run on most distros without problems, it's just the installer that will give you grief. The two things to watch out there that I'm aware of are that one of the cpio options has reversed its meaning - you may have to remove or add an argument, and of course the rc.d environment varies between distros so you may have to configure the UV startup stuff manually. We will be putting low user count development test environments on it and ultimately we are trying to emulate as closely as possible our AIX 5.3 production environment. We are currently using Fedora Core 6 32bit but I feel we should go up to a newer 64 bit platform. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Linux / file create default permissions
In message ofa49394a8.99d841e0-on862575d7.005ef02c-862575d7.005f8...@ntn-bower.com , charles_shaf...@ntn-bower.com writes The umask tells the file system which bits to NOT set when th user creates a file. I believe the default is 777, OR the permissions of the parent directory. Then the umask is applied. The umask is usually set in the users .profile, but can be globally set in the /etc/profile depending on the flavor of UNIX. Try man chmod and play. I've just taken a look and modes 4000 and 2000 look promising although they're not quite what I thought. They'll force owner (and maybe group) to be what you want. It might not work for dynamic files, though :-( Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] UniVerse Admin Tool for Window
In message d76bac9cbfb0f040b820c12e55b227ca01351...@mail.corp.rammutual.com, Amy Raisanen araisa...@rammutual.com writes I had the same issue happen. I had upgraded Universe to 10.2.10 on Windows then installed UniAdmin. None of the locks or users showed in Uniadmin, when I contacted support on the issue they suggested I back down a version of Uniadmin. I think the bad version was either 1.3.6 or 1.3.7. We did back down a version to get the functionality back but have since upgraded to 1.3.8 and those now things work fine. If that's the case, might it be an idea to download the latest evaluation client CD from the IBM website? I *don't* *think* the client tools are timebombed. Just don't install them on the server itself unless someone at IBM confirms that's not a problem :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] [u2u] UV on Windows question
In message 4a2473a1.7010...@comcast.net, Scott Richardson cheetah...@comcast.net writes Thunderbird does the list on reply w/o intervention. PS: Wol? any better luck with LinkedIn over across the pond? Hi Scott, Still no luck. I can view your invitation just fine. Click on accept, and it dies ... Anybody on the list got any ideas? One thing I should add - Demon have just upgraded my exchange to ADSL2, whether that's got anything to do with it I don't know - my internet keeps crashing in the evening ... Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] [u2u] UV on Windows question
In message 004001c9e38a$84e5bb50$8eb131...@com, Symeon Breen syme...@gmail.com writes My meaning is that you would run uv on windows server behind a firewall and the server would not have a user browsing the internet, or be accepting and opening emails. Viruses do not just appear on a windows box. The difference here being server systems of whatever o/s rarely get the exposure that allows a virus, this is completely different to desktop systems. Viruses for linux do exist but the linux desktop market is very small so the virus market is very small. Viruses do exist, yes, but not only is the desktop market small, also there are very few vulnerabilities for a virus to exploit. I understand one virus did spread very well, but it targeted a known vulnerability in a COTS router. The reason that spread was that, unlike most linux installs, routers aren't regularly updated to remove vulnerabilities. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] [u2u] UV on Windows question
In message snt110-ds1230a0f57ff48ef6e6dd52cf...@phx.gbl, jpb-u2ug jpb-u...@hotmail.com writes I just rebooted my Linux systems last week the first time in about 6 months. The reason, because the windows servers on our network have been changed so many times that they were finally starting to affect Samba on the Linux server. I couldn?t just change the IP addresses anymore and restart Samba, I actually had to reboot the system. I have had the system up so long that I got a message like Lee?s but I try to reboot at least once a year just so the heads don?t weld themselves to the disk surface if I do have to bring them down for some reason. I don?t think we have ever had a windows server stay up more than a week so I think you are either lucky or you very seldom install a patch on your server. The standard with Windows is one server app per server. Then they're very stable. But try running SQL-Server, Exchange, file serving, UV, Domain Services etc on one box, and it'll be up and down like you-know-what... Jerry Banker Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Best algorithm for UV part files
In message 0f0fc5f04b472746b01fa2c4897cf972beb5364...@excl01.mouser.lan, Baker Hughes baker.hug...@mouser.com writes Very helpful Stuart (and Brian) thanks. 1 follow-up question: given your algorithm below, when you edit or copy those records over, what does the key look like. E.g. do you have to prefix the key generated by your algorithm, with the part file number or name? To edit the record would you simply: AE HAIRY.PF.NAME B123456 Or AE HAIRY.PF.NAME 23-B123456 ? Neither of these would work ... if you don't change the key, then the record will not move file ... you'd have to do something like AE HAIRY.PF.NAME B123456 SAVE 23-B123456 FD I can't remember exactly what I did ages ago, but I remember that our system wrote a sequential key when it created new records. The part-file algorithm looked for a double-barrelled key - date-number - and if it found the date it returned a date-specific part, otherwise it returned 0. The archive routine selected part file 0, scanned it for records to archive, checked that the date-specific partfile existed, then renamed the record from number to date-number (in basic, it did a read, write, delete). When working with part files, there are two things that you REALLY want to avoid, unless you're prepared to give your users a fair bit of grief. Either (a) changing the part algorithm, or (b) an algorithm that can come up with a non-existent part in normal use. We got round it by defaulting all records into part 0, and only moving them elsewhere when we archived them. Another way would be for the part number to grow sequentially, but MAKE SURE you always are on top of it, or simply to make the original number of parts so large you don't think you'll ever have to add any more :-) Cheers, Wol 74 Clear regards from Ft Worth, -Baker -Original Message- From: owner-u2-us...@listserver.u2ug.org [mailto:owner-u2-us...@listserver.u2ug.org] On Behalf Of Boydell, Stuart Sent: Wednesday, May 27, 2009 7:34 PM To: u2-users@listserver.u2ug.org Subject: RE: [U2] Best algorithm for UV part files Well, it's a 'how long's a piece of string' type of question but we had a requirement to archive old data from a heavily used file. The file has several indices too many and is in use 24/7 This makes copy/delete a fairly heavy impact on the system and archiving was always an issue. The IDs in the file are a mix of numeric and prefixed keys eg 123456 or B123456 from different systems. We decided that a part file would solve the issue because you can remove the part from the DF and because the indices are attached to the part it's a very quick, clean action. If for some reason you need to put it back - it's also very quick and easy. So, we came up with a part file system which puts 10 sequential IDs into a part then moves onto the next one. For our requirement numeric keys go in a different part from prefixed keys. Theoretically this makes it easy to archive chunks of data by moving a whole part and also to find specific date ranges for records based on their keys within a particular part. @PART.ALGORITHM = OCONV(@ID,'MCN':@VM:'MR05')[1] + 1;IF @ID MATCH '0N' THEN @1 ELSE @1 + 10 Regards, Stuart Boydell -Original Message- Does anyone have an opinion about the best algorithm to use for UV distributed files? The goal is ease of moving the records in the distributed files to other files. ** This email message and any files transmitted with it are confidential and intended solely for the use of addressed recipient(s). If you have received this communication in error, please reply to this e-mail to notify the sender of its incorrect delivery and then delete it and your reply. It is your responsibility to check this email and any attachments for viruses and defects before opening or sending them on. Spotless collects information about you to provide and market our services. For information about use, disclosure and access, see our privacy policy at http://www.spotless.com.au Please consider our environment before printing this email. ** --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/ This communication, its contents and any file attachments transmitted with it are intended solely for the addressee(s) and may contain confidential proprietary information. Access by any other party without the express written permission of the sender is STRICTLY PROHIBITED. If you have received this communication in error you may not copy, distribute or use the contents, attachments or information in any way. Please destroy it and contact the sender. --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/ -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley
[U2] LinkedIn
Is anyone else having trouble? The site's totally unusable for me - it seems to be serving static pages okay but anything that requires a database access just dies with a server error. As that includes the how to contact us page, I can't even get in touch with them to ask what's going on! (I'd like to accept and send some join my networks, but of course that's completely up the duff, I've been sitting on stuff since the middle of last week!) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Opening UniBasic Source Code - Tab Key
In message ae8acc132c0f1c4de288c5fa7b72a53e.squir...@webmail.ourldsfamily.com, Karl Pearson ka...@ourldsfamily.com writes CHAR(8) is a tab, IIRC. That's a backspace, actually. Tab is char(9) Karl Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] PCL problem on U2
In message e8209a9427264095bc3fe78d6908d...@it3, David Hoover dhoo...@illinoislock.com writes I am trying to place a logo that is stored at /area1/dh.temp/M100_3Logo.pcl on a form that I am creating in PCL5. I have no problem with the shading, boxes, or data. I am having a problem getting the logo to be printed. Base information: OS - RedHat Linux Running U2 Printer is IBM InfoPrint 1464 Color Logo file original name: 3Logo.jpg After running thru IPX20xx converter from IBM: M100_3Logo.pcl (Set this up as Macro 100) Here's the code that I am currently trying to work with: 0001: EXECUTE 'SETPTR 11,, NHEAD, FORM IBM, AT PR33' 0002: PRINTER ON 0003: * 0004: PRINT ON 11 CHAR(27):E 0005: PRINT ON 11 CHAR(27):f100Y 0006: PRINT ON 11 CHAR(27):f0X 0007: PRINT ON 11 CHAR(27):a540h780V 0008: PRINT ON 11 CHAR(27):*t300R 0009: PRINT ON 11 CHAR(27):*r100A Okay, this is setting up your print job ... 0010: EXECUTE SH -c 'lp -d PR33 /area1/dh.temp/M100_3Logo.pcl' This is sending the logo as a DIFFERENT print job. That's your problem. I'm not sure how you'll fix this, but you need to read the logo file into U2, and then dump it to your print job without the binary getting corrupted. The other way to do it (I'm pretty sure it's possible but I've never done it) is to download the logo to the printer as a logo or font or whatever, then have a code in your print job to print the prestored logo. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] UV to SAP migration disaster
In message 214234.57230...@web51710.mail.re2.yahoo.com, Don Robinson donr_w...@yahoo.com writes Perhaps a bigger push into the open source space would help. I recently read that MySQL has a long history of bugs, like losing all your data, and some of them will not be fixed. Yet it remains the most popular free database. For those who haven't been following the news, it's just been announced that Oracle are buying Sun. Sun bought MySQL a year or so ago. So what will happen to MySQL is debateable ... Oh - and as for bugs, and some of them will not be fixed, I doubt that's true any more. It *WAS* a deliberate design decision that MySQL was optimised to give a very fast read-only backend for things like web databases. It's now turning into a full-blown industrial-strength database. We have a better mouse trap, why doesn't the world know it? Because they've been conned by the relational, it's maths so it must be right mantra, without realising that the 12 rules underneath it are rubbish. Just, my 2 cents worth. BTW, I'm looking for programming work! Universe, Unidata, jBASE, etc. Linux, Aix, Windows, Mac. So am I, unfortunately ... :-( Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] UV to SAP migration disaster
In message 461165.65382...@web58106.mail.re3.yahoo.com, Laurie Blain laurie_bl...@yahoo.com writes I heard that IBM bought Unidata and Universe to kill them off. To their surprise many of us are refusing to let the U2 die. I think the problem WAS (apparently the people involved have moved on) that the top people in the DB2 group (of which U2 is a part) were very heavily emotionally involved in DB2 itself. They've now been replaced with people who are a bit more business-minded, and who realise that the U2 databases are valuable. Maybe U2 will do an Informix on IBM (remember when Informix took over Ardent it turned into a reverse-takeover, and it was Informix that came off worse :-) Laurie Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] UV to SAP migration disaster
In message 367f3b3911959c4da9991c00166f620c01b90...@euducex2.europe.ad.flextronics. com, Glenn Sallis glenn.sal...@nl.flextronics.com writes It's yet another story that makes people who know and understand the multivalue database model cringe. Often one of the reasons for migrating seems to be due to decisions being made upstairs by people who have not bothered to consult the people with the knowledge to inform of the technical realities and work involved in such a major change. Quite often because the existing system has been starved of resources... The in-house UV system I worked on a few years back wasn't upgraded because management wouldn't let us, then they decided to outsource building its replacement. We could easily have done what the contractors did, we just weren't allowed to try. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Does anybody have a hot backup server? {Unclassified}
In message of087fe71c.cb64f47e-on8525757f.0077b5a9-8525757f.0078f...@us.ibm.com, Neil Morris morr...@us.ibm.com writes Hi Mike, Your inquiry was brought to my attention today. Could you possibly clarify what is meant by headers are flushed when they are updated in regard to UniVerse dynamic files? Is this in reference to the information in the T30FILE structure in the UniVerse disk shared memory segment being written to the header of the file when an update would result in changing one of the dynamic file parameters stored in the header? Or does this comment refer to the updated information in the header of the file being flushed to the physical disk via a sync operation? If the former, then it is my understanding that UniVerse has always updated the header immediately when a dynamic file parameter changed and there is no specific release in which this capability was added. If the latter, the only way to force the update to be flushed to disk immediately is to have the file activated for logging with UniVerse Transaction Logging and have Checkpoint mode enabled. Perhaps I'm not understanding the question so any clarification you can provide would be appreciated. Hi Neil, It's long been common knowledge (which doesn't mean it's correct :-) that UV read the dynamic file parameters into memory when the file was opened, and didn't always write changed information back until the file was closed. You were therefore advised that it was not safe to copy even a quiescent dynamic file at OS level, because the header and data might not be in sync. I remember a posting (from an IBM'er) saying that this had changed round about rev 10, and that header and data should now always be in sync on disk. (ie the post was about rev 10, the change might have been earlier. But the implication was that it *had* been changed.) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Does anybody have a hot backup server?
In message 36de7cd3e51d93408bad9eb5fe56a477ae2...@sbsprod.infodata.lu, Manu Fernandes e...@infodata.lu writes Hi, Under Universe/Linux we use Data Replication Block Device. www.drbd.com (open source linux product) A low level software called between logical partition and hdd partition. Cluster over LAN. We define it on partitions where Universe db is installed (including universe himself). On the hot-standby server, the same partition is created and not mounted (universe is not started ~ for this raison, we don't need IBM's license) On fail-over, we switch the ip adress to the hot-standby server, we mount the partition, we boot uv and users can work ... 2 minutes. One restriction, we don't use DYNAMIC files because the headers are not permanently flushed when write occurs files. I believe (iirc) IBM have confirmed that, as of the latest versions of UV, the headers are flushed when they are updated so dynamic files are now safe ... Coupled with Universe Transaction Logger set on strategic files to ensure record update integrity. My two pence. Manu Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] time Verb {Unclassified}
In message 000301c9a111$d8994890$89cbd9...@on.net, Ken Wallis ken_wal...@internode.on.net writes Mike, I'm not sure that is still the way it works. Windows systems may still keep time in localtime, but they have mechanisms to compute UTC or any other localtime from that via the registry. TZEDIT.exe is your friend. That is, I believe, still the DEFAULT - the hardware clock is set to local time. However, I think MS have had enough grief from people moaning about servers being physically in one time zone, but logically in several different/other zones, that it is now possible to configure Windows to keep the hardware clock in UTC. Despite booting several OS's on my home pc, I haven't bothered to work out how this works, so when the clocks change the time on my system tends to go haywire for an hour or two, until the timeserver software cleans up the mess. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] ODBC Connection to Universe on HP-UX
In message 291595756f51e34ea52a45b3a191dae3f5b...@anorpexbe01.auca.corp, Martel, Henry henry.mar...@wearguard-crest.com writes We are thinking of changing our application connect to Universe through ODBC connections. Has anyone had experience doing this? OleDB or ODBC? Are there any performance issues or considerations that we need to be aware of? If you can use OleDB you're probably a lot better off. I'm sure somebody will correct me if I'm wrong :-) but as I understand it OleDB is the main connectivity thing, and ODBC is a layer on top of it. Best avoided if you don't need to use it. While it still makes an awful lot of sense to run HS.SCRUB, with OleDB you don't need @SELECT (all your fields and files are visible, which may or may not be a good thing :-), you don't need HS.UPDATE.FILEINFO, and generally there's just less to go wrong! And in my experience, things do go wrong ... accessing UV using the MS-Access ODBC as a front end is currently passing dud SQL to UV :-( Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] ODBC setup error
In message 47c1b98393896f41a7bfe435442394e510f...@gmsdc.gerzio.ca, u2ug simpson-u...@gerzio.ca writes We have an account setup to allow odbc access to specific files. We added access to a new file and ran HS.UPDATE.FILEINFO which processed a few files but then aborted the session with that old standard : Abnormal termination of UniVerse Fault type is 11. Layer type is Unknown. The files that were processed are accessible via odbc but no others. I tried removing all data files references from the VOC , so there shouldn't be any files to update and it still gave me this error. I also tried disabling and re-enabling file access via HS.ADMIN but the same result - we get the same error when the enable tries to run the HS.UPDATE.FILEINFO command. Quite likely. What version of UV are you on? I am assuming that it is hitting a bad file and blowing up, but I don't know what that file might be as the order that files are processed seems to be pretty random. Try SELECT VOC WITH TYPE EQ F and see if that's the order it's processing them in (I think it probably is). This is a major issue. Can anybody shed some light on this ? Look on the UniVerse page on PickWiki. There's a page about setting up ODBC - it's pretty out-of-date because it's based on 9.6, but it may well give you some clues - especially about how to get HS.UPDATE.FILEINFO into verbose mode, which stands a good chance of telling you which file is blowing up. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] HP, Cron, Como, Execute, Capturing . Not
In message 619f7eea7a19a945876205ff14ee8de10116e...@win2003.office.stamina.com.au, Ross Ferris ro...@stamina.com.au writes Have you changed the code yet to avoid problem? You haven't mentioned version of UV -- if not current, check later GTARs to see if issue identified/resolved. Get the customer going with a redirection/read ... obviously (?) new code, but an ls -l /tmp/unique_filename if you want the COMO to get it as well, try ls -l | tee /tmp/unique_filename will work across ANY *nix platform -- assume cron is running as same user/permissions as when you fire from TCL I'm not sure whether tee is available across all nix platforms, though. Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] UniData LIMITs
In message 4983be3c.4070...@advantos.net, Bill Haskett wphask...@advantos.net writes Ken: I think so. I'm looking at this as the best (or easiest) alternative. Thanks very much for the thought. Bill I've rethought my idea a bit. What I don't quite get is how you build up that massive list of clients in the first place. Am I correct in thinking you own the customers, the customers own the clients? Okay. How does the database know which clients are owned by which customers? Sensible design says the CLIENT file will have a field called CUSTOMER.OWNER (or similar). Certainly there won't be a list of clients in the CUSTOMER file - that could be a massive mv list and is poor poor poor design ... So you should be able to do a EVAL LOCATE( CUSTOMER.ID , TRANS( CLIENT, CLIENT.ID, CUSTOMER.OWNER, X), 1, 1, 1) in your query. Note I'm not sure of the locate syntax, I'm looking for a sub-value (assuming CUSTOMER.OWNER is multi-valued and TRANS does a LOWER). If this returns true, your customer can see this client. There's your select, and it'll probably run faster, too :-) Cheers, Wol -- Anthony W. Youngman pi...@thewolery.demon.co.uk 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Speeding up processing through large dynamic table
In message [EMAIL PROTECTED], Joshua Gallant [EMAIL PROTECTED] writes When running through an array with a for next loop the last item processed isn't remembered so the program needs to traverse the entire array for each record and will slow down as you get to records later in the process. I thought UV had optimisations so that (as long as it is FIELDS) the last item processed IS remembered and it carries on scanning from where it left off. Instead of this: A11 = DCOUNT(IN.TAB,@AM) FOR A1 = 1 TO A11 IN.LINE = IN.TABA1 SWAP @VM WITH @AM IN IN.LINE CUST.NUM = IN.LINE1 CUST.DESC = IN.LINE2 NEXT A1 Seeing as this snippet does nothing much, I'd guess that the problem is actually somewhere in the code you've snipped - do you assign back to IN.TAB at any point? That *really* will get slow. Try something like this instead: LOOP REMOVE IN.LINE FROM IN.TAB SETTING MARK SWAP @VM WITH @AM IN IN.LINE CUST.NUM = IN.LINE1 CUST.DESC = IN.LINE2 WHILE MARK DO REPEAT That will keep track of where you were in the array and pick up where you left off. Let me know how that works out for you. I'd do the following (INFORMATION-style flavours only!) A11 = DCOUNT(IN.TAB,@AM) DIM IN.TAB.M(A11) MATPARSE IN.TAB.M FROM IN.TAB, @FM FOR A1 = 1 TO A11 IN.LINE = IN.TAB.M(A1) SWAP @VM WITH @AM IN IN.LINE CUST.NUM = IN.LINE1 CUST.DESC = IN.LINE2 NEXT A1 If you've been assigning to IN.TAB, you can then get it all back with MATBUILD IN.TAB FROM IN.TAB.M, @FM (Check my syntax - I've probably got MATPARSE/MATBUILD wrong - I look them up in the manual when I need them :-) Something to bear in mind ... using dimensioned arrays will ALWAYS beat dynamic arrays for speed. Which is more important to you? The flexibility of dynamic or the speed of dimensioned? And in PI-style accounts you can have the advantages of both - dimensioned arrays can be defined on the fly, at a small cost in clarity-of-code. With small arrays I can never be bothered with dimensioned arrays. As soon as I start heavily manipulating arrays, or they start getting big, I look at dimensioned arrays and wonder are they worth it. Sometimes they are, sometimes they aren't. Looking at your case, they almost certainly are! Cheers, Wol -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Universe Triggers
In message [EMAIL PROTECTED], Ken Wallis [EMAIL PROTECTED] writes UniData has a VOC_READONLY environment variable which can be set that allows it to run successfully in a directory where the VOC has no write permission. Perhaps UniVerse has the same or similar. Bear in mind the OP of this particular thread said If I can update the VOC, then your comment is irrelevant, but from what Susan says, I presume she has some magic way of dealing with that. I can't conceive of a way to make this safe without setting suitable OS level permissions. I can see how it could be made safe enough that you couldn't get round it with normal verbs and editors, but I can't see how it could be fixed so that a programmer couldn't get around it, or even a savvy user setting up VOC pointers from another account. Exactly. How on earth is Remote VOC going to prevent me from messing about with the system if I have update capability and can recreate an unprotected, vanilla, VOC entry for ED or COPY or whatever verb happens to take my fancy ... ? Cheers, Wol -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Updating UV/UD PE
In message [EMAIL PROTECTED], Tony G [EMAIL PROTECTED] writes Anthony Youngman wrote: Point 6 ... I know you think that c:\ibm is nasty, but be aware that (I think it's UV itself) does NOT like spaces in pathnames. I always accept the defaults, I've had some weird problems installing in Program Files, the space means various services fail to start. Cheers, Wol Ignoring the technical focus for a minute: Doesn't anyone there know how to use short paths? Can't someone at IBM modify the installation to prevent the use of long pathnames if the product can't handle them? This is installation code for an IBM product, not a hack by some kid who writes Java code to work his way through school. Why do people tolerate an issue like this from one release to another? This is UD v7 and UV v10 in 2008, not v1 in 1995 (pre-long-path). I'm not quite sure what you're getting at here, but never mind ... Don't forget UV comes from a heritage where space was an illegal character in file names. (And it still is, for quite a lot of other programs that run under Windows, too. Spaces cause me grief, even in Windows itself!) I know at least one U2 administrator here that has site protocols about putting data into the C drive. I can see a lot of sites not accepting defaults and just needing to know that there are issues with spaces in paths. Where I work, we actually put both UV itself, and the data, on the D: drive by default (we don't put the accounts under \ibm\uv, which is where they go by default aiui). I keep all data in a separate \Data path anyway, separating databases from the software that runs under them. I can understand if the U2 DBMS' don't like spaces in paths to data, but there's no excuse for such restrictions on the OS code and admin utilities. Those restrictions make cross-platform implementation a bit easier. Note that MS have reportedly stripped the spaces from default pathnames in Vista or Windows 7 (I can't say for certain - I have yet to have the pleasure of fighting either of those OS's :-) Cheers, Wol -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Change columns from 80 to 132 and from 132 to 80
In message [EMAIL PROTECTED], Ed Clark [EMAIL PROTECTED] writes Is this universe or unidata? Looks like universe defines the @(-29) function as setting 80 column mode, and the @(-30) function as setting 132 column mode. These may not be defined for all term types, but they work for me in vt420. I did exactly this for wIntegrate and PT250 emulation. What you need to do (or check it works) is multifold ... :-) Make sure those @ functions actually do something. Try setting VAR = @(-29) in a basic program and check the result isn't null. If they don't do anything you'll need to edit your terminfo database ... Then check your .wis file and make sure that, when it receives those sequences it actually makes wIntegrate change column size. Lastly, edit the TERM command (the source is in the UV account somewhere - APP.PROGS?) so that it responds to the WIDTH option. I wrote a couple of spool utilities - in particular one I called WIDELIST, that threw wIntegrate into 132 column mode to list a file and then reverted back to 80 when it had finished. It's a simple job but fiddly. I might be able to find some of the sources and post them. Cheers, Wol On Sep 12, 2008, at 3:49 PM, T Stokes wrote: I need to add this is from Wintegrate. T Stokes wrote: I am trying to find a way to change the columns on my screen from 80 132 and from 132 to 80. Either by scripts or by program. -- View this message in context: http://www.nabble.com/Change-columns- from-80-to-132-and-from-132-to-80-tp19462277p19462889.html Sent from the U2 - Users mailing list archive at Nabble.com. --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/ --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/ -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] SIGDANGER
In message [EMAIL PROTECTED], Michael Martin [EMAIL PROTECTED] writes Found this on IBM's website. http://publib.boulder.ibm.com/infocenter/systems/index.jsp?topic=/com.ib m.aix.baseadmn/doc/baseadmndita/page_space_trouble.htm Paging space troubleshooting The most common problem regarding paging space is caused by running out of allocated space. The total amount of paging space is often determined by trial and error. One commonly used guideline is to double the RAM size and use that figure as a paging space target. If paging space runs low, processes can be lost, and if paging space runs out, the system can panic. The following signal and error information can help you monitor and resolve or prevent paging space problems. I got involved in a little spat over these recommendations. I know I'm talking about Linux not AIX here, but ... Early in either the 2.4 or 2.6 series, Linus decided that the memory management code needed sorting out, so he deleted all the error handling / optimisation / what to do if swap is tight sort of code. What came out of this was that the standard Unix swap algorithm NEEDS twice ram of page space to function efficiently. So if people tell you that the twice ram rule is just a rule of thumb you can ignore, well it may or may not be. There certainly WERE very good reasons for it - whether those reasons are still valid I don't know, but certainly on any system still using the old traditional Unix algorithms it's still a very good rule you *should* follow - and as a *minimum*, not a *target*. Me - disk space is cheap so I always have swap as a minimum of twice my motherboard's maximum ram. Actually, on my current system (max ram = 768Mb) I've got a 10Gb swap, but I've also got /tmp mounted as a tmpfs file system which stuffs the whole lot in ram (or swap, as the case may be :-) Cheers, Wol -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] Editing Subsequent Value Marks
In message [EMAIL PROTECTED], Brutzman, Bill [EMAIL PROTECTED] writes I have what looks like ^253^253 that needs to be just ^253. Are you sure they're value marks, and not } or whatever they display as in your editor? Without luck I have tried... 1: C/^253^253/^253 and 1: C\^253^253\^253 Suggestions would be appreciated. Try ^ to put ED into display control character mode and check that they really are value marks... --Bill Cheers, Wol -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] RE: SELECT problem with quote characters
In message [EMAIL PROTECTED], Israel, John R. [EMAIL PROTECTED] writes You need the double quotes on the outside of the ... SELECT FILE WITH FIELD LIKE ...LLOYD'S... Alternatively, you can often get away with no quotes. Loosely speaking, you only NEED quotes on the outside if you have spaces in your search string, though if you have a single quote in your search string, that would likely confuse it. The problem is, this is a program constructing a select string. I have no control over what the user types in, and I want to do a literal match. What if they enter something like 6A'S? Basically, I need to quote the search string to get a literal match, which means I need to put quotes outside the dots, which means I need to use all three different quote characters, which then doesn't work :-( Cheers, Wol -- Anthony W. Youngman [EMAIL PROTECTED] 'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998 Visit the MaVerick web-site - http://www.maverick-dbms.org Open Source Pick --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
