RE: [U2] Universe, 32bit and file size
Why are we talking about D3 in here - it's like talking about steam engines in the Ferrari garage ;) -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tony G Sent: 23 October 2008 23:17 To: u2-users@listserver.u2ug.org Subject: RE: [U2] Universe, 32bit and file size As I mentioned in CDP, this has nothing to do with D3 but it's always fun to blame D3 for whatever ails ya. David is using the %functions from BASIC which are wrappers around raw C functions. The limitations of these functions are that of the OS, not the DBMS wrappers. A quick Google for fopen 2gb will show tons of postings on this topic from non-MV developers. My posting to CDP also asked about the main reason for taking this technical approach so that someone can provide a better approach to solving the problem, but the last time I asked why here I lost a bit of my backside. I've also posted a note on David's behalf to the TigerLogic D3 Linux forum to ask if the %functions allow for a complete passthru to the OS. If so then he may be able to make use of 64bit (or Long) functions that are available to C developers but not in the current D3 documentation. Tony Gravagno Nebula Research and Development TG@ remove.pleaseNebula-RnD.com --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
RE: [U2] [UV] PHP
Hi The PDO driver looked a good start, but it may be too limited as it currently stands. It also depends on how you are hosting the PHP: most external provides won't let you load custom modules like this one. I did ask IBM last year whether they would be willing to publish the UniObjects protocol, so I could try to write a PHP UO library (PHP has built in socket support). They declined gracefully. But you can communicate using web services with the nuSOAP libraries. I haven't tried that with the IBM web services, but I've used that with mvScript (my web page/web service scripting tool) and it works fine so it should equally work with the IBM ones, or with a three tier solution e.g. exposing the web service through e.g. .Net. Regards Brian -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robert Porter Sent: 23 October 2008 21:48 To: u2-users@listserver.u2ug.org Subject: Re: [U2] [UV] PHP But ... a quick google... http://www.ibm.com/developerworks/db2/library/techarticle/dm-0 701xia/index.html U2 PDO Driver, Part 2: Write PHP applications to access U2 data Brutzman, Bill [EMAIL PROTECTED] 10/23/2008 3:03 PM Is there a way to do PHP with UniVerse? I just found out that there is a way to do Adobe Flex with PHP. Thus, perhaps it is possible to do Flex - PHP - UV. Suggestions would be appreciated. --Bill --- 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/ --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] Triggers on a NFS file?
Hi George, If I have a NFS mounted directory (on a third non-UV system) which houses a UV format file. I have (2) linux system that have that directory mounted, which each have UV that have a VOC pointer to that file. 1F , 2 /mnt/directory/filename , 3 /mnt/directory/D_filename Both UV systems can read/write to this file, but only one system will be writing to the file, the other will only be reading.I realize locking is not respected from one system to another. Be careful! I have seen (and repaired) files that have been damaged in broadly similar situation. As you say, the locking system will not work as the file is being accessed by two completely independent UV systems. The problem that you may have is not with record locking but with group locking that protects the internal structure of the file. Even with only one system writing, it is possible (likely?) that there will be times when the system doing the reading will access a group at the same moment that the other system is modifying an overflow chain. If I defined a trigger on this file on the system that will only be doing the reading, will it detect any changes to the file when the other UV makes changes? No. Triggers are fired somewhere deep in the UV file system. There is no way in which one system can know about changes applied by the other system. Even with uv/net to avoid the locking problems, the triggers are not going to get run on both systems for an update. Basically System1 will be making changes to that file. And System2 will need to update on it's system based on those changes. Sounds like you need to investigate data replication, changing the network structure if necessary. Martin Phillips Ladybridge Systems Ltd 17b Coldstream Lane, Hardingstone, Northampton, NN4 6DB +44-(0)1604-709200 --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
[U2] RE: Triggers on a NFS file?
Owww Firstly I doubt the trigger on one system will spot changes made by the other system. But, rather more seriously, we have had UV file corruption which was laid pretty conclusively down to two systems accessing the same dynamic file - oh and one of them, allegedly, only had read access! If you're doing this, I'd say the first rule is DON'T use dynamic files. Preferably use a type 1 file. And only use the file as a staging post. Even then, don't expect it to work all the time, every time. If you control the third system, look at firewalling the UV ports and using port forwarding - with a bit of effort you should be able to get the first two systems to network transparently. Cheers, Wol -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of George Gallen Sent: 23 October 2008 22:05 To: u2-users@listserver.u2ug.org Subject: [U2] Triggers on a NFS file? If I have a NFS mounted directory (on a third non-UV system) which houses a UV format file. I have (2) linux system that have that directory mounted, which each have UV that have a VOC pointer to that file. 1F , 2 /mnt/directory/filename , 3 /mnt/directory/D_filename Both UV systems can read/write to this file, but only one system will be writing to the file, the other will only be reading.I realize locking is not respected from one system to another. If I defined a trigger on this file on the system that will only be doing the reading, will it detect any changes to the file when the other UV makes changes? Basically System1 will be making changes to that file. And System2 will need to update on it's system based on those changes. The reason I'm using NFS is that the two UV systems can't interact easily because they are on different networks, but both have access to the system which is exporting the UV formatted file. George Gallen Senior Programmer/Analyst Accounting/Data Division [EMAIL PROTECTED] ph:856.848.9005 Ext 220 The Wyanoke Group http://www.wyanokegroup.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/
Re: [U2] RE: Triggers on a NFS file?
Hi Wol, But, rather more seriously, we have had UV file corruption which was laid pretty conclusively down to two systems accessing the same dynamic file - oh and one of them, allegedly, only had read access! This is an even worse case of what I brought out in my earlier reply. A dynamic file keeps information about its current modulus and load values (etc) in shared memory. Having this on two separate systems is a recipe for disaster. UV introduced the ALLOWNFS configuration option long ago with a default setting of 0 (no). When I deliver UV training, I take great care to emphasise that setting this to 1 (yes) also means it's all my fault if the file gets corrupted. Martin Phillips Ladybridge Systems Ltd 17b Coldstream Lane, Hardingstone, Northampton, NN4 6DB +44-(0)1604-709200 --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] PHP
On Fri, Oct 24, 2008 at 9:03 AM, Brutzman, Bill [EMAIL PROTECTED]wrote: Is there a way to do PHP with UniVerse? I just found out that there is a way to do Adobe Flex with PHP. Thus, perhaps it is possible to do Flex - PHP - UV. Here is the thread from a while back here Charles was using php/java bridge and uniobjects. http://www.nabble.com/Uniobjects-for-Java-td19002493.html I'm not sure if it is what you are after as you still have to do the java coding as well. Regards, Adrian --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] PHP
Bill, We are using PHP to access Unidata here. Universe should be similar. There was a fair amount of setup involved, but once it was up and running, it has worked great. There is some java coding, but it is minimal. Java and the Java-Bridge are needed in our case only because we are using Linux servers and only Uniobjects for Java is available. If you are on Windows, you may be able to call Uniobjects (COM version) or .NET directly from PHP, without Java/Java-Bridge. Unfortunately no one has developed a PDO for U2, at least as far as I know. There is an article out there about writing a PDO, but that requires socket programming which I wasn't anxious to get into. If you would like more detailed information, let me know. Hope this helps. By the way, Adrian's tip about closing the connection in the Finally clause was dead-on. I was leaving connections dangling. Thanks again, Adrian. Charles Shaffer Senior Analyst NTN-Bower Corporation Adrian Merrall [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 10/24/2008 03:57 AM Please respond to u2-users To: u2-users@listserver.u2ug.org cc: Subject:Re: [U2] [UV] PHP On Fri, Oct 24, 2008 at 9:03 AM, Brutzman, Bill [EMAIL PROTECTED]wrote: Is there a way to do PHP with UniVerse? I just found out that there is a way to do Adobe Flex with PHP. Thus, perhaps it is possible to do Flex - PHP - UV. Here is the thread from a while back here Charles was using php/java bridge and uniobjects. http://www.nabble.com/Uniobjects-for-Java-td19002493.html I'm not sure if it is what you are after as you still have to do the java coding as well. Regards, Adrian --- 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/
Re: [U2] [UV] PHP
quote who='Symeon Breen' date='Thursday 23 October 2008' I believe someone has a php library that uses the intercall libraries, it has been discussed on here, and there is actually a white paper on it on the ibm web site. I wrote a small driver for use in Perl using Intercall, but it can be used for PHP as well. -- Jeff Butera, Ph.D. Administrative Systems Hampshire College [EMAIL PROTECTED] 413-559-5556 Where I'm from, we believe all sorts of things that aren't true. We call it history. The Wizard, Wicked --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
RE: [U2] Triggers on a NFS file?
The NFS mount is actually set to ro (read only), so I wouldn't think (not to be read as it couldn't happen) the files would get corrupted. Stranger UV things have happened. The file itself is not on a UV system. What I did was a CREATE-FILE then copied the file to the shared directory. Then I did a DELETE-FILE so it was no longer in the original UV's systems linkages. Now the only reference is via a VOC remote pointer. I will have to play around with this (Warnings taken, not lightly). Thanks for the info. George -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Martin Phillips Sent: Friday, October 24, 2008 5:05 AM To: u2-users@listserver.u2ug.org Subject: Re: [U2] RE: Triggers on a NFS file? UV introduced the ALLOWNFS configuration option long ago with a default setting of 0 (no). When I deliver UV training, I take great care to emphasise that setting this to 1 (yes) also means it's all my fault if the file gets corrupted. Martin Phillips --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] PHP/Perl
quote who='[EMAIL PROTECTED]' date='Friday 24 October 2008' Would you be willing to share the details? Sure. Like many, I was looking for a means to get Perl/Php to connect to Unidata. After someone pointed me to the API documentation (Admin Supplement for Client APIs), I did some reading on the various APIs available. I chose to use Intercall as I had more experience with C/C++ than I did with Java. While I am comfortable in C/C++, I also knew that I did NOT want to place a lot of code that needed to be maintained in C/C++. I already have gobs of code in Unibasic that I want to leverage, and I also had plenty of code in Perl - I didn't want to have yet a third layer between them that also needed to be maintained as I added projects, etc. Hence, the C program was written once and I *never* touch it after it was debugged. To add functionality, I add to Unibasic code and Perl - that's it. In a nutshell, I wrote a small C program which leverages Intercall to make connections to the database. This is a lightweight program which requires little to no maintenance. In short, the C program is merely the connection between Perl and Unidata, passing commands from Perl to a single subroutine on the Unidata side. That subroutine, in turn, does all the work and then returns the data back to Perl. This isn't perfect. I don't deal with connection pooling so all requests from web forms make separate, direct connections to the database. With 1700+ end users (students, faculty, staff) I've never run into a license issue, but it's surely possible. Also, if I had more time and experience, I could've used perlxs to compile the Intercall code more tightly with Perl - but I don't. So currently the C program is stand alone which suffers from slight performance issues because each request from Perl is independently firing off the program, which requires some overhead. Again, with my usage here this hasn't proved to be a problem, but could be in larger implementations. But what I got in the end is exactly what I wanted: A realtime interface with Unidata that returns (possibly) multivalued data as that - no unnesting or flattening to ODBC-ify the data. If I'm using a MV database then I want raw MV data in my applications. I do all my web development with Perl. But I have shared this with another college who got it working under Php and even with my limited php skills, you can use the same mechanism as I'm using in Perl. For more info, Server: http://adsys.hampshire.edu Username: adsys_guest Password: foobar In the upper left corner perform a search under 'Documentation' for 'Intercall'. By the way, the site above is entirely driven using perl and my C program. -- Jeff Butera, Ph.D. Administrative Systems Hampshire College [EMAIL PROTECTED] 413-559-5556 Where I'm from, we believe all sorts of things that aren't true. We call it history. The Wizard, Wicked --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] PHP/Perl
Jeff, Charles, I basically used the same strategy when I wrote u2pipe, which is a C program that uses InterCall to connect with the U2 database. To add functionality, I add a Unibasic subroutine and a method in whatever web language -- right now ColdFusion and PHP, and looking to add ASP.NET soon. U2pipe relies on an external socket listener: xinetd for linux, inetd for AIX, and wininetd for Windows. The socket listener does not have to be on the same machine as U2 as long as it's reachable via the network. Using config files, multiple connections can be made (one socket port mapped to one U2 connection). I'm working on a connection pooling version, but it's not ready for prime time. The web server doesn't need to be on the same network. All of our deployments use a remote web hosting solution. So the web server is hosted in one state, while the actual U2 database is in another. The code and documentation is on www.PickWiki.com. Jeff Butera wrote: In a nutshell, I wrote a small C program which leverages Intercall to make connections to the database. This is a lightweight program which requires little to no maintenance. In short, the C program is merely the connection between Perl and Unidata, passing commands from Perl to a single subroutine on the Unidata side. That subroutine, in turn, does all the work and then returns the data back to Perl. --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re: [U2] [UV] PHP/Perl
I'll also add that I know Chaman University (CA) has a java based socket server that does implement connection pooling. If anyone wants a contact there, let me know. -- Jeff Butera, Ph.D. Administrative Systems Hampshire College [EMAIL PROTECTED] 413-559-5556 Where I'm from, we believe all sorts of things that aren't true. We call it history. The Wizard, Wicked --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
RE: [U2] [UV] PHP/Perl
I was playing with a perl program that would use named pipes to talk to a phantom, which would do the read/writing/selecting etc and pass it back to the perl program again through a name pipe. There would be a perl subroutines that would send/receive command to the local perl listener, and convert the data into hased arrays and varibles depending on what was used. I stopped working with UV at the time, but now I'm back,so I'll most likely finish that up. Basically, the phantom program would track the files and records in use and lock if needed (and of course unlock after a set amount of time went by and not unlocked). It would have been kludgy, but it would have gotten the job done. I was creating it so a website could access data directly from UV via a .cgi George -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Rex Gozar Sent: Friday, October 24, 2008 2:01 PM To: u2-users@listserver.u2ug.org Subject: Re: [U2] [UV] PHP/Perl Jeff, Charles, I basically used the same strategy when I wrote u2pipe, which is a C program that uses InterCall to connect with the U2 database. To add functionality, I add a Unibasic subroutine and a method in whatever web language -- right now ColdFusion and PHP, and looking to add ASP.NET soon. U2pipe relies on an external socket listener: xinetd for linux, inetd for AIX, and wininetd for Windows. The socket listener does not have to be on the same machine as U2 as long as it's reachable via the network. Using config files, multiple connections can be made (one socket port mapped to one U2 connection). I'm working on a connection pooling version, but it's not ready for prime time. The web server doesn't need to be on the same network. All of our deployments use a remote web hosting solution. So the web server is hosted in one state, while the actual U2 database is in another. The code and documentation is on www.PickWiki.com. Jeff Butera wrote: In a nutshell, I wrote a small C program which leverages Intercall to make connections to the database. This is a lightweight program which requires little to no maintenance. In short, the C program is merely the connection between Perl and Unidata, passing commands from Perl to a single subroutine on the Unidata side. That subroutine, in turn, does all the work and then returns the data back to Perl. --- 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/
RE: [U2] [UV] PHP/Perl
From: Jeff Butera I'll also add that I know Chaman University (CA) has a java based socket server that does implement connection pooling. If anyone wants a contact there, let me know. And Ian Renfrew wrote JScript for U2 which I think was ahead of its time. I don't believe it's maintained any longer but you can find some info here: http://www3.sympatico.ca/ian_renfrew/JS4U2/docframe.html This discussion prompted me to edit and post a blog entry I've been sitting on for a while: removeNebula-RnD.com/blog/tech/mv/2008/10/mvbindings1.html (please remove the text 'remove') Tony Gravagno Nebula Research and Development TG@ remove.pleaseNebula-RnD.com --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re[2]: [U2] Universe, 32bit and file size
Tony, Running D3 on a 64bit version of Linux does not allow reading. This is not because of the OS. The blame is where it belongs .. Does it really matter why i need the data? If we need to bring the data into the system then THAT IS what needs to be done. Of course I could always contract with someone to write a wrapper that could then be called from D3. Or i could just write it in c myself. You have never lost your back side from my posts unless you start selling your services instead of offering information. I know plenty of consultants .. during a small wind they fall out of the trees G If 'tiger logic' has a good solution then i hope to hear it. About a month ago we ran into a similar issue and in fact it caused a core dump .. imagine not protecting your code G. One of the contacts at the company asked about it and even sent the code .. never heard anything back. SO hope you fare better. Thanks Thursday, October 23, 2008, 3:17:13 PM, you wrote: TG As I mentioned in CDP, this has nothing to do with D3 but it's TG always fun to blame D3 for whatever ails ya. David is using the TG %functions from BASIC which are wrappers around raw C functions. TG The limitations of these functions are that of the OS, not the TG DBMS wrappers. A quick Google for fopen 2gb will show tons of TG postings on this topic from non-MV developers. My posting to CDP TG also asked about the main reason for taking this technical TG approach so that someone can provide a better approach to solving TG the problem, but the last time I asked why here I lost a bit of TG my backside. I've also posted a note on David's behalf to the TG TigerLogic D3 Linux forum to ask if the %functions allow for a TG complete passthru to the OS. If so then he may be able to make TG use of 64bit (or Long) functions that are available to C TG developers but not in the current D3 documentation. TG Tony Gravagno TG Nebula Research and Development TG TG@ remove.pleaseNebula-RnD.com From: Martin Phillips I'm not quite sure why I would want to read/write a sequential file over 2Gb but, given that this originated as a D3 comment and has found its way on to the U2 site, let me just add that QM can handle sequential files over 2Gb. The argument used by UV is that because integer values are stored as 32 bit numbers internally, file addresses over 2Gb cause a problem. A 64 bit version of UV still stores integer values as 32 bits because to do otherwise would potentially impact some applications (think about bitwise operations, for example). Also remember that at the operating system level, 64 bit file addressing does not require a 64 bit architecture, just use of long long int variables or, in some cases, a pair of 32 bit variables. TG --- TG u2-users mailing list TG u2-users@listserver.u2ug.org TG To unsubscribe please visit http://listserver.u2ug.org/ TG __ Information from ESET NOD32 Antivirus, version of TG virus signature database 3550 (20081023) __ TG The message was checked by ESET NOD32 Antivirus. TG http://www.eset.com -- DSig ` David Tod Sigafoos ( O O ) ___oOOo__( )__oOOo___ Academic chairs are many, but wise and noble teachers are few; lecture-rooms are numerous and large, but the number of young people who genuinely thirst after truth and justice is small. -- Albert Einstein. 1879-1955 --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re[2]: [U2] Universe, 32bit and file size
Robert, ding ding ding ding ... great answer Robert. Thanks Thursday, October 23, 2008, 11:46:41 AM, you wrote: RP Look at uconfig (but note the ONLY valid on 64-bit capable platforms comment) RP # 64BIT_FILES - This sets the default mode used to RP # create static hashed and dynamic files. RP # A value of 0 results in the creation of 32-bit RP # files. 32-bit files have a maximum file size of RP # 2 gigabytes. A value of 1 results in the creation RP # of 64-bit files (ONLY valid on 64-bit capable platforms). RP # The maximum file size for 64-bit RP # files is system dependent. The default behavior RP # may be overridden by keywords on certain commands. RP 64BIT_FILES 1 -- DSig ` David Tod Sigafoos ( O O ) ___oOOo__( )__oOOo___ A democracy will continue to exist up until the time that voters discover they can vote themselves generous gifts from the public treasury. -- Alexander Tyler --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re[2]: [U2] Universe, 32bit and file size
Yes .. trying to read in a os level file that is gt 2 gig. and yes %fopen fails Thursday, October 23, 2008, 12:10:56 PM, you wrote: SB I am not sure what you mean when you say ...D3 can only access files 2gig SB and smaller? SB Certainly, there is no 2gb limit on D3 files themselves (I have many SB instances of D3 files with modulos in excess of 1 million). Are you saying SB that you can't read native OS files from inside D3 when those file(s) are SB greater than 2gb? I would not suggest trying to read an entire large file SB into D3 this way. Have you tried the %open, %read functions to read and SB process the file in reasonably sized chunks, say 10K at a time? You can use SB %write to build large files this way as well. -- DSig ` David Tod Sigafoos ( O O ) ___oOOo__( )__oOOo___ When buying selling are controlled by legislation, the first thing to be bought sold are the legislators - P.J. O'Rourke --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
Re[2]: [U2] Universe, 32bit and file size
MP I'm not quite sure why I would want to read/write a sequential file over 2Gb MP but, given that this originated as a D3 comment and has found its way on to MP the U2 site, let me just add that QM can handle sequential files over 2Gb. I guess the reason is that I need the data from the file. It is good to hear that QM can handle a file of this size. MP Also remember that at the operating system level, 64 bit file addressing MP does not require a 64 bit architecture, just use of long long int variables MP or, in some cases, a pair of 32 bit variables. Working from the D3 environment I was just trying to use what tools they have. I am surprised that they don't have and are not talking about a 64bit version but that's cool if that is the direction they are taking. The reason to ask here is that I was wondering if IBM had an alternate path. --- u2-users mailing list u2-users@listserver.u2ug.org To unsubscribe please visit http://listserver.u2ug.org/
