INFO-VAX Sun, 15 Jul 2007 Volume 2007 : Issue 384 Contents: Re: EDT Replacement Re: HoffmanLabs: High-performance Interconnects, MPI? RE: HoffmanLabs: High-performance Interconnects, MPI? Internet2 Shibboleth on OpenVMS RE: Internet2 Shibboleth on OpenVMS Re: UCX Printer connection Via LPD ---------------------------------------------------------------------- Date: Sun, 15 Jul 2007 08:52:56 -0700 From: AEF Subject: Re: EDT Replacement Message-ID: <1184514776.483224.242350@d55g2000hsg.googlegroups.com> On Jul 14, 4:41 am, "P. Sture" wrote: > In article <46982E8D.38D51...@spam.comcast.net>, > David J Dachtera wrote: > > > "Michael D. Ober" wrote: > > > > Is there a free or low cost replacement for EDT on VMS 8.3? Preferably one > > > that can be configured to use the same 10-key keypad (Gold) keys as EDT. > > > I agree with those who maintain the EDT is as usable as it ever was. > > > I do wish TPU/EVE could process a "large" file without reading the entire > > thing into memory, though. > > > EDT mode of EVE is quite usable for me, though I use it only when EDT pukes > > on long records. > > While we are here, I just found an old message here of mine which claims > that EDT can't handle files with > 65K records. This could be the OP's > problem. Wait, there's more! I used EDT to create a new file containing the word LINE. I used EDT's repeat function to duplicate the line 32767 times and appended a line containing the number 1. Then I copied that block to the end of the buffer and changed the 1 to 2. Then repeated that process using 3 and 4. Then I used COPY and APPEND to create a final file containing over 1 million lines! This is a 1-megaline file! Now, let's see how EDT and EVE deal with it. $ SH SYM DIR D*IR == "DIRECTORY/SIZE=ALL/DATE /WIDTH=(FILE=29,SIZE=7)" $ DIR LINES.DAT Directory DISK$DATA1:[FELDMAN.EDT] LINES.DAT;4 12288/12288 15-JUL-2007 13:46:57.55 LINES.DAT;3 6144/6144 15-JUL-2007 13:46:43.84 LINES.DAT;2 3072/3072 15-JUL-2007 13:46:34.65 LINES.DAT;1 1536/1536 15-JUL-2007 13:45:34.25 Total of 4 files, 23040/23040 blocks. $ SEAR LINES.DAT 1,2,3,4/NUM 32768 1 65536 2 98304 3 131072 4 163840 1 196608 2 229376 3 262144 4 . . . 884736 3 917504 4 950272 1 983040 2 ****** 3 ****** 4 $ EDIT/EDT LINES.DAT 1 LINE *F E *SH BUF =MAIN 1048576 lines PASTE No lines *-3 1048574 LINE * 1048575 LINE * 1048576 4 * [EOB] *-1 1048576 4 *S-4-NEW4 1048576 NEW4 1 substitution *EX DISK$DATA1:[FELDMAN.EDT]LINES.DAT;5 1048576 lines It worked! EDT can handle a file with over 1 million records! Now let's try EVE: $ SH SYM EVE EVE == "EDIT/TPU" $ EVE LINES.DAT %SYSTEM-F-ACCVIO, access violation, reason mask=05, virtual address=7FEB17DC, PC=0005AC9F, PSL=03C00000 Improperly handled condition, image exit forced. Signal arguments Stack contents %TPU-F-ACCVIO, access violation, reason mask=00, virtual address=7FEC3A00, PC=833E7195, PSL=03C00004 -TPU-I-FROMLINE, called from line 357 -TPU-I-FROMLINE, called from line 310 -TPU-I-FROMLINE, called from line 2655 -TPU-I-FROMLINE, called from line 2329 -TPU-I-FROMLINE, called from line 2225 -TPU-I-FROMLINE, called from line 406 -TPU-I-FROMLINE, called from line 159 -TPU-I-FROMLINE, called from line 5719 -TPU-I-FROMLINE, called from line 28 -TPU-I-FROMLINE, called from line 4360 $ SHOW WORK Working Set /Limit=512 /Quota=1024 /Extent=20500 Adjustment enabled Authorized Quota=1024 Authorized Extent=20500 $ HAH! Take that, you EDT-bashers!!! :-D [...] > -- > Paul Sture AEF ------------------------------ Date: Sun, 15 Jul 2007 05:44:32 -0700 From: "Tom Linden" Subject: Re: HoffmanLabs: High-performance Interconnects, MPI? Message-ID: On Sat, 14 Jul 2007 21:52:31 -0700, Ron Johnson wrote: > On 07/14/07 21:18, Tom Linden wrote: > [snip] >> 4 times that of VAX. Abandoning VAX was a serious blunder. Had VAX >> been >> continued (and extended to 64 bit architecture) it could certainly >> clock at >> the same frequency as others, let's say 4 GHz (IBM will soon have 6GHz) >> and >> that would run the same as a 12GHZ Alpha! > > Didn't we have this same conversation last month? > Yes. -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Sun, 15 Jul 2007 14:20:53 +0000 From: "Main, Kerry" Subject: RE: HoffmanLabs: High-performance Interconnects, MPI? Message-ID: > -----Original Message----- > From: bill@triangle.cs.uofs.edu [mailto:bill@triangle.cs.uofs.edu] > On Behalf Of Bill Gunshannon > Sent: July 14, 2007 8:50 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: HoffmanLabs: High-performance Interconnects, MPI? > [snip...] > > > > At the US University I was at, by far the biggest dept was Arts > and Science=3D > > , followed by Engineering. Both of those had fair to good > ratings. CS was o=3D > > ne of the Depts that had one of the worst Operations environments > that we r=3D > > eviewed. When asked about their archiving strategy, they answered > "well, we=3D > > are working on improving that area, but right now xxx takes > tapes home eve=3D > > ry couple of weeks and brings the old ones back..." .. I almost > fell of my =3D > > chair. > > > > Their central ITS dept was the best with formal strategies, well > run Operat=3D > > ions environments. > > > > So, please don't tell me that University CS depts reflect the > future of com=3D > > puting. > > Well, you have pretty much proven my point that HP (assuming you > are > typical at HP and not atypical) just doe not understand the edu > world. > > CS department don't reflect, they drive. The next generation of > CIO's > and IT managers are going to go out with what they learned in their > CS education. If all they know is C, what will they have their > people > use? Today, it is Java. Tomorrow, who knows, but you can bet it > will > be whatever the next programming language du jour is. > They will use what ever the standard is within the company. Hotshot program= mers from CS or outsiders from other companies just do not come in an decid= e to change the company standard because that is what they learned in Colle= ge or University. Reality check - These days, changing language stds is a huge, huge expense = and this is at a time when BU's are "demanding" (a couple of years ago they= were "asking") that IT reduce their costs in big ways. [snip...] > > > > See my notes above on CS Depts. They also tend to change their > technical st=3D > > rategies with the latest trends. From what I understand, C and > C++ and C# w=3D > > ere being replaced with Java as the primary programming > languages. > > And where did the IT world go? Surprise!! > And it also benefited OpenVMS (and other enterprise OS's) because now Cust'= s can develop on their client of choice (Windows primarily, Linux another o= ption albeit used much less than Windows) and then deploy on rock solid ser= ver platforms that are not Windows/Linux. Reference:(Netbeans and OpenVMS) http://h71000.www7.hp.com/openvms/products/ips/netbeans/overview.html http://h71000.www7.hp.com/openvms/products/ips/netbeans/distnb.html > > > > So, all these Java graduates are potential OpenVMS programmers. > Simply copy=3D > > the Java programs to OpenVMS Java environment and run. > Stability, clusteri=3D > > ng and no monthly security patches to worry about. Nice picture. > > > >:-) > > Yeah, well here's another surprise. They are pushing Java cause > they > used in college. They are not using VMS in college, they are using > Windows and Linux. Hmmmm..... What direction is the industry > taking? > The Java directions has next to zero to do with CS Colleges. Java was perce= ived as an option to write once, run anywhere which is a powerful message (= at least in marketing speak). It was and is, being pushed by IBM, Sun, HP, = Oracle and all of the other major vendors as they perceive it to be a way t= o reduce their costs of maintaining applications running on different platf= orms. At least that was the theory. In short, major vendors and industry "analysts" are the primary drivers, al= ong with the massive push to reduce IT costs - the CS in Universities are s= imply jumping on the band wagon. [snip...] > > > > If I were in charge of IT, I would fire any internal programmer > that publis=3D > > hed details on the company internal computing environment on the > external w=3D > > eb site. Imho, that is a gross security violation. That includes > any platfo=3D > > rm details. > > > > Most secure environments do not publish much, if anything about > their inter=3D > > nal computing strategies. > > This isn't the Pentagon, it's a college. And only the academic > side of > that. We have to tell people what we run. They come to open- > houses and > ask. Can you imagine the reaction when we said "none of your > business"? > So you think Universities should have lower security stds than what most no= rmal businesses do? I would think your own University security people would disagree with you. Especially with all of the budding "crackers" that are prevalent in Univers= ities - yes, many of which are in CS. At the University I was at, I found t= heir security folks to be very good. They were very much in control of what= was happening. The FERPA, SOX, HIPPA regulatory world has hit them hard in= the last year or two. Along with other unfortunate incidents that were wel= l publicized. [snip...] > > > > Do you really understand the benefits of active-active clustering > with a na=3D > > tive common file system and batch sub-system that allows > programs, batch jo=3D > > bs and DCL routines to run against the same data at the same time > on any sy=3D > > stem with no third party prog's like expensive Oracle RAC in the > mix? Or th=3D > > e benefits of not having to worry about 5-20 security patches per > month? > > Jeezz Kerry, give it a rest. None of that has any relevance to the > discussion at hand, even if it were true. I don't have the budget > for > "active-active clustering with a native common file system and > batch > sub-system that allows programs, batch jobs and DCL routines to run > against the same data at the same time on any system with no third > party prog's like expensive Oracle RAC in the mix". And it wold > really > be overkill for my 14 students. They needed yhings like a > wordprocessor > and spreadsheet a bit more than "active-active clustering". And, I > don't know how to break it to you but I have never had to worry > about > your "5-20 security patches per month". But then, everybody keeps > telling you that, you just don't want to hear it. > Bill - lets leave that point alone. You deal primarily with client WS stuff= and perhaps a small number of servers likely less than 50) and I deal with= large to very large server DC environments. Two different worlds. [snip..] Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Sun, 15 Jul 2007 22:50:47 +0930 From: Mark Daniel Subject: Internet2 Shibboleth on OpenVMS Message-ID: <139k7q2j43os8e6@corp.supernews.com> Anyone tinkering with this? http://shibboleth.internet2.edu/ (And yes, I remember it took a decade to get Kerberos onto VMS :-) ------------------------------ Date: Sun, 15 Jul 2007 14:59:40 +0000 From: "Main, Kerry" Subject: RE: Internet2 Shibboleth on OpenVMS Message-ID: > -----Original Message----- > From: Mark Daniel [mailto:mark.daniel@vsm.com.au] > Sent: July 15, 2007 9:21 AM > To: Info-VAX@Mvb.Saic.Com > Subject: Internet2 Shibboleth on OpenVMS > > Anyone tinkering with this? > > http://shibboleth.internet2.edu/ > > (And yes, I remember it took a decade to get Kerberos onto VMS :-) Mark, As I suspect you know :-), there have been numerous attempts at providing m= ulti-vendor, multi-platform distributed security solutions over the years -= each with varying degrees of success. Web Service standards and offerings = continue to struggle big time with this as well. I could not agree more with the driver to do this. In fact, thx for the poi= nter as I will read up more on this. However, those involved in projects like this need to have some background = in order that they not repeat the many mistakes made in the past for these = types of distributed security initiatives. A review of why DCE and other in= itiatives like Microsoft's public program failed (I forget their code name)= are likely in order. Why did Kerberos not gain wider acceptance for exampl= e. There are also some very good commercial solutions with dedicated teams of = SW engineers working on them that would likely be considered as the competi= tion. (Yes, cheap is not a term for these solutions). At the US University I was at doing a DC and server optimization review, we= quit counting after about 10 in terms of the different user authentication= DB's in use at the University (HR out of syc with Banner out of sync with = Parking out of sync with facilities out of sync with individual Colleges us= ing different AD's like AFS, Microsoft AD etc). They were working to resolve it, but it was a real tough battle for them be= cause the real issues with distributed common solutions in general are not = technical, but rather data ownership issues. You get the same situation in most med to large companies today. This solution seems to take a federated approach, so it does sound promisin= g, but will need to review this some more. Bottom line is that distributed security solutions are just really, really = tough to do because security is people + process + technology and the techn= ology is the one that typically gets blamed when people or process issues f= ail. Many Cust's have totally given up on SSO (single sign on) as a theoretical = concept that will never really be reached in most environments. Hence, the = new buzz word often touted about is RSO (reduced sign-on) with a consolidat= ion of auth db's, but recognize there will continue to be some App's etc th= at will continue to use their own DB or repository for this. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Sun, 15 Jul 2007 17:00:33 -0000 From: ravindhar.uppada@gmail.com Subject: Re: UCX Printer connection Via LPD Message-ID: <1184518833.868848.11590@e16g2000pri.googlegroups.com> On Jul 2, 8:51 am, Chris wrote: > Paul Anderson wrote: > > In article > > <7dd80f60706200848y58bfcdefmfe656d2e474e7...@mail.gmail.com>, > > "Ken Robinson" wrote: > > >> You could try using DCPS. The latest version have the capability of > >> using LPD for printing. > > > Yeah, and we even send the right number of control and data files. > > > But I find it hard to believe that TCP/IP Services is doing the wrong > > thing. What version are you running? > > > Paul > > Found out this is an older version of OpenVMS and TCPIP. > VMS 7.2-1 > TCPIP 5.1 ECO4 > > Any one Know if thisDoubledfAin an LPR stream thing was specifically > addressed? or do you have a link to where I can research it? > > They are being *difficult* about considering updating... > > I appreciate your responses Paul, Printing is a bit of a lost art these > days. > > tia, Chris Hi, >>They are being *difficult* about considering updating... I would still recommend have them upgraded to the latest supported versions. >>Any one Know if this Double dfA in an LPR stream thing was specifically addressed? or do you have a link to where I can research it? Apparently this has been addressed very long back ago i.e prior to the TCPIP version 5.1. The LPD does handle the CF and DF in any order and with any number of DFs per CF. You may wanted to turn on all the LPR/LPD diagnostics for further information. That is, define system wide logical names TCPIP$LPD_DEBUG and TCPIP$LPD_RCV as 15. Please note that Leaving these diagnostics on during normal use might affect the performance of LPD and produce large log files. Regards Ravi OpenVMS TCP/IP Engineering Hewlett-Packard Company ------------------------------ End of INFO-VAX 2007.384 ************************