INFO-VAX Thu, 09 Aug 2007 Volume 2007 : Issue 434 Contents: Re: Alpha/Integrity versions of VMS. Re: Alpha/Integrity versions of VMS. Barnes & Noble "The Minimum You Need to Know to Be an OpenVMS Application Develo Re: Easy DCL question PURGE vs. DELETE Re: F$FAO question Re: Maximum Java Heap Size/OpenVMS Java Experiences? Re: OpenVMS and Smart Array Controllers Re: OpenVMS and Smart Array Controllers Re: OpenVMS and Smart Array Controllers Re: OpenVMS and Smart Array Controllers Stillstand ist =?ISO-8859-15?Q?R=FCckschritt=2E?= Re: TPU on MAC OS-X ? Re: TPU on MAC OS-X ? Re: VMS cluster behind a *NIX firewall What VMS does RIGHT! Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion RE: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion Re: Wonderful things happen to an OS when it has an internal champion ---------------------------------------------------------------------- Date: Thu, 09 Aug 2007 03:59:37 -0700 From: "AlexNOSPAMDaniels@themail.co.uk" Subject: Re: Alpha/Integrity versions of VMS. Message-ID: <1186657177.559951.211380@d55g2000hsg.googlegroups.com> On 8 Aug, 02:21, "FredK" wrote: > New device support might actually find itself on Alpha releases - but not > supported. Consider theUSBsupport. We do NOTHING to prevent the code > from building and shipping on Alpha - but we do not QUALIFY it on Alpha - > unless there was a compelling CUSTOMER reason for formal support. The SPD lists USB as being supported on Marvel Alpha's. http://h18000.www1.hp.com/info/xav12x/xav12xpf.pdf "OpenVMS supports the Universal Serial Bus (USB) technology on the AlphaServer ES47, ES80, and GS1280 systems, and the rx1600, rx2600, and rx4640 Integrity Server systems. Support for the USB interconnect enables OpenVMS systems to connect to multiple supported USB devices using a single USB cable. OpenVMS supports the fully qualified USB devices listed in the appropriate AlphaServer platform configuration and options web site: http://h18002.www1.hp.com/alphaserver/ " Alex ------------------------------ Date: Thu, 9 Aug 2007 09:14:56 -0400 From: "FredK" Subject: Re: Alpha/Integrity versions of VMS. Message-ID: "AlexNOSPAMDaniels@themail.co.uk" wrote in message news:1186657177.559951.211380@d55g2000hsg.googlegroups.com... > On 8 Aug, 02:21, "FredK" wrote: >> New device support might actually find itself on Alpha releases - but not >> supported. Consider theUSBsupport. We do NOTHING to prevent the code >> from building and shipping on Alpha - but we do not QUALIFY it on Alpha - >> unless there was a compelling CUSTOMER reason for formal support. > > The SPD lists USB as being supported on Marvel Alpha's. > > http://h18000.www1.hp.com/info/xav12x/xav12xpf.pdf > > "OpenVMS supports the Universal Serial Bus (USB) > technology on the AlphaServer ES47, ES80, and > GS1280 systems, and the rx1600, rx2600, and rx4640 > Integrity Server systems. Support for the USB interconnect > enables OpenVMS systems to connect to multiple > supported USB devices using a single USB cable. > OpenVMS supports the fully qualified USB devices > listed in the appropriate AlphaServer platform configuration > and options web site: > http://h18002.www1.hp.com/alphaserver/ " > > Yes, EV7 supports USB. The "fully qualified USB devices" on the EV7 platforms are essentially the KB and the Mouse. On Integrity, for example, we support a variety of serial line muxes - we do not qualify or support them on EV7 - nonetheless, the drivers for them are on Alpha - but not as "fully qualified USB devices". ------------------------------ Date: Thu, 09 Aug 2007 09:26:14 -0700 From: yyyc186 Subject: Barnes & Noble "The Minimum You Need to Know to Be an OpenVMS Application Develo Message-ID: <1186676774.915763.285090@j4g2000prf.googlegroups.com> It seems Barnes & Noble is having a 20% computer book sale that somewhat coincides with the prep for OpenVMS at 30. They've really cut the price on this book for the duration of the sale. ------------------------------ Date: Thu, 9 Aug 2007 08:58:09 -0400 From: norm.raphael@metso.com Subject: Re: Easy DCL question PURGE vs. DELETE Message-ID: "P. Sture" wrote on 08/09/2007 01:18:17 AM: > In article > , > norm.raphael@metso.com wrote: > > > > > > > > > > > On fait ce q'on peut. > > > > > > > > > More accurately: "One does what one can." Quote from a French Officer who > > boarded the Patna after she drifted and who stayed on board some 36 hours. > > (The Captain was German.) Those "little gray cells" sometimes distort > > recall. > > > > > Talking of "little grey cells", mine would have written "qu'on" rather > than "q'on". > > A Google search produces both forms, so which is the more correct > French, or are both acceptable? Thanks, Paul. Actually, that was a typo on my part. I looked it up and Conrad did write "(On fait ce qu'on peut.)" I had no idea the other form was in use. I guess we are done for now with the real topic of this thread as I must confess that I hijacked it myself onto this tangent. Sorry. > > -- > Paul Sture > > Sue's OpenVMS bookmarks: > http://eisner.encompasserve.org/~sture/ovms-bookmarks.html ------------------------------ Date: 9 Aug 2007 10:36:29 -0500 From: cornelius@encompasserve.org (George Cornelius) Subject: Re: F$FAO question Message-ID: In article , VAXman- @SendSpamHere.ORG writes: > > You can right justify an ascii format but it requires a little more > than just the F$fao. Without following your link, this admittedly rather ugly approach comes to mind: $ write sys$output f$fao("!''f$int(12-f$len("Stuff"))'* !AS","Stuff") Longer, but cleaner: $ s="Stuff" $ n=12-f$len(s) $ if n.lt.0 then n=0 ! Negative n is a no-no $ write sys$output f$fao("!''n'* !AS",s) > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM -- George Cornelius cornelius(at)eisner.decus.org cornelius(at)mayo.edu ------------------------------ Date: Thu, 09 Aug 2007 13:18:22 +0200 From: "P. Sture" Subject: Re: Maximum Java Heap Size/OpenVMS Java Experiences? Message-ID: In article , John Santos wrote: > P. Sture wrote: > > In article , John Santos > > wrote: > > > > > >>P. Sture wrote: > >> > >>>In article <1185990365.187329.66340@i38g2000prf.googlegroups.com>, > >>> sean@obanion.us wrote: > >>> > >>> > >>> > >>>>Out of sheer curiosity and thinking it would be easy, I tried to find > >>>>in the on-line documentation what the current maximum WSMAX value is, > >>>>since it looks like that's the next limit. > >>> > >>> > >>>If you do: > >>> > >>>$ MCR SYSGEN SHOW WSMAX > >>> > >>>you get something like this (taken on V8.3, Alpha) > >>> > >>> > >>>Parameter Name Current Default Min. Max. Unit Dynamic > >>>-------------- ------- ------- ------- ------- ---- ------- > >>>WSMAX 393216 8192 1024 134217728 Pagelets > >>> internal value 24576 512 64 8388608 Pages > >>> > >>>$ write sys$output 134217728/2048 ! max in megabytes > >>>65536 > >>> > >>>i.e. 64 GB, for Alpha running V8.3. You should repeat this on Itanium > >>>to check that SYSGEN uses the same maximum value there. > >>> > >> > >>On I64 V8.3: > >> > >>$ mcr sysgen show wsmax > >>Parameter Name Current Default Min. Max. Unit > >>Dynamic > >>-------------- ------- ------- ------- ------- ---- > >>------- > >>WSMAX 1089536 131072 16384 134217728 Pagelets > >> internal value 68096 8192 1024 8388608 Pages > >> > >>Same maximum, but much higher default and min values. "Current" value came > >>from AUTOGEN with feedback, AFAIK. > >> > > > > > > Thanks for that. Ditto here for the "Current" vaalue. Surprisingly, I > > don't see anything (on Alpha) about WSMAX in > > SYS$SYSTEM:AGEN$PARAMS.REPORT, though AUTOGEN is obviously calculating > > it. > > > > The same on Itanium. I didn't set it explicitly, and there's nothing > in MODPARAMS.DAT or in AGEN$PARAMS.REPORT. My rx2620 came from the > porting workshop, and was pre-loaded with V8.2-1 (since upgraded to V8.3). > Maybe HP set it, or maybe AUTOGEN set it on the initial install and > the report is long gone. A look at AUTOGEN shows that it does check to see whether WSMAX is too large or too small, and there's a lot of code there; parameters affecting fluid memory are taken into account to ensure there is enough of that. Looking back at my upgrade to V8.3, WSMAX wasn't in MODPARAMS.DAT generated by the upgrade procedure, nor the accompanying AGEN$PARAMS.REPORT (or any since). WSMAX isn't included in AGEN$HISTORY.DAT either, so I conclude that it is calculated by AUTOGEN each time it is run. -- Paul Sture Sue's OpenVMS bookmarks: http://eisner.encompasserve.org/~sture/ovms-bookmarks.html ------------------------------ Date: Thu, 09 Aug 2007 09:12:11 -0400 From: Robert Deininger Subject: Re: OpenVMS and Smart Array Controllers Message-ID: In article , dittman@dittman.net wrote: > In any case, as it's not supported I'd rather use the much cheaper > LSISAS1068 controller instead. That's fine, but don't expect the same performance you'd get from a SmartArray controller. The cheap card has is basically a SAS controller with limited RAID thrown in as an afterthought. > > Last I heard, in informal use, HP SATA drives work just as well as HP > > SAS drives with these controllers. To maximize the likelihood of > > success, make sure you have: > > 1. VMS V8.3 with the latest patches. > > 2. The latest FW and EFI driver for your controller. > > 3. The latest FW for your disks. > > I have loaded the latest of everything. I suspect that the actual > revision of the LSISAS1068 might be the issue. The one I have (a > SAS3080X-HP) has an A1 rev. The latest firmware for the LSISAS1068 > only supports the B0 and B1 revs. I'm going to keep my eye out for > a controller with a newer rev. to test. The FW should be released in parallel for the A and B revs. I think it is in separate packages. And the A rev of the chip IS supported. > Is the controller on the rx2660 and rx3600 integrated or on a > separate board? Integrated on the rx2660. On rx3600 and rx6600, the card goes in one of the reserved core PCI slots. ------------------------------ Date: Thu, 09 Aug 2007 13:34:58 GMT From: dittman@dittman.net Subject: Re: OpenVMS and Smart Array Controllers Message-ID: <6KEui.4156$Aj6.540@trnddc01> Robert Deininger wrote: > In article , dittman@dittman.net wrote: > > In any case, as it's not supported I'd rather use the much cheaper > > LSISAS1068 controller instead. > That's fine, but don't expect the same performance you'd get from a > SmartArray controller. The cheap card has is basically a SAS controller > with limited RAID thrown in as an afterthought. I haven't tested the throughput of the P600 but the LSISAS1068 controller had about 75% of the performance of the built-in U320 controllers with the HP 73GB 15K RPM U320 drives in my quick BACKUP test. I'll test the P600 to compare (and run some better tests on both). Fortunately the speed is adequate for the purposes. I'm using this for storing stuff I don't need fast access to but don't want to have to pull off tape or CD/DVD. > > > Last I heard, in informal use, HP SATA drives work just as well as HP > > > SAS drives with these controllers. To maximize the likelihood of > > > success, make sure you have: > > > 1. VMS V8.3 with the latest patches. > > > 2. The latest FW and EFI driver for your controller. > > > 3. The latest FW for your disks. > > > > I have loaded the latest of everything. I suspect that the actual > > revision of the LSISAS1068 might be the issue. The one I have (a > > SAS3080X-HP) has an A1 rev. The latest firmware for the LSISAS1068 > > only supports the B0 and B1 revs. I'm going to keep my eye out for > > a controller with a newer rev. to test. > The FW should be released in parallel for the A and B revs. I think it > is in separate packages. And the A rev of the chip IS supported. I was talking about the firmware released by LSI for that specific controller. They only have firmware for the B0 and B1 revs. in the latest firmware. I'll try switching to the firmware used for the HP controller (if it's available separately). Another thing I'll try is getting a small SAS drive just to see if the problem is SAS drives vs. SATA drives. > > Is the controller on the rx2660 and rx3600 integrated or on a > > separate board? > Integrated on the rx2660. On rx3600 and rx6600, the card goes in one of > the reserved core PCI slots. Is that card available separately? -- Eric Dittman dittman@dittman.net ------------------------------ Date: Thu, 9 Aug 2007 09:52:39 -0400 From: "David Turner, Island Computers" Subject: Re: OpenVMS and Smart Array Controllers Message-ID: <13bm70bjtlr6b71@news.supernews.com> Oopth ! wrote in message news:uqpui.1999$Aj6.1475@trnddc01... > David Turner, Island Computers wrote: >> Well, eric, I have 20+ in stock > > That was Steve that was having trouble finding them. > -- > Eric Dittman > dittman@dittman.net ------------------------------ Date: Thu, 9 Aug 2007 17:44:03 +0000 (UTC) From: Rick Jones Subject: Re: OpenVMS and Smart Array Controllers Message-ID: > > Integrated on the rx2660. On rx3600 and rx6600, the card goes in > > one of the reserved core PCI slots. > Is that card available separately? Armed with the part number (which I do not know offhand) you could try partsurfer.hp.com. rick jones -- web2.0 n, the dot.com reunion tour... these opinions are mine, all mine; HP might not want them anyway... :) feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH... ------------------------------ Date: Thu, 09 Aug 2007 11:05:05 +0200 From: Martin Krischik Subject: Stillstand ist =?ISO-8859-15?Q?R=FCckschritt=2E?= Message-ID: <46bad8c2$1@news.post.ch> P. Sture schrieb: > In article <1615375.22NOzJ5N3m@linux1.krischik.com>, > Martin Krischik wrote: > >> Bob Koehler wrote: >> >>> As many products in many industries have proven, being stable is not >>> the same as being dead. >> Depends on your definition on dead and stable. Mine is defined around the >> German proverb "Stillstand ist Rückschritt" - which translates to "To stand >> still is to go back" (because everybody else around you move forward). >> >> Which in turn means that to keep stable you have to move forward at a >> moderate rate and if you don't move forward at all you are - or soon will >> be - dead. >> > > Software is usually a compromise. Getting it right can be knowing when > to stop. I think you missed the point of "Stillstand ist Rückschritt". You can't stop because everybody else moves. Or to put it different: There are two ways of stopping: 1) Absolute - then the majority of the industry moves passed you and you soon are left behind. 2) Relative - You move forward at the average rate of the rest of the industry. And you can look at it any way you like: If you choose option 1 then you will become obsolete - depending on which part of the industry it will happen in 5, 10 or 20 years. But it will happen. As for the topic: Our two newest team members have chosen Vim over Eve or LSEdit. Martin -- mailto://krischik@users.sourceforge.net Ada programming at: http://ada.krischik.com ------------------------------ Date: Thu, 09 Aug 2007 11:10:29 +0200 From: Martin Krischik Subject: Re: TPU on MAC OS-X ? Message-ID: <46bada06@news.post.ch> Bob Koehler schrieb: > If you don't have the TPU code to do your favorite editor trick, > that's not TPU's fault. Then do tell me how to implement syntax highlight in an editor language without colour support. Even syntax folding which is supported by lsedit is not at all that useful without a 2nd colour to highlight the folded lines. Martin -- mailto://krischik@users.sourceforge.net Ada programming at: http://ada.krischik.com ------------------------------ Date: 9 Aug 2007 07:34:08 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: TPU on MAC OS-X ? Message-ID: In article <46bada06@news.post.ch>, Martin Krischik writes: > Bob Koehler schrieb: > >> If you don't have the TPU code to do your favorite editor trick, >> that's not TPU's fault. > > Then do tell me how to implement syntax highlight in an editor language > without colour support. DON'T! I hate it when editors change the colors to illegible. ------------------------------ Date: Thu, 09 Aug 2007 11:37:14 -0000 From: Thomas Dickey Subject: Re: VMS cluster behind a *NIX firewall Message-ID: <13blv3acm3eu320@corp.supernews.com> David J Dachtera wrote: > Thomas Dickey wrote: >> >> David J Dachtera wrote: >> > You'll want to look into SMG. See the following... >> >> >> but all of them are complex >> >> > Not much worse than "termcap" and "curses", really... >> >> ...but certainly far more limiting. If the terminal supports features >> not explicitly in one of DEC's terminals, SMG doesn't support it. > D'y'ever look at SYS$SYSTEM:SMGTERMS.TXT? certainly (I mentioned that a few years ago). > Need support for a terminal not in there? ditto - it can't do what I want. (google is your friend) -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net ------------------------------ Date: Thu, 09 Aug 2007 16:43:04 +0000 From: "Paul Raulerson" Subject: What VMS does RIGHT! Message-ID: In a recent conversation, I keep feeling like I am forced to denigrate VM= S, which really isn't what I meant at all. I'm just rather "cold blooded"= when I am evaluating products or software for one fit or another. I tend= to use what I judge to be *best* for any particular need. That judgement= can get complex sometimes. :) So here is what, with limited experience, I just VMS is BEST, or at least= VERY good at. (1) Resource Utilization: VMS seems to sip resources. It runs acceptable= well on a emulator, and very well on a modest Alpha server. Seems to run= great on an Itanium machine. (2) Langauges: I;m a COBOL bigot, and I >>like<< the COBOL compiler. It h= as a few warts (Y'all look into cutting down the number of error messages= generated because of a single typo please! :) but the language support is superb. I= t is easily used for even odd things (to COBOL) like CGI processing. To a= lesser degree, I like the Pascal and Fotran compilers as well. The Pasca= l compiler actually makes Pascal a useful langauage! The Fortran compiler= seems tight too. Better than Fortran on an ancient Perkin/Elmer! :) (3) Clustering: This is the best non-hardware based non-custom-vrsion-of-= the-OS clustering technology I have seen. Best overall that is, in terms = of setup, capability, and managability. -Paul ------------------------------ Date: Thu, 09 Aug 2007 15:20:43 +0000 From: "Paul Raulerson" Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: ----=_vm_0011_W7201327024_11552_1186672843 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable In article <005601c7da23$bb530840$31f918c0$@com>, "Paul Raulerson" > writes: >> >> To answer just this little snippet, UNIX security is not bad, and is >= roughly >> on par with VMS. > >ROTFLOL. You think that VMS has high security? Well, it does, but in = part, that is security through obscurity. Does it have kernel based firewall capabilities? What about third party m= onitoring tools that meet SOX requirements? How about little things like = really erasing the data on a DASD unit? Or how about published ways on VM= S to handle packet spoofing, tcp sequencing, etc? The two systems are roughly equivalent; speaking of a hardended UNIX site= and a hardened VMS site. Each system does some things better than the ot= her. None of them approaches the security level of say, z/OS secured with RACF= That's built in security that goes down to the level of THIS user can see= THIS field ONLY when logged into THIS terminal and authenticated THIS wa= y, with full reporting and so forth. And it carries through the entire sy= stem. Now if you want to see something that is really a joke- look at "Windows = Security." If nothing else, most places leave physcial access open to th= e users. -Paul ----=_vm_0011_W7201327024_11552_1186672843-- ------------------------------ Date: Thu, 09 Aug 2007 16:28:28 +0000 From: "Paul Raulerson" Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: ----=_vm_0011_W7617223970_9212_1186676908 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable > >To this day IBM does not have a distributed lock manager nor a >distributed transaction manager. Both of which are required to >perform fault tolerant transactions. UH- CICS, now known as "Transaction Server" has been around near on 40 ye= ars now and supportd distributed confgurations as well as isloated config= urations. A very large percentage of anything like real time financial tr= ansacitons in the world goes through CICS. The HP ADCM product is somewha= t similar. This is something that VMS does well, don't take me wrong, but how does i= t do it differently or better than IBM? I see VMS has fitting in much bet= ter in smaller, "batch" distributed systems, with many smaller nodes size= d appropriately to the location. You can do that with IBM gear of course,= but it scales differently. Or to say that differently, VMS scales like a micro z/OS into many situat= ions, with appropriate security and processing power. z/OS just doesn't s= cale *down* all that well, because it takes significant skill just to get= it installed, much less operating with tuned specs. VMS is much easier i= ndeed. A BIG selling point in the SMB market place. Even though there are some very large VMS sites out there (I'm impressed)= I'm not sure that VMS is the target of choice for VL environments. I am = willing to be convinced otherwise. :) >No version of Un*x has these >capabilities either. No version of Un*x actually clusters. This is a >fact Oracle is finding out in court over their RAC10 product now. It >will be a damned big check they write indeed! That was a very public >outage with measurable financial loss. Well- CICS runs fine under Unix, AIX in particular. I think it is Charles= Swchab (?) that is running all the operations on a fleet of AIX machines= running CICS. (Well, Transaction Server, but IBM renames things just to = shake up the market... :) Unix clustering is primitive compared to VMS; VMS is very good there, esp= ecially scaling down to little bitty machines! On the other hand, look at the latest supercomputers built on Apple xServ= ers. It would be difficult indeed to say they are not completely clustere= d. However, distributed processing and "clustering" are not the same thin= g. Unix works fine as a distributed processing load system. -Paul ----=_vm_0011_W7617223970_9212_1186676908-- ------------------------------ Date: Thu, 09 Aug 2007 16:32:24 +0000 From: "Paul Raulerson" Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: ----=_vm_0011_W76983619_11016_1186677144 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable > > None of them approaches the security level of say, z/OS secured with RA= CF. > That's built in security that goes down to the level of THIS user can s= ee >THIS field ONLY when logged into THIS terminal and authenticated THIS= way, >with full reporting and so forth. And it carries through the entir= e system. >ACLs used correctly and applied in detail to RDB databases provide >exactly that level of security. It was the first OS and relational >database to provide such capability and it can do it across an entire >globe spanning cluster, not a single "moat and fortress" site. I have never seen "ACLs" - meaning access control lists tied to a filesys= tem able to do what RACF does. I do not think we are talking about the same t= hing here. BTW: DB2 has had federated databases for more than a decade, well before = Oracle. RDB is file based, I'm not sure how it can do that, especially ov= er slower global WAN connections. -Paul ----=_vm_0011_W76983619_11016_1186677144-- ------------------------------ Date: 9 Aug 2007 12:27:26 -0500 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: In article , "Paul Raulerson" writes: > In article <005601c7da23$bb530840$31f918c0$@com>, "Paul Raulerson" > @raulersons.com> writes: >>> >>> To answer just this little snippet, UNIX security is not bad, and is >= > roughly >>> on par with VMS. >> >>ROTFLOL. > > You think that VMS has high security? Well, it does, but in = > part, that is security through obscurity. Do you mean that since the source listings are only on CDROM these days, VMS security is obscured from those who have only a microfiche reader and no CDROM-capable computer ? > Does it have kernel based firewall capabilities? Your use of the term "kernel based" implies that you are "not from around here". Absolutely the VMS kernel firewalls all communication protocols implemented in the the VMS kernel. That would be SCS, right ? Can you think of any others ? > What about third party monitoring tools that meet SOX requirements? If you had been paying attention on DECUServe, you would have read the testimony that SOX compliance means "whatever this year's SOX auditor thinks it means." If you want to consider technical compliance with a standard more rigid that Jello, try NIST 800-53. My company would be happy to provide software ready-to-go for monitoring it. http://www.LJK.com/ljk/800-53.html If you want, you can tune that software to match the technical characteristics _you_ think are mandated by SOX. > How about little things like really erasing the data on a DASD unit? What constitutes _really_ erasing is up to your own DAA, which is why VMS lets you provide your own erase pattern. Of course that capability has only been around for the past 20 years or so, and if you are still running VMS V3.0 on your VAX 11-780, you do not have it. > Or how about published ways on VMS to handle packet spoofing, > tcp sequencing, etc? That depends on which TCP/IP stack you have installed on your machine. If you have complaints about the documentation of that stack, speak to your TCP/IP vendor. > None of them approaches the security level of say, z/OS secured with RACF= (Note to readers who do not follow IBM operating systems - RACF is an add-on product. The security is added on to the operating system.) > That's built in security that goes down to the level of THIS user can see= > THIS field ONLY when logged into THIS terminal and authenticated THIS wa= > y, with full reporting and so forth. And it carries through the entire sy= > stem. Don't those "fields" come from an add-on product as well ? I gather you have not looked at Rdb security on VMS. ------------------------------ Date: 9 Aug 2007 12:37:12 -0500 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: In article , "Paul Raulerson" writes: > ----=_vm_0011_W76983619_11016_1186677144 > Content-Type: text/plain; charset="utf-8" > Content-Transfer-Encoding: quoted-printable Please stop doing that. > I have never seen "ACLs" - meaning access control lists tied to a filesys= > tem > able to do what RACF does. I do not think we are talking about the same t= > hing here. Certainly not. Rdb is a database management system, not a file system. > BTW: DB2 has had federated databases for more than a decade, well before = > Oracle. Your term "federated database" is not clear, but when you say "well before Oracle" you must be talking about "Classic" Oracle, which nobody here has proposed as an example of database security. > RDB is file based, I'm not sure how it can do that, especially over > slower global WAN connections. Rdb is a relational database that had the security features we are discussing well before it was purchased from Digital by Oracle, and that was quite a bit more than 10 years ago. Are you perhaps confusing the acronym "Rdb" with the acronym "RMS" ? Generally participants here are reasonably careful about the distinctions between the two. ------------------------------ Date: Thu, 09 Aug 2007 09:03:37 +0200 From: Michael Kraemer Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: David J Dachtera schrieb: > Entry into those markets currently closed by VMS's foothold? Such as ? Cerner ? > Now - to flip that coin over: > > What could IBM bring to VMS that it currently lacks? > > - A return to virtualization (LPARS) > - A return to marketing > - A return to profitability > - A return to a respectable market share They already have all that with e.g. AIX. ------------------------------ Date: 9 Aug 2007 07:35:11 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: RE: Wonderful things happen to an OS when it has an internal champion Message-ID: In article <005601c7da23$bb530840$31f918c0$@com>, "Paul Raulerson" writes: > > To answer just this little snippet, UNIX security is not bad, and is roughly > on par with VMS. ROTFLOL. ------------------------------ Date: Thu, 09 Aug 2007 08:29:06 -0600 From: Keith Parris Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: David J Dachtera wrote: > What could IBM bring to VMS that it currently lacks? ... > - A return to profitability Profitability has never been a problem for OpenVMS. So IBM couldn't achieve "a return to profitability" for OpenVMS. ------------------------------ Date: Thu, 09 Aug 2007 08:36:15 -0700 From: yyyc186 Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: <1186673775.817381.100430@i13g2000prf.googlegroups.com> > To answer just this little snippet, UNIX security is not bad, and is roughly > on par with VMS. OMG! What have you been drinking/smoking/imbibing? Security is a complete joke on all flavors of Un*x including UX. > > In regard to the IBM question: IBM would not be interested in VMS because, > to be brutally honest, > VMS does not have anything IBM already does not have. And IBM has a lot that > VMS does not have. A > powerful lot indeed. If it ran on zSeries hardware they would snap it up > though, because that is > one more powerful mainframe operating system they could run. To this day IBM does not have a distributed lock manager nor a distributed transaction manager. Both of which are required to perform fault tolerant transactions. No version of Un*x has these capabilities either. No version of Un*x actually clusters. This is a fact Oracle is finding out in court over their RAC10 product now. It will be a damned big check they write indeed! That was a very public outage with measurable financial loss. ------------------------------ Date: Thu, 09 Aug 2007 08:40:12 -0700 From: yyyc186 Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: <1186674012.435195.108880@x40g2000prg.googlegroups.com> > > None of them approaches the security level of say, z/OS secured with RACF. > That's built in security that goes down to the level of THIS user can see THIS field ONLY when logged into THIS terminal and authenticated THIS way, with full reporting and so forth. And it carries through the entire system. ACLs used correctly and applied in detail to RDB databases provide exactly that level of security. It was the first OS and relational database to provide such capability and it can do it across an entire globe spanning cluster, not a single "moat and fortress" site. ------------------------------ Date: 9 Aug 2007 16:05:03 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: <5i0s9fF3mmkmfU1@mid.individual.net> In article <1186673775.817381.100430@i13g2000prf.googlegroups.com>, yyyc186 writes: >> To answer just this little snippet, UNIX security is not bad, and is roughly >> on par with VMS. > > OMG! What have you been drinking/smoking/imbibing? Security is a > complete joke on all flavors of Un*x including UX. Keep telling yourself that. It gives the industry a real good laugh everytime they hear it. Meanwhile, your legacy OS is slowly dying while Unix (who's death was announced more than a decade ago right right there on the cover of Byte Magazine) continues to grow market share. > > >> >> In regard to the IBM question: IBM would not be interested in VMS because, >> to be brutally honest, >> VMS does not have anything IBM already does not have. And IBM has a lot that >> VMS does not have. A >> powerful lot indeed. If it ran on zSeries hardware they would snap it up >> though, because that is >> one more powerful mainframe operating system they could run. > > To this day IBM does not have a distributed lock manager nor a ::>The OpenDLM Project ::> ::>About OpenDLM ::> ::> ::>OpenDLM is a distributed lock manager. It has interfaces for use at ::> both user and kernel level. ::> ::> ::>We are mostly members of OpenGFS and the Open Clustering Framework ::>who are working on a locking mechanism that will be suitable for ::>all types of clustering. ::> ::> ::>We started from a very good starting point with IBM's DLM, and are ::>extending it to fit the needs of the various different types of ::>clusters. One eventual goal is to be a reference implementation ::>for the OCF locking API (once it is standardized). Another is to ::>be used for inter-node lock support for OpenGFS, and perhaps other ::>clustered filesystems. ::> > distributed transaction manager. Both of which are required to > perform fault tolerant transactions. Probably hasn't been a need up to this point, but once some Unix vendor (IBM?) decideds it is needed they will write it. Unlike some other OSes we know, Unix is not stagnating. It continues to move forward. (Look at the release docs for the next step in FreeBSD.) > No version of Un*x has these > capabilities either. See above. I know it runs on Linux. I would bet it works on pretty much any modern Unix. (at least anything newer than Ultrix-11.) > No version of Un*x actually clusters. Probably depends on your definition of cluster. But let's assume the VMS definition. This has been asked before but never answered. Forgetting Hobbyist systems where they cluster for the sake of clustering, what percentage of real, production VMS systems are clusters as opposed to stand alone systems? Why would Unix bother implementing something that has little if any need? When a Beowulf style cluster became needed, it was written (and actually, that's not the only kind available on Unix.) And then we also had Amoeba Clusters, which, if nothing else, showed how little interest there really was in clustering beyond looking at it from an academic standpoint. > This is a > fact Oracle is finding out in court over their RAC10 product now. It > will be a damned big check they write indeed! That was a very public > outage with measurable financial loss. I don't know the facts of the case you are referencing, however, if an outage is going to cost them money that is a sign of poor lawyers, not technical superiority of anybody's product. If they sold a product with a license that left them open to a suit like that they deserve it. Why do you think Compaq/HP have never been sued over the Alphacide, even though it was abrupt and totally unexpected? Good lawyers who saw to it that nothing was ever promised. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Thu, 09 Aug 2007 11:30:10 -0600 From: Keith Parris Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: Bill Gunshannon wrote: > Forgetting Hobbyist systems where they cluster for the sake of > clustering, what percentage of real, production VMS systems are > clusters as opposed to stand alone systems? The last time I saw figures for this, many years ago, about 50% of VMS systems were clustered. Now that you can form a cluster over any flavor of Ethernet, and don't have to buy proprietary CI or DSSI hardware, the percentage may be even higher today. ------------------------------ End of INFO-VAX 2007.434 ************************