1 INFO-VAX	Wed, 21 Jul 2004	Volume 2004 : Issue 401       Contents:" Re: 68 pin SCSI on 50 pin SCSI bus/ Re: Anyone have any HSZ to HSG upgrade stories? / Re: Anyone have any HSZ to HSG upgrade stories? / Re: Anyone have any HSZ to HSG upgrade stories? / Re: Anyone have any HSZ to HSG upgrade stories? - Attaching tk50 to simh emulator running on pc  decamds 1 Re: HSx ECB (cache battery) troubleshooting/specs 7 Re: Is it possible to route or tunnel the SCS protocol? - Re: Itanium Licensing and Hobbyist on Itanium $ Re: Just curious about MAXPROCESSCNT Re: Looking for a Video Card Re: Looking for a Video Card Re: Looking for a Video Card Re: Looking for a Video Card Re: Looking for a Video Card Re: Looking for a Video Card Re: Looking for a Video Card Re: Looking for a Video Card* Re: Math Problems on latest survey results6 Re: OpenVMS Management Station -  cluster storage ???? Re: OpenVMS security?  Re: OpenVMS security?  Re: OpenVMS security?  Re: OpenVMS security?  Re: OpenVMS security?  Re: OpenVMS security?  Re: OpenVMS security? ! Re: Power switch for DEC Alpha PC 2 Sending page messages from VMS using a Unix server6 Re: Sending page messages from VMS using a Unix server6 Re: Sending page messages from VMS using a Unix server6 Re: Sending page messages from VMS using a Unix server6 Re: Sending page messages from VMS using a Unix server Solaris to Itanium...  Re: Telnet without timestamp1 Re: Turn-key OpenVMS E-mail, web server solution? 1 Re: Turn-key OpenVMS E-mail, web server solution? 1 Re: Turn-key OpenVMS E-mail, web server solution? 1 Re: Turn-key OpenVMS E-mail, web server solution? 1 Re: Turn-key OpenVMS E-mail, web server solution? 1 Re: Turn-key OpenVMS E-mail, web server solution?  VMS/Cluster vs. Polyserve ?  Re: VMS/Cluster vs. Polyserve ?  [OT] Haggis in comp.os.vms :-)" Re: [OT] Haggis in comp.os.vms :-)" Re: [OT] Haggis in comp.os.vms :-)" Re: [OT] Haggis in comp.os.vms :-)B [OT]: Microsoft Patent Cross License - Open Source Software Impact  F ----------------------------------------------------------------------  % Date: Wed, 21 Jul 2004 08:42:58 -0400 ; From: "Island Computers USA" <dbturner@islandco.com.nospam> + Subject: Re: 68 pin SCSI on 50 pin SCSI bus 0 Message-ID: <10fsolr2tjm5he1@news.supernews.com>  + We have been thtough this a number of times   * You can do this with OLDER UW and U2 Disks/ U3 Drives do not like being bumped down to 8Bit   L On many wide drives, you need to diable FW negotiation (most IBM drives have this) I Compaq used IBM drives in their faster storageworks narrow canisters with & the FW NEgotiation disabled by default  K We sell the 50 to 68 pin converters here alsoo which we use all the time on 
 Alpha for $15 & 50pin Narrow Female to 68P Wide Female   DT   --   David B Turner Island Computers US Corporation  2700 Gregory St., Suite 180  Savannah GA 31404  DIRECT#: 912 232 1729  Tel: 912 447 6622  Fax: 912 201 0402  Email: dbturner@hpaq.net http://www.hpaq.net   < "Stuart Norris" <stuie_norris@yahoo.com.au> wrote in message7 news:51262235.0407171352.4a97d539@posting.google.com...  > Hello All, > % > I have an Alpha station 200i 4/100.  > L > I am finding it very hard to get 50 pin SCSI disks that are good enough to use. > 6 > I have several RZ29 drives with a 68 pin interfaces. > K > If I buy a female 50 pin to female 68 pin SCSI adapter can I connect this 
 adapter toK > the mother board and run the 68 pin drives.  The 68 pin SCSI cable I have  states> > that all the drives including the last must be unterminated. > - > From a gender point of view this will work.  > + > Will this work from a SCSI point of view?  >  > Stuart   ------------------------------  + Date: Wed, 21 Jul 2004 07:51:58 +0000 (UTC) 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)8 Subject: Re: Anyone have any HSZ to HSG upgrade stories?0 Message-ID: <newscache$84z61i$jcp$1@news.sil.at>  f In article <b3531425.0407201219.6c01ba36@posting.google.com>, twnews@kittles.com (Thomas Wirt) writes:E >I have 2 AlphaServer 4100 with 5GB mem. and 2 300 MHz CPUs.  My main E >disk space is on a pair of HSZ70 controllers.  We have plenty of CPU C >and memory available all of the time, but we clearly seem IO bound F >bound at many busy points during each day.  I may need to increase myF >total computing capacity by about 40%.  I am entertaining the idea ofB >adding an HSG80 cabinet (everything redundant) to increase our IOD >throughput.  My question is, has anyone upgraded similar (remotely)D >hardware from HSZ SCSI based to HSZ SAN based?  I do not want to doG >this only to find out that my AS4100 can not keep up with my HSG80.  I C >understand that the AS 4100 may become the bottleneck (CPU wise or B >even IO wise), but I need to know that in some IO bound scenarios? >adding an HSG80 could give me a significant performance boost.   D I do not have experiences with HSZ. I do however read here, that theI HSG is not that fast as V7.3-2 (and as written even V7.3-1) could be, and G VMS can (and obviously have) overrun its firmware with I/Os (but it was G a earlier ACS firmware version, I don't know if a newer version already C fixed this problem, check the HSG firmware release notes yourself).   Y http://groups.google.com/groups?threadm=5a85bce2.0309120734.7b37cbdf%40posting.google.com @ http://www.compaq.com/products/storageworks/softwaredrivers/acs/  L A HSV110/EVA5000 is about twice as fast as a HSG80, has virtual storage (andM seems problematic). It may be a overkill for you, cause it's new & expensive.   J HSG80 are quite cheap now (but not by HPQ). You may buy a ESA8000/ESA12000J (with a HSG80-A) or even a EMA8000/EMA12000 (with a HSG80-B) at an auctionK on EBAY and try it yourself. And if it's worth, you may ask HPQ for support  (or another HSG80)...   A eg. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5702610251     I You may however try also an MSA1000 (but then again, could be a overkill)   A eg. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5710683452    -Peter  F PS: 1GBit/s FC Controller are also cheaper now as 2GBit/s are current.A eg. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5704904293  --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Wed, 21 Jul 2004 09:23:31 -0400 ; From: "Island Computers USA" <dbturner@islandco.com.nospam> 8 Subject: Re: Anyone have any HSZ to HSG upgrade stories?0 Message-ID: <10fsr1sk6hh6b5e@news.supernews.com>  L If you are thinking of going fiber then you should entertain the entry level MSA1000    We have them in stock   H Fully configured (dual controllers but only run in 1x active 1 x standby9 mode) with 14 x 146GB Disks, dual 512MB Cache controllers I dual SAN switch and Tru64 version of firmware you still get change out of  $40,000 4 That includes the 2Gbit controllers (2 of them too!)     Regards        --   David B Turner Island Computers US Corporation  2700 Gregory St., Suite 180  Savannah GA 31404  DIRECT#: 912 232 1729  Tel: 912 447 6622  Fax: 912 201 0402  Email: dbturner@hpaq.net http://www.hpaq.net   C "Peter 'EPLAN' LANGSTOEGER" <peter@langstoeger.at> wrote in message * news:newscache$84z61i$jcp$1@news.sil.at...? > In article <b3531425.0407201219.6c01ba36@posting.google.com>, ( twnews@kittles.com (Thomas Wirt) writes:G > >I have 2 AlphaServer 4100 with 5GB mem. and 2 300 MHz CPUs.  My main G > >disk space is on a pair of HSZ70 controllers.  We have plenty of CPU E > >and memory available all of the time, but we clearly seem IO bound H > >bound at many busy points during each day.  I may need to increase myH > >total computing capacity by about 40%.  I am entertaining the idea ofD > >adding an HSG80 cabinet (everything redundant) to increase our IOF > >throughput.  My question is, has anyone upgraded similar (remotely)F > >hardware from HSZ SCSI based to HSZ SAN based?  I do not want to doI > >this only to find out that my AS4100 can not keep up with my HSG80.  I E > >understand that the AS 4100 may become the bottleneck (CPU wise or D > >even IO wise), but I need to know that in some IO bound scenariosA > >adding an HSG80 could give me a significant performance boost.  > F > I do not have experiences with HSZ. I do however read here, that theK > HSG is not that fast as V7.3-2 (and as written even V7.3-1) could be, and I > VMS can (and obviously have) overrun its firmware with I/Os (but it was I > a earlier ACS firmware version, I don't know if a newer version already E > fixed this problem, check the HSG firmware release notes yourself).  >  > Y http://groups.google.com/groups?threadm=5a85bce2.0309120734.7b37cbdf%40posting.google.com B > http://www.compaq.com/products/storageworks/softwaredrivers/acs/ > I > A HSV110/EVA5000 is about twice as fast as a HSG80, has virtual storage  (andD > seems problematic). It may be a overkill for you, cause it's new &
 expensive. > L > HSG80 are quite cheap now (but not by HPQ). You may buy a ESA8000/ESA12000L > (with a HSG80-A) or even a EMA8000/EMA12000 (with a HSG80-B) at an auctionE > on EBAY and try it yourself. And if it's worth, you may ask HPQ for  support  > (or another HSG80)...  > C > eg. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5702610251  >  > K > You may however try also an MSA1000 (but then again, could be a overkill)  > C > eg. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5710683452  >  > -Peter > H > PS: 1GBit/s FC Controller are also cheaper now as 2GBit/s are current.C > eg. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5704904293  > --   > Peter "EPLAN" LANGSTOEGER ' > Network and OpenVMS system specialist  > E-mail  peter@langstoeger.atH > A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 21 Jul 2004 09:22:28 -0700& From: twnews@kittles.com (Thomas Wirt)8 Subject: Re: Anyone have any HSZ to HSG upgrade stories?= Message-ID: <b3531425.0407210822.1219f929@posting.google.com>    Alan wrote:    > Thomas Wirt wrote: > B >> I have 2 AlphaServer 4100 with 5GB mem. and 2 300 MHz CPUs.  My mainC >> disk space is on a pair of HSZ70 controllers.  We have plenty of  CPU E >> and memory available all of the time, but we clearly seem IO bound - >> bound at many busy points during each day.  >  > C > Are you already running VMS v7.3-2 with the XFC?   If not, do so. @ > At one of my sites, it cut a ten minute I/O bound job to a ten > second job.   YMMV >  > Alan >   C We are running VMS 7.3-1 and XFC cache had a positive impact on our E system.  Unfortunately our DB and transactions are not all cache-able C (lots of single record once a day reads writes).  The XFC gave us a @ 10% to 25% IO increase.  I am hoping that an upgrade to an HSG80A system will give us at least a 200% IO increase.  I know that the E HSG80 should give us at least that speed increase over our HSZ70, but E I need to know if I should expect that the AS 4100 can keep up enough  to see the IO speed increase.  --     Thomas Wirt  Systems Manager  Kittle's Home Furnishings  Indianapolis, IN   ------------------------------    Date: 21 Jul 2004 10:29:32 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) 8 Subject: Re: Anyone have any HSZ to HSG upgrade stories?- Message-ID: <LLd+BnJxeX9+@malvm7.mala.bc.ca.>   > In article <b3531425.0407210822.1219f929@posting.google.com>, -      twnews@kittles.com (Thomas Wirt) writes:  > E > We are running VMS 7.3-1 and XFC cache had a positive impact on our G > system.  Unfortunately our DB and transactions are not all cache-able E > (lots of single record once a day reads writes).  The XFC gave us a B > 10% to 25% IO increase.  I am hoping that an upgrade to an HSG802 > system will give us at least a 200% IO increase.  K    In some simple tests I did ( on a DS20 ) the read throughput on an HSG80 H was 3 to 4 times that of an HSZ70 ( BACKUP/PHYSICAL of disks to NL: with
 no CRC ).    >  I know that theG > HSG80 should give us at least that speed increase over our HSZ70, but G > I need to know if I should expect that the AS 4100 can keep up enough  > to see the IO speed increase.   M    On the DS20 doing the above noted tests the CPU time was extremely low, so J the actual work of moving data off the HSG80 is minimal. Without knowing aH lot about your database it's hard to say what the effect will be in yourI situation ( ie how much processing does your database have to do for each J block read? ). Is CPU anywhere near a limiting factor now ( eg what's your" average idle time with MON CPU? ).   ------------------------------    Date: 21 Jul 2004 06:57:35 -0700' From: rcwood0000@yahoo.com (Roy C Wood) 6 Subject: Attaching tk50 to simh emulator running on pc< Message-ID: <22e43e71.0407210557.d9c0623@posting.google.com>  B I know the tk50 is scsi device and in theory can be hung off a pc.H  If I install the hardware, can a pc running vax-vms under simh actuallyO  read tapes? If it can, how do I set it up. I already have vax-vms 7.3 running  
  on my pc.    RC Wood  former DECCIE....   ------------------------------  # Date: Wed, 21 Jul 2004 14:29:20 GMT # From: "John N." <JNixon@cfl.rr.com>  Subject: decamds= Message-ID: <45vLc.36609$KP6.2194716@twister.tampabay.rr.com>   K In a pure VMS environment, can we continue to run DECAMDS, or do we need to  go to Availability Manager?   I I have had AMDS running on our VMS 7.3-1 cluster nodes for some time.  We H just added another cluster node at VMS 7.3-2.  I cannot get this node toC display on our VMS 7.3-1 system where the AMDS analyzer is running.   K Is this because it the analyzer is older than the collector?  Based on this H assumption, and the fact that the newer collector and analyzer should beK "better" I want to upgrade all my nodes to the latest version of AMDS.  But 4 I cannot find the AMDS installation kit.  Any ideas?   ------------------------------  % Date: Wed, 21 Jul 2004 08:26:04 +0100 & From: Nic Clews <spamthis@[127.0.0.1]>: Subject: Re: HSx ECB (cache battery) troubleshooting/specs' Message-ID: <cdl5og$mll$1@lore.csc.com>    David J Dachtera wrote:  > Bob Kaplow wrote:  > b >>In article <40F89A49.79DE2D76@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: >>K >>>Good luck. Rumor has it that cache batteries turn to mush if they're not 4 >>>kept charged, ragardless of the voltage readings. >>M >>That's been my experience. They have an in-use life of about 2 years, and a K >>sit on the shelf life of at most a few months. Lead Acid needs to be kept  >>fully charged or it's trash. >  > C > Wonder if NiMH or Lithium-ion would have been better, had it been  > feasible/available...   C If you really want to see people lose it, go to a forum discussing   mobile phone batteries...   G Lithium-Ion (or the later polymer cells*) have a limited life as well,  G of generally around 2 years, they just wear out. Their prime advantage  I is storage capacity versus weight. In practice though, the Lithium cells  L aren't too bad, in my cellphone, I'm using one that is around 2.5 years old.  H NiMH usually offer a higher storage density than NiCad, but having used 0 both NiCad and NiMH, NiMH is not a magic bullet.  @ Store changed, discharged, or do you periodically do a charge / E discharge cycle to prolong the life, does it damage the cells if you  " repeatedly charge, trickle charge?  H The issue I think is that the charging circuitry is different. *Lithium I cells have a "base" voltage of 3.6V, NiCad and NiMH, 1.2 of course. Lead  F Acid is base 2.0V, NiCad and NiMH have different but similar, but not G sufficiently similar you can use NiMH cells in a NiCad charger, you'll  ( shorten the life of the costlier NiMH's.  D In practice, I've seen some NiCads last a very, very long time, and F under quite "abusive" conditions. I've also seen NiMH have very short E lives. For my money, NiCads have the most predictable usage pattern,  E therefore determining the probable failure point is easier, so still  & hold real value even in today's world.  K Digital I believe used to have their own dedicated "battery" manufacturing.    --  E Regards, Nic Clews a.k.a. Mr. Car Park Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  % Date: Wed, 21 Jul 2004 17:15:28 +0100 & From: Nic Clews <spamthis@[127.0.0.1]>@ Subject: Re: Is it possible to route or tunnel the SCS protocol?' Message-ID: <cdm4p8$31e$1@lore.csc.com>    David J Dachtera wrote:    > Nic Clews wrote: >  >>David J Dachtera wrote:  >> >> >>>JF Mezei wrote: >>>  >>>    >>>My take:  >>> 8 >>>Early 1990's processor speeds <100MHz, Ethernet = 3Mb >>> 0 >>>2004 processor speeds >500MHz, Ethernet = 1Gb >>> = >>>Yes, I'd expect to see significant differences in latency.  >>G >>First supported interconnects were CI, remember that these have LOCAL J >>(i.e. non CPU dependent) processing capability, so comparing chip speeds >>is pointless.  >  > ? > Huh? "Star couplers" are passive devices - totally unpowered.   H The CI adapters in VAXen have on-board processing. I'm not sure if this I is true of the Alpha PCI variety (CIPCA's). [I wasn't talking about star  G couplers, I know they are passive devices with "pulse" transformers in   them].    F >>Ethernet got supported, and remember it's 10 meg half duplex but theH >>processing of this messaging happens not at process priority, but at a! >>more basic and faster priority.  >  > 1 > Early Ethernet - at the dawn of LAVc - was 3Mb.   C ... and half duplex, receive OR transmit. with full duplex you can  F receive and transmit at the same time so, even if it was just 3 Mb it I could be 3Mb in AND 3Mb out, "total" of 6 Mb bandwidth. However my point  F was that compared to today's networking, its primitive, and slow. The F effective maximum quoted if often 0.3 to 0.4 of a circuit, due to the C way it works in terms of utilization, the rate is 10Mb, but packet  E sizes, and all things being equal other devices on the ether, you're  G averaging 0.3 to 0.4 of that 1.0. Having said that, I've seen it happy  G enough at 0.6 to 0.7 with the collision light looking like an activity  C light, and it still delivers reasonable performance. I was sort of  G impicticly saying this, and of course you know the maximum length of a  E segment, due to the "speed of light" characteristics of the physical  B signalling, the timing of the CSMACD sequence is set to work with D maximum lengths, and any devices outside that would fail. [You will I never see half duplex with gigabit circuits, it simply can't hang around   waiting that long.]   F The point I was making is that, 3Mb per second, and that's the design G tolerance, is low by today's standards. I've glossed the details above  5 of course, so don't shoot me! 3Mb is 6 * 56k modems*.   F A DSL at 512kb easily exceeds the 300kb that LAVC clustering operates I over. However of course, typically you're connecting at Ethernet speeds,  D but if you're 100Mb, there's no problem. Theoretically, a couple of E hobbyists could *cluster* over the internet, tunelling between their  A respective 512Mb links. It does of course depend on the upstream  C capability, and as Hoff's pionted out, IP clustering isn't a great  E thing, becuase it's not in IP's philosophy to have delivered latency   guarantees.   8 * OK, yes I know that a 56k modem upstream is only 33.6k  G ATM is quite well suited to clustering. The laymans simplification is,  F if you remember the old films where you saw people passing buckets of E water to each other in a line to put out a fire, the ATM packets are  G like those buckets, and they move at the same rate from A to B if they  ? have data in them or not, it's a guaranted latency / bandwidth.   G Ethernet is like a tightrope that can only have one person standing on  F it at a time. Full duplex is like an "in" and an "out" tightrope, but - there are time limits on reaching either end.   F However, the real point is the cluster has a limited tolerance. ABove G all data integrity is important, and it's quite definitely a "needs of  H the many outweigh the needs of the few" so you've 3 nodes, and if their F data integrity is dependent on a node which responds slowly, then the G ultimate delivered performance is reduced, but you also get to a point  H where the two nodes decide, is that third node just too slow, or has it E died? I think I'm in the realms of teaching you to suck eggs so I'll  
 stop here.   >>[snip]: >>Verel Boaen is the main man for this sort of discussion. >  >  > How does one reach him/her?   G I'll drop you a note with his email... Just say the DSSI advocate from   England sent you ;-)   --  E Regards, Nic Clews a.k.a. Mr. Car Park Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  % Date: Wed, 21 Jul 2004 09:07:06 +0300 0 From: "Guy Peleg" <guy.peleg@remove_this_hp.com>6 Subject: Re: Itanium Licensing and Hobbyist on Itanium* Message-ID: <40fe0ad4@usenet01.boi.hp.com>  ? "David J Dachtera" <djesys.nospam@comcast.net> wrote in message % news:40FAAE1D.ED128966@comcast.net...  > Alex Daniels wrote:  > > K > > With some of the talk about what will (and will not) be included in the J > > different licensing options with Itanium and the comments about it not being J > > definitively decided yet, I had a look around on an pre 8.2 base level > > machine. > > J > > The products in the different categories were no shock and as previousK > > stated by hp, but looks like hobbyists won't have to load quite so many L > > licences any more, and more importantly shows that they are committed to: > > catering for hobbyist's from the start. Well done hp.. > >  > > From the early 8.2 box...  > >  > >    FOE     Foundation 5 > >    OPENVMS-I64,OPENVMS-USER,DVNETEND,DW-MOTIF,UCX > > >    TDC,DCOM-MIDL,X500-ADMIN-FACILITY,X500-DIRECTORY-SERVER > >  > >    EOE     Enterprise # > >    DECRAM,RMSJNL,VOLSHAD,SYSMGT  > >  > >    MCOE    Mission Critical + > >    RTR-SVR,VMSCLUSTER,VMSCLUSTER-CLIENT  > >  > >    HOE     Hobbyist  > >    OPENVMS-HOE > > E > > It has been documented prior that each category will also include 
 everything" > > from the preceding categories. > I > It is interesting to note that the most common place for VMS to live is J > in the back-end server tier, which of course requires the MCOE licensingJ > tier. That is, once again: minimal acceptable OpenVMS licensing bundle =E > prohibitive cost-to-acquire, while affordable (read: competitive) =  > virtually useless. > J > These supposed business school graduates just don't get it, and probably
 > never will.  >  > D.J.D.  L It is possible to purchase the FOE pak and then purchase separately specificB products from other bundle. So you can have an rx2600 with FOE andC VMSCLUSTER license. Without going into too much details (as I'm not E a sale person) the price of the VMSCLUSTER license for rx2600 will be 	 different  than the price for a Superdome.    FWIW,  Guy    ------------------------------    Date: 21 Jul 2004 08:33:36 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) - Subject: Re: Just curious about MAXPROCESSCNT 3 Message-ID: <RHQPqtEBsgCk@eisner.encompasserve.org>   Z In article <cdjfl9$h1l$1@news.doit.wisc.edu>, "Jim" <jim.nospam.gould@charter.net> writes:K > I have noticed that no matter what I set MAXPROCESSCNT to, AUTOGEN always M > sets BALSETCNT to exactly 2 less.  This is on three different clusters, one N > running 7.3-2, the other two running 7.3-1, and one lone server running 6.2.K > Is the calculation of BALSETCNT based on other parameters or is it always  > just 2 less.  E    MAXPROCESSCNT = BALSETCNT + 2, the other two being the swapper and G    the null process (even though the null process no longer exists, its     process header does).  G    I've forced the values to be different and witnessed the undesirable     behaviour.  It's real.    ------------------------------  % Date: Wed, 21 Jul 2004 08:39:53 +0200 E From: Bitnissen <bitnissen#SPAMMER_FORMAT_YOUR_HARDDISK#@hotmail.com> % Subject: Re: Looking for a Video Card 8 Message-ID: <0p3sf0d95q0srd7mad5skkpuu2cveh66vj@4ax.com>   Hi Bobby  < Try Island Computers (http://www.islandco.com/graphics.html)   /Jimmi  3 On Tue, 20 Jul 2004 17:24:22 -0400, "Bobby Coleman"  <colemanrl@ornl.gov> wrote:    >Hi, > K >We have two DS20E Alpha workstations that came with DEPVZ-AA multifunction K >video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768.  We K >would like to upgrade these to 1280x1024 or 1600x1200.  We are running VMS I >Ver 7.3.  Would someone be so kind as to recommend a card that we should I >consider?  We really only care about 2D graphics and text.  I've not had ; >much luck with "phone sales" with HP when it comes to VMS.  > K >The issue at hand is really just a matter of gaining more screen space for H >the multiple inputs and display windows that are being used.  The otherM >option would be to configure for two lower-resolution monitors but I have no + >experience with this type of setup on VMS.  > & >Any suggestions are most appreciated. >  >Bobby >  >  >    ------------------------------  # Date: Wed, 21 Jul 2004 12:30:39 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> % Subject: Re: Looking for a Video Card 2 Message-ID: <PltLc.6191$V42.3555@news.cpqcorp.net>  I If you want to buy it on the secondary market, the 3DLabs Oxygen VX1 card L fits your bill.  It is great at 2D.  My DS20 is running it at 1920x1200 with 16bpp.  H If you want it directly from HP, then the Radeon 7500 is the one to get.J This card can also do 3D.  But you will need to get up to V7.3-1 or better (V7.3-2) for this card.   " Both should work well on the DS20.      5 "Bobby Coleman" <colemanrl@ornl.gov> wrote in message " news:cdk2i6$fdj$1@sws1.ornl.gov... > Hi,  > L > We have two DS20E Alpha workstations that came with DEPVZ-AA multifunctionL > video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768.  WeL > would like to upgrade these to 1280x1024 or 1600x1200.  We are running VMSJ > Ver 7.3.  Would someone be so kind as to recommend a card that we shouldJ > consider?  We really only care about 2D graphics and text.  I've not had< > much luck with "phone sales" with HP when it comes to VMS. > L > The issue at hand is really just a matter of gaining more screen space forI > the multiple inputs and display windows that are being used.  The other K > option would be to configure for two lower-resolution monitors but I have  no, > experience with this type of setup on VMS. > ' > Any suggestions are most appreciated.  >  > Bobby  >  >  >  >    ------------------------------  % Date: Wed, 21 Jul 2004 08:39:10 -0400 ; From: "Island Computers USA" <dbturner@islandco.com.nospam> % Subject: Re: Looking for a Video Card 0 Message-ID: <10fsoen9ult1l35@news.supernews.com>  K The replacement for most of the older cards is now the Radeon 7500 PCI Card = I believe resolution is 2048 x 1920 maximum at 24 Bit color -   C The ATI Radeon has a SVGA and DVI out - rock solid product for $189 + The ATI requires VMS 7.3-1 and a patch kit.     F The DEPVZ though, has an on board SCSI controller (LVD/U2) so you will  either need to replace this cardL with the 3X-KZPCA-AA U2 SCSI PCI (uses the same 53C895 chip as the DEPVZ) or my suggestion would be to F upgrade to VMS 7.3-1 and install the newest card 3X-KZPEA-DB U160 SCSI Controller.     L As for other controllers, the SN-PBXGD-AE Powerstorm 350 (we have the latest& rev D909 for DS20e) will also give youG very high resolution graphics, thougn it is considerably more expensive  $1200 currently.= You would not need to upgrade your VMS version with this card    Regards to all at COV          --   David B Turner Island Computers US Corporation  2700 Gregory St., Suite 180  Savannah GA 31404  DIRECT#: 912 232 1729  Tel: 912 447 6622  Fax: 912 201 0402  Email: dbturner@hpaq.net http://www.hpaq.net   5 "Bobby Coleman" <colemanrl@ornl.gov> wrote in message " news:cdk2i6$fdj$1@sws1.ornl.gov... > Hi,  > L > We have two DS20E Alpha workstations that came with DEPVZ-AA multifunctionL > video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768.  WeL > would like to upgrade these to 1280x1024 or 1600x1200.  We are running VMSJ > Ver 7.3.  Would someone be so kind as to recommend a card that we shouldJ > consider?  We really only care about 2D graphics and text.  I've not had< > much luck with "phone sales" with HP when it comes to VMS. > L > The issue at hand is really just a matter of gaining more screen space forI > the multiple inputs and display windows that are being used.  The other K > option would be to configure for two lower-resolution monitors but I have  no, > experience with this type of setup on VMS. > ' > Any suggestions are most appreciated.  >  > Bobby  >  >  >  >    ------------------------------  % Date: Wed, 21 Jul 2004 09:19:53 -0400 ; From: "Island Computers USA" <dbturner@islandco.com.nospam> % Subject: Re: Looking for a Video Card 0 Message-ID: <10fsqr1deiifrad@news.supernews.com>   Fred# DO you know something I don't or...   H The 3DLabs Oxygen was discontinued and is extremely rare in the reseller marketH If there is anyone out there with these then we would love to buy them !   --   David B Turner Island Computers US Corporation  2700 Gregory St., Suite 180  Savannah GA 31404  DIRECT#: 912 232 1729  Tel: 912 447 6622  Fax: 912 201 0402  Email: dbturner@hpaq.net http://www.hpaq.net   D "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message, news:PltLc.6191$V42.3555@news.cpqcorp.net...K > If you want to buy it on the secondary market, the 3DLabs Oxygen VX1 card I > fits your bill.  It is great at 2D.  My DS20 is running it at 1920x1200  with > 16bpp. > J > If you want it directly from HP, then the Radeon 7500 is the one to get.L > This card can also do 3D.  But you will need to get up to V7.3-1 or better > (V7.3-2) for this card.  > $ > Both should work well on the DS20. >  >  > 7 > "Bobby Coleman" <colemanrl@ornl.gov> wrote in message $ > news:cdk2i6$fdj$1@sws1.ornl.gov... > > Hi,  > > @ > > We have two DS20E Alpha workstations that came with DEPVZ-AA
 multifunction J > > video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768. WeJ > > would like to upgrade these to 1280x1024 or 1600x1200.  We are running VMS L > > Ver 7.3.  Would someone be so kind as to recommend a card that we shouldL > > consider?  We really only care about 2D graphics and text.  I've not had> > > much luck with "phone sales" with HP when it comes to VMS. > > J > > The issue at hand is really just a matter of gaining more screen space for K > > the multiple inputs and display windows that are being used.  The other H > > option would be to configure for two lower-resolution monitors but I have > no. > > experience with this type of setup on VMS. > > ) > > Any suggestions are most appreciated.  > > 	 > > Bobby  > >  > >  > >  > >  >  >    ------------------------------  % Date: Wed, 21 Jul 2004 14:36:14 +0000 - From: David B Sneddon <dbsneddon@bigpond.com> % Subject: Re: Looking for a Video Card * Message-ID: <40FE7F5E.1020100@bigpond.com>  % Fred Kleinsorge was overheard to say: K > If you want to buy it on the secondary market, the 3DLabs Oxygen VX1 card N > fits your bill.  It is great at 2D.  My DS20 is running it at 1920x1200 with > 16bpp. > J > If you want it directly from HP, then the Radeon 7500 is the one to get.L > This card can also do 3D.  But you will need to get up to V7.3-1 or better > (V7.3-2) for this card.  > $ > Both should work well on the DS20.  B Will the Radeon 7500 work on a Digital Personal Workstation 600au?+ The Oxygen cards required an EV6 or better.    Regards, Dave.  --  I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/ I DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------  # Date: Wed, 21 Jul 2004 14:52:28 GMT 6 From: Gib Copeland <copeland@not.jenni.path.uiowa.edu>% Subject: Re: Looking for a Video Card . Message-ID: <MqvLc.138614$JR4.93675@attbi_s54>   Bobby Coleman wrote: > Hi,  > L > We have two DS20E Alpha workstations that came with DEPVZ-AA multifunctionL > video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768.  WeL > would like to upgrade these to 1280x1024 or 1600x1200.  We are running VMSJ > Ver 7.3.  Would someone be so kind as to recommend a card that we shouldJ > consider?  We really only care about 2D graphics and text.  I've not had< > much luck with "phone sales" with HP when it comes to VMS.  >  @ According to the specs, this card supports 1280x1024 resolution. Try adding the lines:    $  decw$xsize_in_pixels == 1280o $  decw$ysize_in_pixels == 1024   , to sys$manager:decw$private_server_setup.com   Gibu   ------------------------------  # Date: Wed, 21 Jul 2004 17:47:09 GMTe9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>e% Subject: Re: Looking for a Video Card 2 Message-ID: <x_xLc.6226$Ir2.4273@news.cpqcorp.net>  @ Nope.  I just assumed they must be available on the used market.    F "Island Computers USA" <dbturner@islandco.com.nospam> wrote in message* news:10fsqr1deiifrad@news.supernews.com... > Fred% > DO you know something I don't or...n > J > The 3DLabs Oxygen was discontinued and is extremely rare in the reseller > marketJ > If there is anyone out there with these then we would love to buy them ! >e > -- e > David B Turner! > Island Computers US Corporation  > 2700 Gregory St., Suite 180V > Savannah GA 31404r > DIRECT#: 912 232 1729e > Tel: 912 447 6622i > Fax: 912 201 0402  > Email: dbturner@hpaq.net > http://www.hpaq.net  >IF > "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message. > news:PltLc.6191$V42.3555@news.cpqcorp.net...H > > If you want to buy it on the secondary market, the 3DLabs Oxygen VX1 cardK > > fits your bill.  It is great at 2D.  My DS20 is running it at 1920x1200, > with
 > > 16bpp. > >eL > > If you want it directly from HP, then the Radeon 7500 is the one to get.G > > This card can also do 3D.  But you will need to get up to V7.3-1 orO better > > (V7.3-2) for this card.] > >i& > > Both should work well on the DS20. > >. > >s > > 9 > > "Bobby Coleman" <colemanrl@ornl.gov> wrote in message & > > news:cdk2i6$fdj$1@sws1.ornl.gov...	 > > > Hi,- > > >-B > > > We have two DS20E Alpha workstations that came with DEPVZ-AA > multifunctionrL > > > video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768. > WeL > > > would like to upgrade these to 1280x1024 or 1600x1200.  We are running > VMSbG > > > Ver 7.3.  Would someone be so kind as to recommend a card that wen shouldJ > > > consider?  We really only care about 2D graphics and text.  I've not hadi@ > > > much luck with "phone sales" with HP when it comes to VMS. > > >sL > > > The issue at hand is really just a matter of gaining more screen space > for G > > > the multiple inputs and display windows that are being used.  Theh other1J > > > option would be to configure for two lower-resolution monitors but I > have > > no0 > > > experience with this type of setup on VMS. > > > + > > > Any suggestions are most appreciated.6 > > >a > > > Bobby0 > > >i > > >r > > >q > > >t > >w > >a >e >    ------------------------------  # Date: Wed, 21 Jul 2004 17:48:28 GMT29 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>,% Subject: Re: Looking for a Video Card12 Message-ID: <M%xLc.6228$hz2.6037@news.cpqcorp.net>  K Of course, you are right.  This is the 3DLabs P2 chip, and I think there is ( enough memory on that card for 1280x1024  C "Gib Copeland" <copeland@not.jenni.path.uiowa.edu> wrote in messageo( news:MqvLc.138614$JR4.93675@attbi_s54... > Bobby Coleman wrote: > > Hi,v > >t@ > > We have two DS20E Alpha workstations that came with DEPVZ-AA
 multifunctionaJ > > video/ehternet/SCSI cards.  These have a max resolution of 1024 x 768. WeJ > > would like to upgrade these to 1280x1024 or 1600x1200.  We are running VMS4L > > Ver 7.3.  Would someone be so kind as to recommend a card that we shouldL > > consider?  We really only care about 2D graphics and text.  I've not had> > > much luck with "phone sales" with HP when it comes to VMS. >  > >6B > According to the specs, this card supports 1280x1024 resolution. > Try adding the lines:t >G! > $  decw$xsize_in_pixels == 1280 ! > $  decw$ysize_in_pixels == 1024  >t. > to sys$manager:decw$private_server_setup.com >a > Gib    ------------------------------    Date: 21 Jul 2004 07:00:56 -07000 From: keith.cayemberg@arcor.de (Keith Cayemberg)3 Subject: Re: Math Problems on latest survey resultsg< Message-ID: <f64e50ca.0407210600.d1cccd0@posting.google.com>  A I don't want to be impolite, but I have the impression my earlier B Email to this thread was not noticed for some technical reason, or maybe the way I posted.i$ Although I do see my post in Google.  ` http://groups.google.com/groups?selm=40fce7ab%240%245927%249b4e6d93%40newsread4.arcor-online.net    OpenVMS had *104 votes* not 014!   Cheers!m   Keith Cayembergs  u norm.raphael@metso.com wrote in message news:<OF1E47FB56.D080A180-ON85256ED7.006C8900-85256ED7.006CBDD8@metso.com>...a@ > "John Smith" <a@nonymous.com> wrote on 07/20/2004 03:08:01 PM: >  > > Barry Treahy, Jr. wrote:J > > > Besides that VMS is extremely low, doesn't anyone else see a problemE > > > with this math?  The VMS counter must be over flowing, or being  > > > truncated,  somewhere!:m > > >l1 > > > The results from the last instapoll follow:g > > >hB > > > For your future server consolidation needs, how important isJ > > > supporting multiple operating systems on a single server: Check one. > > >  > > > Very important - 41a > > > Somewhat important - 74r > > > Not important at all - 26f > > > Does not apply - 5 > > >eI > > > If very or somewhat important, what collection of operating systems1H > > > would you like to see supported on a single server (check all that
 > > > apply):  > > >  > > > Windows - 88 > > > Linux - 100  > > > HP-UX - 46 > > > OpenVMS - 014) > > > Other - 25 > > >a > > >s > > > Total - 146I > >t > >0" > > There are three kinds of lies: >  > Make that four kinds of lies:b# > > Lies, damn lies, and statistics. > and campaign promises. >  > >  -- Mark Twain > >r > >S   ------------------------------  + Date: Wed, 21 Jul 2004 08:08:25 +0000 (UTC)c6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)? Subject: Re: OpenVMS Management Station -  cluster storage ????i0 Message-ID: <newscache$nvz61i$5ep$1@news.sil.at>  ` In article <40FDCB26.82DB3ABE@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes:! >Peter 'EPLAN' LANGSTOEGER wrote: r >> In article <a3c44af1.0407191100.92ce5f7@posting.google.com>, dave.baxter@bannerhealth.com (Dave Baxter) writes:# >> >Is HSZTERM a 3rd party product?. >> n= >> No. It is part of the StorageWorks Command Console (SWCC).e4 >> http://www.compaq.com/products/storageworks/swcc/ >c >Oohhh... Better check that...  O Yup. I should have written "to the best of my knowledge it was/is part of SWCC"1( And maybe then, I'm still wrong, too ;-)  8 >> Eg. OVMS 8.7B Platform Kit for HSG80 (31-Mar-2004) on] >> ftp://ftp.hp.com/pub/softlib/software3/COL6232/co-19283-1/SOL_HSG80_v87_VMS_279807-B21.zip W >> contains in directory AGENT the file SWCC25.EXE which is a zip archive with the file # >> DEC-AXPVMS-SWCC-V0205-106-1.PCSI7 >2E >I dissected that kit using PRODUCT COPY/FORMAT=REFERENCE and did not 5 >find the HSZTERM$SCSPAD executable in the .PCSI kit.o  9 Indeed. Shame on me, I didn't unpack it before posting...   G btw: Does anyone know, if SET HOST/SCSI was ever documented somewhere ?l Appearantly not in VMS Help...   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialistd E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  + Date: Wed, 21 Jul 2004 07:15:14 +0000 (UTC)d From: david20@alpha2.mdx.ac.uk Subject: Re: OpenVMS security?) Message-ID: <cdl562$rn3$2@news.mdx.ac.uk>   ` In article <40FDD654.B8D1823E@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: >Undisclosed wrote:i >>  K >> first off, just want to thank everyone for their very informative posts.t >>   >> now,n >>   >> Bob Ceculski wrote:p >> > Undisclosed <nomail@dontbeaweaselspammer.com> wrote in message news:<AMWdnWKtYYEgnWTdRVn-ig@comcast.com>... >> >J >> >>couple of questions from a OpenVMS newbie with a little (real little) >> >>security knowledge. >> >>oJ >> >>3. OpenVMS's native access control would be considered discressionaryB >> >>(sp..) like Unix's is, correct? Except that it has much finer* >> >>granularity (more priviledge levels). >> > >> >H >> > Yes, but first redesign and rewrite your unix to cleanly catagorize >> > and separateiK >> > Kernel Mode from Supervisor Mode and from User Mode. Three modes are aa >> > minimumI >> > for a correct ring protection system. The use of three or more rings. >> > happens togH >> > be a fully patented methodology by OpenVMS Engineering. OpenVMS has
 >> > four.J >> > OpenVMS also has 40 groups of higher mode functionality classified as >> > requiring >> > special named privileges. >> y( >> This is tangential to my questions... >> rJ >> but how can the VMS team have the use of 3 or more rings patented, whenN >> the ring concept originated with Multics, which had 6 hardware-based rings? > B >Um, Bob C. tends to be a bit, well, overly enthusiastic at times. >Careful how you take him.  4 In this instance Bob is just quoting Keith Cayemberg    
 David Webb Security Team Leader CCSS Middlesex University   ------------------------------    Date: 21 Jul 2004 05:29:25 -0700( From: bob@instantwhip.com (Bob Ceculski) Subject: Re: OpenVMS security?= Message-ID: <d7791aa1.0407210429.4e7f7f14@posting.google.com>   f young_r@encompasserve.org (Rob Young) wrote in message news:<zCD$f7y4mmvn@eisner.encompasserve.org>...h > In article <Y4ednf5iy_AwTGDdRVn-hg@comcast.com>, Undisclosed <nomail@dontbeaweaselspammer.com> writes: > L > > first off, just want to thank everyone for their very informative posts. > >  > > now, > >  > > Bob Ceculski wrote:c >  hJ > >>    holes - buffer-overflows. The secure Calling Standard is a central > >> design2 > >>    theme in OpenVMS.n > > K > > does the descriptor calling standard guarantee integer-based overflows  7 > > (really, integer wraparounds) cannot occur somehow?  > > K > > that's a definite vuln in C and I know there's no way to really fix it vA > > in that language, I don't know if Bliss has similar problems.  > > F > > all the fixes I've seen for integer overflows in C involve either  > > complicated C > > mostly theoretical hacks that constantly check the processor's  K > > equivalent of EFLAGS to see if a overflow flag has been tripped (which 1K > > would gut speed), or involve compiler modifications to create programs nG > > that detect the often huge memory shifts required to get a integer   > > overflow working.  > > L > > since that class of vulnerability was only made public in 2000, I would : > > be really impressed if OpenVMS already had it blocked. > >  >  > ? > 	Keith penned a definitive explantion of why VMS mostly isn'td? > 	vulnerable to smash the stack and buffer overflows.  Instead ) > 	of lifting portions, here is the link:m >  > http://tinyurl.com/6uooc > 	 > 				Robs   care to chime in Andrew? :)-   ------------------------------  % Date: Wed, 21 Jul 2004 08:28:47 -0400m- From: "John E. Malmberg" <wb8tyw@qsl.network>: Subject: Re: OpenVMS security?1 Message-ID: <49WdnX_j6PQd_GPdRVn-pw@adelphia.com>m   Undisclosed wrote: > J > but how can the VMS team have the use of 3 or more rings patented, when M > the ring concept originated with Multics, which had 6 hardware-based rings?s  I I am ignorant if Digital patented the use of the four rings, or licensed t it from someone else.t  > VAX had the four rings implemented in hardware, so does Alpha.  G The Intel 80286 and later also has four rings implemented in hardware. mD Some of the specifics differ in the specific access modes permitted.  G There seems to be a persistent myth that the x386 processors only have  D two rings, which is of course easily disproved by RTFM for them.  I F remember in the early specifications for the 80286 how the four rings + were clearly documented even with pictures.d  H This myth may be based on other operating systems like LINUX only using  two of the four rings.  I > does the descriptor calling standard guarantee integer-based overflows  5 > (really, integer wraparounds) cannot occur somehow?g   No.   I Code is required to inspect the descriptor and make sure that the bounds oF are honored.  Privileged code is required to make sure that addresses E passed by the user are ones that the user has permission to write to.   H Elimination of such bugs is a function of good programming practice and  quality control.  H It is more efficient to do those checks with a descriptor than it is to H search a memory range of undetermined length for a terminator character.  8 This removes the temptation to trust arbitrary pointers.  K > this is also the case when you are using a nonstandard system using some xG > of the compiler and kernel security mods that are out there for Unix l > variants.x > I > using SSP/Propolice and other compiler mods makes a lot of stack-based -E > attacks merely annoying crashes as opposed to critical emergencies..  I In the case of VMS, the applications that are typically attacked are not 6G running as "root", they are running with less privileges, so even if a t> stack attack succeeds, it only corrupts the local application.  H The applications typically do not have privileges to modify the various G logs on the system, so even if a hacker found a vulnerability, usually oE about the worse they could do is fill up a the disk quota while they  ( leave clear tracks on how to trace them.  G Again though errors are still possible, but it is quality control, and eI paying attention to lessons learned that prevent attacks from succeeding.   I > what about the heap? that's the usual next target 1337 haX0rs go after .( > once you plug the stack smashing hole. > K > does OVMS impliment anything like FreeBSD's heap "canaries", or seperate 0 > heaps a la OpenBSD?E  I I have no idea about those.  Heap allocations are in the mode requested, cG and the memory is private to the process.  That is just a variant of a < buffer overrun attack.  G The stack is protected from the heap by guard pages, so a heap overrun u' can not become a stack overrun exploit.n  D There are quite a few methods of heap management that depend on the ) application and the programming language.e  I For some heap allocations, guard pages are automatically allocated.  For  G others, the application or the language manages it's own sets of heap, r& which may or may not have guard pages.   -Johnl wb8tyw@qsl.network Personal Opinion Onlye   ------------------------------    Date: 21 Jul 2004 08:22:57 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)o Subject: Re: OpenVMS security?3 Message-ID: <g0InCEIhs4Ys@eisner.encompasserve.org>r  f In article <Y4ednf5iy_AwTGDdRVn-hg@comcast.com>, Undisclosed <nomail@dontbeaweaselspammer.com> writes: > I > does the descriptor calling standard guarantee integer-based overflows  5 > (really, integer wraparounds) cannot occur somehow?e  A    Integer overflows can be detected by the CPU and dealt with inaE    an exception-handling manner.  All languages supported by VMS haveaB    access to exception-handling capabilities.  That eliminates anyG    need to write special algorithms or use up CPU looking for potentials    problems.  G    Where and how much of this is made use of in VMS is not clear to me.,E    Since I don't know how the integer overflow exploit works, its notM    clear whether its a problem.    ------------------------------  % Date: Wed, 21 Jul 2004 17:50:04 +0200l0 From: Keith Cayemberg <keith.cayemberg@arcor.de> Subject: Re: OpenVMS security?A Message-ID: <40fe90ac$0$5930$9b4e6d93@newsread4.arcor-online.net>s   Rob Young wrote:     > 	sniph >  > ? > 	Keith penned a definitive explantion of why VMS mostly isn't,? > 	vulnerable to smash the stack and buffer overflows.  Insteado) > 	of lifting portions, here is the link:o >  > http://tinyurl.com/6uooc > 	 > 				Robn >   D Sorry, but I'm a different Keith. That excellent posting concerning E stack and buffer overflows was written by the acknowledged expert of -5 fully-shared disaster-tolerant multi-site clusters...: Mr. Keith Parris.    Cheers!s   Keith Cayemberg>( IBM Business Services - Hanover, Germany   ------------------------------  % Date: Wed, 21 Jul 2004 17:36:19 +0200 0 From: Keith Cayemberg <keith.cayemberg@arcor.de> Subject: Re: OpenVMS security?A Message-ID: <40fe8d74$0$5936$9b4e6d93@newsread4.arcor-online.net>'   Bob Koehler wrote:h > In article <Y4ednf5iy_AwTGDdRVn-hg@comcast.com>, Undisclosed <nomail@dontbeaweaselspammer.com> writes: > I >>does the descriptor calling standard guarantee integer-based overflows e5 >>(really, integer wraparounds) cannot occur somehow?  >  > C >    Integer overflows can be detected by the CPU and dealt with in G >    an exception-handling manner.  All languages supported by VMS have D >    access to exception-handling capabilities.  That eliminates anyI >    need to write special algorithms or use up CPU looking for potentialt >    problems. > I >    Where and how much of this is made use of in VMS is not clear to me.tG >    Since I don't know how the integer overflow exploit works, its notc! >    clear whether its a problem.t >   = An Integer overflow as a class of OS-level security risks or  I vulnerabilities would be new to me. And, I believe it would be a new one pH to most security professionals.   Math operations taking place on known @ types are well understood and thoroughly controlled operations. G Mathematical operations leading to a mathematical overflow are handled m@ by some form of exception often already recognized in hardware. I Certainly, mathematical errors can result in application errors. Whether r@ that error is to be regarded as a security risk in a particular G application or by an organisation, I regard to be outside of the scope nE of my Operating System security architecture essay from last January.h  H Maybe one could regard Integer Overflow as a class of OS-level security G risks or vulnerabilities if there were OS-level structures inwhich the sA use of an address, privilege or a resource were dependent on the nF integer's value which is determined or changeable from user-mode with C math operations. Such a dependency is not known to me in any OS. I  F suppose it's not impossible, but would require also having additional C unusual privileges (to break inherent OS structural integrity) and tH atypical operations (on privileged OS stuctures) in the scenarios I can  imagine at this time.a  F I was referring to the well-known class of OS-level security risks or F vulnerabilities known as "buffer overflows" which are in reference to G multi-byte structures whose length is not determined by type alone (as s( usually applied to "Strings" as a type).   Cheers!    Keith Cayemberg,   ------------------------------  # Date: Wed, 21 Jul 2004 17:17:39 GMTr& From: jlsue <jefflsxxxz@sbcglobal.net> Subject: Re: OpenVMS security?8 Message-ID: <d89tf0dm1u842cfq5pqm83eb29o6db6shp@4ax.com>  4 On Tue, 20 Jul 2004 21:35:00 -0500, David J Dachtera" <djesys.nospam@comcast.net> wrote:     >t7 >The root vulnerabitity doe snot exist in VMS, TTBOMK. n   Making it very slippery indeed.y   ;-)    --- jlsi0 The preceding message was personal opinion only.6 I do not speak in any authorized capacity for anyone,  and certainly not my employer.- (get rid of the xxxz in my address to e-mail)t   ------------------------------    Date: 21 Jul 2004 08:30:31 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)a* Subject: Re: Power switch for DEC Alpha PC3 Message-ID: <cpftGAWeQgmX@eisner.encompasserve.org>n  h In article <8bdc0085.0407200103.393c4dd9@posting.google.com>, shofu_au@yahoo.com.au (Mark Smith) writes: > H > I am now wanting to have a power switch on my case at the front, but IE > cannot determine the correct power switch to buy.  I have looked ateG > modern PC switches and they are only momentary, and I doubt that theye > can carry the amperage.g  E    On the outside of the box is a tag with the power requirements forl>    your box, somewhere on it should be the maximum amps drawn.  F    Take that down to your hardware store and get a cheap toggle switchE    rated at at least that many amps.  You may need to drill a hole int    the case to install it.   ------------------------------  # Date: Wed, 21 Jul 2004 10:26:19 GMTo& From: Lee Mah <lytmah@telusplanet.net>; Subject: Sending page messages from VMS using a Unix servern, Message-ID: <fxrLc.53099$iw3.22955@clgrps13>  C Currently, I'm paging text messages from VMS by running ftp to send 9 the appropriate file to an NT machine which runs winbeep.4@ However, the NT box has no failover or redundancy.  It failed to> respond shortly after midnight and the situation has been likeG that for several hours.  This is not the first time that the NT machine@D has failed.  Unfortunately, an application running on the VMScluster? is relying on the NT box for passing on semi-critical messages.h; I'm considering requesting that the NT machine be made moree1 reliable but I would prefer a different solution.t> I've found out we have a Unix machine which has Remedy running= on it which sends out pages by using Telalert commands in thei following format:e5   /etc/telalert/bin/telalertc -c $MAIN_PAGER_SERVICE$ 2     -pin $MAIN_PAGER_NR$  -m "$PAGER_TEXT_MESSAGE$ I know very little about Unix.< Is there a way to send a command with parameters from VMS to the Unix machine to page?e   TIA    ------------------------------  # Date: Wed, 21 Jul 2004 11:18:14 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")? Subject: Re: Sending page messages from VMS using a Unix server 6 Message-ID: <00A35262.34C9B16E@SSRL.SLAC.STANFORD.EDU>  U In article <fxrLc.53099$iw3.22955@clgrps13>, Lee Mah <lytmah@telusplanet.net> writes: D >Currently, I'm paging text messages from VMS by running ftp to send: >the appropriate file to an NT machine which runs winbeep.A >However, the NT box has no failover or redundancy.  It failed toe? >respond shortly after midnight and the situation has been likenH >that for several hours.  This is not the first time that the NT machineE >has failed.  Unfortunately, an application running on the VMScluster=@ >is relying on the NT box for passing on semi-critical messages.< >I'm considering requesting that the NT machine be made more2 >reliable but I would prefer a different solution.? >I've found out we have a Unix machine which has Remedy runningk> >on it which sends out pages by using Telalert commands in the >following format:6 >  /etc/telalert/bin/telalertc -c $MAIN_PAGER_SERVICE$3 >    -pin $MAIN_PAGER_NR$  -m "$PAGER_TEXT_MESSAGE$t >I know very little about Unix.e= >Is there a way to send a command with parameters from VMS toe >the Unix machine to page?  H You ought to be able to use RSH or SSH to execute arbitrary commands on 4 the Unix box, if the Unix box is set up to allow it.  O You can also page from the VMS box.  I don't know if there's free software, butsH there are special pager products (RamPAGE) and a pager channel for PMDF.  O If you have an alphanumeric page that works by email (which is what we've got),SM you can send to pagernumber@airtouch.net (or whatever); of course you're then , stuck with email latency, which can be high.   HTH!   -- Alan7   -- nO ===============================================================================p0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025 O ===============================================================================    ------------------------------  # Date: Wed, 21 Jul 2004 13:11:18 GMTe5 From: brad@rabbit.dnsalias.org (Bradford J. Hamilton)s? Subject: Re: Sending page messages from VMS using a Unix serveri. Message-ID: <WXtLc.138003$JR4.49084@attbi_s54>   In article <00A35262.34C9B16E@SSRL.SLAC.STANFORD.EDU>, winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") writes:0V !In article <fxrLc.53099$iw3.22955@clgrps13>, Lee Mah <lytmah@telusplanet.net> writes:E !>Currently, I'm paging text messages from VMS by running ftp to sendd; !>the appropriate file to an NT machine which runs winbeep.x !snip!P !You can also page from the VMS box.  I don't know if there's free software, butI !there are special pager products (RamPAGE) and a pager channel for PMDF.   > We've used (and still use) RamPage, which is a decent product.  P !If you have an alphanumeric page that works by email (which is what we've got),N !you can send to pagernumber@airtouch.net (or whatever); of course you're then- !stuck with email latency, which can be high.s  K We've gone to this solution recently, for notification of environmental and E application events.  I've noticed little or no "email latency"; YMMV.s   !snip!  J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own" K bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-' e0                                          with @"   ------------------------------    Date: 21 Jul 2004 08:37:02 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)n? Subject: Re: Sending page messages from VMS using a Unix server,3 Message-ID: <wm4KGY$mkB3z@eisner.encompasserve.org>a  U In article <fxrLc.53099$iw3.22955@clgrps13>, Lee Mah <lytmah@telusplanet.net> writes:   9 > Unfortunately, an application running on the VMSclustercA > is relying on the NT box for passing on semi-critical messages.o  @    Neither NT not UNIX is going to match the capabilities of VMS$    for dealing with critical events.  A    There are paging packages for VMS, IIRC some are available for B    free.  I suggest you start from the list of free software sitesB    in the faq, which can be found at www.openvms.compaq.com (never7    will be able to remember the HP name for that site).    ------------------------------  % Date: Wed, 21 Jul 2004 12:15:28 -0500e( From: brandon@dalsemi.com (John Brandon)? Subject: Re: Sending page messages from VMS using a Unix server 1 Message-ID: <04072112152859@dscis6-0.dalsemi.com>u  E > Currently, I'm paging text messages from VMS by running ftp to send ; > the appropriate file to an NT machine which runs winbeep.pB > However, the NT box has no failover or redundancy.  It failed to@ > respond shortly after midnight and the situation has been likeI > that for several hours.  This is not the first time that the NT machine.F > has failed.  Unfortunately, an application running on the VMSclusterA > is relying on the NT box for passing on semi-critical messages.l= > I'm considering requesting that the NT machine be made moren3 > reliable but I would prefer a different solution. @ > I've found out we have a Unix machine which has Remedy running? > on it which sends out pages by using Telalert commands in thea > following format:n7 >   /etc/telalert/bin/telalertc -c $MAIN_PAGER_SERVICE$ 4 >     -pin $MAIN_PAGER_NR$  -m "$PAGER_TEXT_MESSAGE$  > I know very little about Unix.> > Is there a way to send a command with parameters from VMS to > the Unix machine to page?I  # Why not use SMTP mail or SMTP_SFF ?l  M With either one you can wrapper the script as needed and run it on VMS ratherl than rely on NT or UNIX.   (SMTP mail)h $ MAIL  
 MAIL> SEND( To:     SMTP%"5556667777@mobile.att.net" Subj:   TESTH Enter your message below. Press CTRL/Z when complete, or CTRL/C to quit: TEST  Exit-  
 (SMTP_SFF)  . $ SMTP_SFF == "$sys$system:tcpip$smtp_sff.exe"
 $ SMTP_SFFI Usage: SMTP_SFF infile_name [-logfile logfile_name] [-loglevel log_level]:K infile_name         : Name of text input file containing SMTP mail to send.mK logfile_name        : Name of text log file to log diagnostics to. (Default. SYS$OUTPUT).G log_level           : Debug log level. 1 or 0 at this time. (Default 0)i! %DCL-E-INVQUAL, invalid qualifierh        D RAMPAGE was also noted - which is a good product - however there areO limitations to any TAP based service - specifically the amount of hardware.  Ifd; you page to different TAP services you have a longer delay.y       J*o*h*n B*r*a*n*d*o*na VMS Systems Administratoru* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  # Date: Wed, 21 Jul 2004 16:27:07 GMTm1 From: "konabear" <konabearg-newsgroups@yahoo.com>p Subject: Solaris to Itanium...< Message-ID: <vPwLc.2435$eu6.1257@newssvr15.news.prodigy.com>  H Interesting Link.   Seems Sum thinks Itanium will be around for a while:  ; http://news.zdnet.co.uk/business/0,39020645,39161196,00.htm    ------------------------------    Date: 21 Jul 2004 01:13:36 -0700) From: its_john_mac@hotmail.com (John Mac)r% Subject: Re: Telnet without timestamp = Message-ID: <fcd7eaa9.0407210013.705ce94e@posting.google.com>i  . Guilty as charged - this is not a vms question  ? FYI  the timestamp is a 32bit package introduced in RFC 1323 tox@ facilitate PAWS. I will give Kermit a go - thanks for your help.   ------------------------------  + Date: Wed, 21 Jul 2004 07:03:58 +0000 (UTC)u From: david20@alpha2.mdx.ac.uk: Subject: Re: Turn-key OpenVMS E-mail, web server solution?) Message-ID: <cdl4gu$rn3$1@news.mdx.ac.uk>    In article <00A35210.02FC02AA@SSRL.SLAC.STANFORD.EDU>, winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") writes:4S >In article <1dKdnUOgd8yQrGjd4p2dnA@igs.net>, "John Smith" <a@nonymous.com> writes:b >>David J Dachtera wrote:yJ >>> Is anyone aware of a company prodiving a turn-key OpenVMS solution forD >>> e-mail, web services and/or both? I thought I followed a link onH >>> Google Groups to one once, but I didn't print anything out from that' >>> site and don't a URL to go back to.  >>> H >>> The time for such seems to be ripening, and if Itanic can repair itsH >>> hull, pump out and put itself to rights, OVMS-I64 may be a practicalI >>> alternative to the trendier UN*X options (which will, undoubtedly, be H >>> available on the soon-to-be ubiquitous x86-64, unlike a certain o.s., >>> we all know of - VMS snoozes, it loses). >>>p
 >>> D.J.D. >> >>J >>Time to DJD to pull together an open-source CD and a self-published book" >>that contains all the foregoing. >> >>;-) (half kidding) >lI >It wouldn't be a bad idea, except that there isn't really an open source"L >mail product.  PMDF costs money.  (It's _worth_ money, but it costs money.) >   J Well MX used to be free. I believe the latest versions cost something like* $50. Again this is available from Process.  
 David Webb Security Team Leader CCSS Middlesex University          t    J >(I was arguing as soon as I first heard of allegedly-really-cheap ItaniumK >servers for HP to sell prepackaged PathWorks server solutions, prepackaged K >webserver solutions, prepackaged mailhub solutions; haven't seen much sign0 >of it happening.) >o >-- Alan >b >-- P >===============================================================================1 > Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDU N > Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056N > Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025P >=============================================================================== >.   ------------------------------  + Date: Wed, 21 Jul 2004 07:22:57 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk: Subject: Re: Turn-key OpenVMS E-mail, web server solution?) Message-ID: <cdl5kh$rn3$3@news.mdx.ac.uk>?   In article <00A3521C.7FA28FC8@SSRL.SLAC.STANFORD.EDU>, winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") writes:ia >In article <40FDCC63.C362F721@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes:e, >>Alan Winston - SSRL Admin Cmptg Mgr wrote: >>> V >>> In article <1dKdnUOgd8yQrGjd4p2dnA@igs.net>, "John Smith" <a@nonymous.com> writes: >>> >David J Dachtera wrote:M >>> >> Is anyone aware of a company prodiving a turn-key OpenVMS solution fornG >>> >> e-mail, web services and/or both? I thought I followed a link onMK >>> >> Google Groups to one once, but I didn't print anything out from thatv* >>> >> site and don't a URL to go back to. >>> >>K >>> >> The time for such seems to be ripening, and if Itanic can repair itsfK >>> >> hull, pump out and put itself to rights, OVMS-I64 may be a practicaleL >>> >> alternative to the trendier UN*X options (which will, undoubtedly, beK >>> >> available on the soon-to-be ubiquitous x86-64, unlike a certain o.s. / >>> >> we all know of - VMS snoozes, it loses).  >>> >>
 >>> >> D.J.D.. >>> >  >>> >nM >>> >Time to DJD to pull together an open-source CD and a self-published book % >>> >that contains all the foregoing.  >>> >s >>> >;-) (half kidding)  >>> L >>> It wouldn't be a bad idea, except that there isn't really an open sourceO >>> mail product.  PMDF costs money.  (It's _worth_ money, but it costs money.)@ >>> M >>> (I was arguing as soon as I first heard of allegedly-really-cheap Itanium$N >>> servers for HP to sell prepackaged PathWorks server solutions, prepackagedN >>> webserver solutions, prepackaged mailhub solutions; haven't seen much sign >>> of it happening.)1 >>B >>Commercial products are not at all incongruous with the proposed >>business plan. >aH >I don't think it's a bad idea at all, especially if you can get ProcessN >to cut you some kind of deal on license costs.  I was responding specificallyN >to "John Smith's" remark about "open-source CD"; you're not going to get PMDF >on one of those.a >n  M The problem with getting Process to cut a deal on PMDF licensing costs is the B history of the takeover of Innosoft (the original owners of PMDF).J Although Process develops and sells PMDF it is under license from SUN (whoF bought Innosoft). Hence any deal must be compatible with that license.L I have been told that is the reason that Process cannot produce a "hobbyist" version of PMDF.  
 David Webb Security Team Leader CCSS Middlesex University        K >(I hope it's clear that this is clarification, not argumentation; when I'mo' >arguing with you, we'll both know it.)7 >1 >-- Alan >  >-- P >===============================================================================1 > Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUeN > Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056N > Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025P >=============================================================================== >    ------------------------------  % Date: Wed, 21 Jul 2004 01:54:59 -0400 ( From: David Froble <davef@tsoft-inc.com>: Subject: Re: Turn-key OpenVMS E-mail, web server solution?* Message-ID: <40FE0533.10007@tsoft-inc.com>   David J Dachtera wrote:   , > Alan Winston - SSRL Admin Cmptg Mgr wrote: > T >>In article <1dKdnUOgd8yQrGjd4p2dnA@igs.net>, "John Smith" <a@nonymous.com> writes: >> >>>David J Dachtera wrote: >>>eJ >>>>Is anyone aware of a company prodiving a turn-key OpenVMS solution forD >>>>e-mail, web services and/or both? I thought I followed a link onH >>>>Google Groups to one once, but I didn't print anything out from that' >>>>site and don't a URL to go back to.  >>>>H >>>>The time for such seems to be ripening, and if Itanic can repair itsH >>>>hull, pump out and put itself to rights, OVMS-I64 may be a practicalI >>>>alternative to the trendier UN*X options (which will, undoubtedly, berH >>>>available on the soon-to-be ubiquitous x86-64, unlike a certain o.s., >>>>we all know of - VMS snoozes, it loses). >>>>
 >>>>D.J.D. >>>> >>>nK >>>Time to DJD to pull together an open-source CD and a self-published bookp# >>>that contains all the foregoing.s >>>c >>>;-) (half kidding)_ >>>.J >>It wouldn't be a bad idea, except that there isn't really an open sourceM >>mail product.  PMDF costs money.  (It's _worth_ money, but it costs money.)  >>K >>(I was arguing as soon as I first heard of allegedly-really-cheap ItaniumoL >>servers for HP to sell prepackaged PathWorks server solutions, prepackagedL >>webserver solutions, prepackaged mailhub solutions; haven't seen much sign >>of it happening.)s >> > B > Commercial products are not at all incongruous with the proposed > business plan. >  > D.J.D. >   I Why is it that many people in education environments place such value on sQ freeware, and want everything to be free,  ....  until you want a copy of one of   the books they write?r   Dave   -- t4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Roadg Vanderbilt, PA  15486.   ------------------------------  # Date: Wed, 21 Jul 2004 11:06:58 GMTtL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr"): Subject: Re: Turn-key OpenVMS E-mail, web server solution?6 Message-ID: <00A35260.A1649B80@SSRL.SLAC.STANFORD.EDU>  J In article <cdl4gu$rn3$1@news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk writes: >In article <00A35210.02FC02AA@SSRL.SLAC.STANFORD.EDU>, winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") writes:T >>In article <1dKdnUOgd8yQrGjd4p2dnA@igs.net>, "John Smith" <a@nonymous.com> writes: >>>David J Dachtera wrote:K >>>> Is anyone aware of a company prodiving a turn-key OpenVMS solution fordE >>>> e-mail, web services and/or both? I thought I followed a link onrI >>>> Google Groups to one once, but I didn't print anything out from that ( >>>> site and don't a URL to go back to. >>>>I >>>> The time for such seems to be ripening, and if Itanic can repair itssI >>>> hull, pump out and put itself to rights, OVMS-I64 may be a practical.J >>>> alternative to the trendier UN*X options (which will, undoubtedly, beI >>>> available on the soon-to-be ubiquitous x86-64, unlike a certain o.s. - >>>> we all know of - VMS snoozes, it loses).t >>>> >>>> D.J.D.  >>>  >>>iK >>>Time to DJD to pull together an open-source CD and a self-published book # >>>that contains all the foregoing.0 >>>  >>>;-) (half kidding)p >>J >>It wouldn't be a bad idea, except that there isn't really an open sourceM >>mail product.  PMDF costs money.  (It's _worth_ money, but it costs money.)i >> >nK >Well MX used to be free. I believe the latest versions cost something liker+ >$50. Again this is available from Process.a  I Is it?  Last I looked, it was $500, and it wasn't available from Process. I (If it's $50 now, I'll buy it for my home VMS box; can't justify that foro $500.)   -- Alan- -- nO =============================================================================== 0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-30565M  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025>O ===============================================================================m   ------------------------------   Date: 21 Jul 2004 12:02:16 GMT, From: bill@gw5.cs.uofs.edu (Bill Gunshannon): Subject: Re: Turn-key OpenVMS E-mail, web server solution?* Message-ID: <2m74a7Fj2norU1@uni-berlin.de>  * In article <40FE0533.10007@tsoft-inc.com>,+ 	David Froble <davef@tsoft-inc.com> writes:p > K > Why is it that many people in education environments place such value on  , > freeware, and want everything to be free,   C Because most of us have smaller equipment budgets than the business1( lunch budget at profit making companies.    K >                                    ....  until you want a copy of one of   > the books they write?e  C Publishers set the price (and make most of the profit) on books.  AoE number of professors here have books currently in print.  Unless your C Stephen King or Robert Ludlum your lucky to get beer money from thepC sales of your book. Fiction and politics sell much better than text  books.   bill   -- 'J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   D   ------------------------------  % Date: Wed, 21 Jul 2004 12:23:22 -0400i# From: "John Smith" <a@nonymous.com>f: Subject: Re: Turn-key OpenVMS E-mail, web server solution?, Message-ID: <9_mdnaX13uTgBWPdRVn-sA@igs.net>  , "Alan Winston - SSRL Admin Cmptg Mgr" wrote:9 > In article <40FE0533.10007@tsoft-inc.com>, David Froble- > <davef@tsoft-inc.com> writes:u >> David J Dachtera wrote: >>. >>> Alan Winston - SSRL Admin Cmptg Mgr wrote: >>>w> >>>> In article <1dKdnUOgd8yQrGjd4p2dnA@igs.net>, "John Smith" >>>> <a@nonymous.com> writes:> >>>> >>>>> David J Dachtera wrote:p >>>>>i@ >>>>>> Is anyone aware of a company prodiving a turn-key OpenVMSA >>>>>> solution for e-mail, web services and/or both? I thought IlG >>>>>> followed a link on Google Groups to one once, but I didn't printnA >>>>>> anything out from that site and don't a URL to go back to.e >>>>>>G >>>>>> The time for such seems to be ripening, and if Itanic can repair>E >>>>>> its hull, pump out and put itself to rights, OVMS-I64 may be a>F >>>>>> practical alternative to the trendier UN*X options (which will,E >>>>>> undoubtedly, be available on the soon-to-be ubiquitous x86-64,TE >>>>>> unlike a certain o.s. we all know of - VMS snoozes, it loses).  >>>>>>
 >>>>>> D.J.D.r >>>>>> >>>>>d: >>>>> Time to DJD to pull together an open-source CD and a: >>>>> self-published book that contains all the foregoing. >>>>>, >>>>> ;-) (half kidding) >>>>>CF >>>> It wouldn't be a bad idea, except that there isn't really an open >>>> source B >>>> mail product.  PMDF costs money.  (It's _worth_ money, but it >>>> costs money.) >>>>F >>>> (I was arguing as soon as I first heard of allegedly-really-cheap >>>> ItaniumC >>>> servers for HP to sell prepackaged PathWorks server solutions,oD >>>> prepackaged webserver solutions, prepackaged mailhub solutions; >>>> haven't seen much signn >>>> of it happening.) >>>> >>>cD >>> Commercial products are not at all incongruous with the proposed >>> business plan. >>> 
 >>> D.J.D. >>>, >>B >> Why is it that many people in education environments place suchF >> value on freeware, and want everything to be free,  ....  until you. >> want a copy of one of the books they write? > E > Gosh, I don't know. It might have to do with our neither having biga	 > budgetsaG > at work with which to buy stuff (so we're interested in free software  > so wecF > can have software at all),  or ogetting paid big money, so potential > incomeE > from books is nice.  (In my particular case, I was obliged to buy a  > homeH > VMS box in order to write the Webserver book.  After I've made as muchB > money on the book as it cost me - what with equipment purchases,
 > vacationH > time, and a bit of unpaid leave - I might consider making it availableG > free.  I do give away the runoff-to-HTML translator I adapted from DPn% > Murphy's runoff-to-RTF translator.)  > E > However, some people in education do put their books on the Web fori > free. > download, like the economist David Friedman. >mC > In the meantime, it was John Smith - not in education as far as Ia > know -D > who introduced the idea of an open source book and disk, and all I > saidA > was that you'd need to spend money on the mail server software. 	 > Nothing G > wrong with packaged solutions where you have to spend _some_ money on  > the software.     $ Correct - I'm not in the .edu space.  K Assume for a moment that there is some economic justification for D.J.D. orMI somebody else in investing the time/effort/money in this sort of thing to L help HP sell more VMS systems (I leave it to the reader of this paragraph toL draw their own conclusions as to just why a $60B corporation needs people toK do things for them for free or nearly so when it comes to VMS), a differentr: approach to the question is a freeware/trialware combo CD.  H Perhaps all the applications on this CD are indeed freeware (web server,F paging, etc) except for the e-mail server and maybe a message queueingH server and AV software. Say that the e-mail software is PDMF and it is aF 30-120 day free trial copy (not unheard of in many commercial softwareI organizations) and that the use beyond the trial period requires payment.T  H D.J.D. (or whomever producing the freeware CD and docs) cuts a deal withL Process, et. al. that when the PDMF/VMS user who has used the trialware buysL a licensed copy the 'key code' that was used on the trial copy identifies itK as a copy that was distributed via the 'freeware CD' and as such D.J.D. (orsK whomever) is entitled to a commission on the sale once payment is received. I Same could hold true for the other commercial packages on the CD. (I knowoD senior IBM mgrs. for MQ Series and WebSphere, but maybe BEA would be interested, etc....)  L The value in the whole process is that a person or corporation that might beK interested in trying VMS has on one CD a complete replacement for MicrosofteE IIS, Exchange, MSMQ, an antivirus package. The only thing is that the J commercial trialware has to have a sufficiently long free use to allow forL learning curve issues for both the application and the new VMS user becoming# familiar with the operating system.o   ------------------------------  % Date: Wed, 21 Jul 2004 17:21:52 +0200x2 From: "Lars Holmstrm" <lars.holmstrom@flysta.net>$ Subject: VMS/Cluster vs. Polyserve ?+ Message-ID: <cdm1nh$37k$1@news.utel.net.ua>o  J Has anyone seen a comparison between OpenVMS/Cluster and the Polyserve for Linux ?a /Larsa   ------------------------------  % Date: Wed, 21 Jul 2004 10:44:37 -0700o+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>c( Subject: Re: VMS/Cluster vs. Polyserve ?' Message-ID: <40FEAB85.1010202@MMaz.com>    Lars Holmstr=F6m wrote:s  J >Has anyone seen a comparison between OpenVMS/Cluster and the Polyserve f= or >Linux ? > =20  >fG Well, their web site polyserve.com has many white papers and product=20tJ datasheets, but just skimming the surface and though it may not be full=20I VMS Clustering, it does appear to be pretty interesting for Linux - It=20iD looks like a SAN solution, the servers see a single image, with a=20K distributing locking mechanism, no doubt using daemons on each server to=20mH accomplish/coordinate the synchronization and locking for each system...   Barry    --=20n  > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                       =20s   ------------------------------  % Date: Wed, 21 Jul 2004 12:49:52 +0100u- From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>d' Subject: [OT] Haggis in comp.os.vms :-) * Message-ID: <2m73loFjnlaoU1@uni-berlin.de>   VAXman- wrote:g > In article <40FB3802.E4E6FB38@applied-synergy.com>, Chris Scheers <chris@applied-synergy.com> writes:c > # >>VAXman-, @SendSpamHere.ORG wrote:h >>N >>>Having just returned from my >fortnight holiday in the UK, I just wanted toN >>>publicly state what a pleasure it was to *finally* meetup with Richard evenN >>>if it was for only an hour.  It looks as if the third time was a charm this >>>time out. >>>rN >>>My wife and I enjoyed beer, pizza and pleasant conversation at the Dicken'sI >>>Inn near Tower Bridge with Richard.  Thanks.  VMS people are the best.a >> >>D >>Did you get to try the famous English corn and tuna pizza?  <grin> >  > L > I prefer tuna in raw form such as sashimi, or extremely rare and seared asB > an ahi tuna steak.  If cooked beyond extremely rare, tuna sucks. > L > No.  But I did order up haggis for a meal one evening and I enjoyed it.  I) > even ordered it again for another meal.o  H Well, this must *surely* be the first mention of "haggis" in the history of com.os.vms !r  B As Brian discovered, haggis is actually very nice, and pretty muchC formed the staple diet for us poor people growing up in Scotland in-B the 1950's and 1960's.  It was cheap (consisting basically of the  "leftover" parts of a sheep).>  H Reminds me of an amusing episode back in, if I remember correctly, 1984.B I was working as VMS system manager etc. in the European MolecularF Biology Laboratory (see http://www.embl-heidelberg.de), in Heidelberg,H Germany (one of the most *perfect* places to work in the universe) where I worked for 14 years.  H There was a fairly large contingent of Scots (and just about every otherE nationality on the planet), and we decided to organise a Burns night,oE in celebration of Robert Burns, Scotland's best-known poet ever (withsC maybe one exception :-).  We had organised Scottish country dancing G lessons for weeks in advance, ordered much fine malt whisky, and a huge-B haggis from Scotland, the haggis to be delivered on the day of theF celebration.  The delivery was meant to arrive at Frankfurt airport atG 11:00.  It got to 12 and still no sign of the haggis ... then 13:00 ... , then 14:00 ... then 15:00 ... still no sign.  @ Eventually about 16:00, we get a phone call from the dear German> customs authorities at Frankfurt to say that this strange meat? product had been quarantined, and would not be allowed to entern> the country (due to strict import laws concerning meat stuff).> We dispatched the Lab driver up to Frankfurt, and he must have; averaged about 200 Km/hour (perfectly legal on the Deutschec= Autobahnen :-).  This guy could talk his way out of anything,o: and he managed to convince the authorities that this was a8 consignment of biological material for research purposesB (remember, this was the European Molecular Biology Lab after all).  @ So the haggis was duly stamped on the outer skin "For Biological> Research Purposes Only, and driven at break-neck speed back to; Heidelberg, where the canteen chef was eagerly awaiting it.,  @ The haggis was cooked just in time as the first whisky was being@ consumed, and a truly wonderful time was had by all.  And no-oneA got sick from eating this "for biological research purposes only"e product :-)t  : To this day, the lab still organises an annual Burns night: celebration, and it is the only one of the "national" ones= to survive, all the others (Swedish, Dutch, English, USofA'anr- etc. etc.) having long fallen by the wayside.i  ? And by the way, the Laboratory has just appointed a fellow Scotf< as the new Director General, a most prestigious appointment;< I'm *sure* it's got something to do with his performances as( Holy Wullie at the many Burns nights :-)   Oh yeah, a joke: Q: how do you make haggis ?tB A: You get a sheep, take all the wool off, stuff your hand and armB down its throat, grab the inside of the stomach wall, pull as hardB as you can, thus turning the sheep inside out, cook, and consume !  : Burns even wrote a most eloquent poem in honour of haggis.8 "To a Haggis" - "Fair fa' yer honest, sonsie face, great! chieftain o' the pudden race ..."r   Enough ... now, stovies :-)   	 Roy Omondd Blue Bubble Ltd.   ------------------------------  # Date: Wed, 21 Jul 2004 13:16:42 GMTl" From:   VAXman-  @SendSpamHere.ORG+ Subject: Re: [OT] Haggis in comp.os.vms :-)e0 Message-ID: <00A3528B.E8395505@SendSpamHere.ORG>  Z In article <2m73loFjnlaoU1@uni-berlin.de>, Roy Omond <Roy.Omond@BlueBubble.UK.Com> writes: >VAXman- wrote:  {...snip...}M >> I prefer tuna in raw form such as sashimi, or extremely rare and seared asrC >> an ahi tuna steak.  If cooked beyond extremely rare, tuna sucks.t >> eM >> No.  But I did order up haggis for a meal one evening and I enjoyed it.  I1* >> even ordered it again for another meal. >cI >Well, this must *surely* be the first mention of "haggis" in the historyT >of com.os.vms ! > C >As Brian discovered, haggis is actually very nice, and pretty much D >formed the staple diet for us poor people growing up in Scotland inC >the 1950's and 1960's.  It was cheap (consisting basically of the a >"leftover" parts of a sheep).  F I think it is this knowledge that tends to turn most people off to itsF consumption.  I'm very open minded when it comes to foods.  I will tryF most anything once.  If I don't like it, at least I have an "informed" opinion. ;)<  E BTW, my haggis was served to me with drambuie, not a fine malt.  Mrs.MF VAXman will attest to my state of being for the rest of the pub crawlsF later that evening after consuming the drambuie.  I have no recollect-C tion of our returning to the guest house in Edinburgh that evening.n  F I highly recommend the hike up the cliff starting in the park near the@ Holyroodhouse Palace.  Quite a workout and well worth the climb. -- sB http://www.legacy-2000.com  for the *best* OpenVMS system securityC                             solutions that others only claim to be.t -- tK VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM.            -5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Wed, 21 Jul 2004 14:53:28 +0100s- From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>r+ Subject: Re: [OT] Haggis in comp.os.vms :-)m* Message-ID: <2m7athFjr4rfU1@uni-berlin.de>   VAXman- wrote:   >i > [...snip...] > H > I think it is this knowledge that tends to turn most people off to itsH > consumption.  I'm very open minded when it comes to foods.  I will tryH > most anything once.  If I don't like it, at least I have an "informed"
 > opinion. ;)a   Excellent !a  G > BTW, my haggis was served to me with drambuie, not a fine malt.  Mrs.rH > VAXman will attest to my state of being for the rest of the pub crawlsH > later that evening after consuming the drambuie.  I have no recollect-E > tion of our returning to the guest house in Edinburgh that evening.i   :-)   H > I highly recommend the hike up the cliff starting in the park near theB > Holyroodhouse Palace.  Quite a workout and well worth the climb.  B That's Arthur's Seat, the place we kids used to get shipped to forB free entertainment, and to wear us out in the long summer holidaysE (I'm from Edinburgh ...).  Yep, Edinburgh's a *great* place to visit,eD and a *fantastic* place to live.  So, when are you moving, Brian :-)* There's still some VMS around in the area.  A You'd really love the music scene there (though, sad to say we'veD@ recently lost 2 of the world's finest folk musicians, Tony Cuffe: and Johnny Cunningham, both dead at such a young age, both  younger than myself - very sad).   ------------------------------  % Date: Wed, 21 Jul 2004 18:15:05 +0200T* From: Paul Sture <nospam@sture.homeip.net>+ Subject: Re: [OT] Haggis in comp.os.vms :-)l* Message-ID: <2m7j4eFjj7d7U1@uni-berlin.de>   Roy Omond wrote: > VAXman- wrote: > E >> In article <40FB3802.E4E6FB38@applied-synergy.com>, Chris Scheers e& >> <chris@applied-synergy.com> writes: >>% >>> VAXman-, @SendSpamHere.ORG wrote:  >>>oG >>>> Having just returned from my >fortnight holiday in the UK, I just e >>>> wanted toD >>>> publicly state what a pleasure it was to *finally* meetup with  >>>> Richard evennF >>>> if it was for only an hour.  It looks as if the third time was a  >>>> charm thisa >>>> time out. >>>>H >>>> My wife and I enjoyed beer, pizza and pleasant conversation at the 
 >>>> Dicken'sRK >>>> Inn near Tower Bridge with Richard.  Thanks.  VMS people are the best.a >>>s >>>  >>>nF >>> Did you get to try the famous English corn and tuna pizza?  <grin> >> >> >>D >> I prefer tuna in raw form such as sashimi, or extremely rare and  >> seared asC >> an ahi tuna steak.  If cooked beyond extremely rare, tuna sucks.k >>G >> No.  But I did order up haggis for a meal one evening and I enjoyed )	 >> it.  Ir* >> even ordered it again for another meal. >  > J > Well, this must *surely* be the first mention of "haggis" in the history > of com.os.vms !r > D > As Brian discovered, haggis is actually very nice, and pretty muchE > formed the staple diet for us poor people growing up in Scotland in=D > the 1950's and 1960's.  It was cheap (consisting basically of the  > "leftover" parts of a sheep).i > J > Reminds me of an amusing episode back in, if I remember correctly, 1984.D > I was working as VMS system manager etc. in the European MolecularH > Biology Laboratory (see http://www.embl-heidelberg.de), in Heidelberg,J > Germany (one of the most *perfect* places to work in the universe) where > I worked for 14 years. > J > There was a fairly large contingent of Scots (and just about every otherG > nationality on the planet), and we decided to organise a Burns night, G > in celebration of Robert Burns, Scotland's best-known poet ever (withSE > maybe one exception :-).  We had organised Scottish country dancinglI > lessons for weeks in advance, ordered much fine malt whisky, and a huge:D > haggis from Scotland, the haggis to be delivered on the day of theH > celebration.  The delivery was meant to arrive at Frankfurt airport atI > 11:00.  It got to 12 and still no sign of the haggis ... then 13:00 ...,. > then 14:00 ... then 15:00 ... still no sign. > B > Eventually about 16:00, we get a phone call from the dear German@ > customs authorities at Frankfurt to say that this strange meatA > product had been quarantined, and would not be allowed to enterr@ > the country (due to strict import laws concerning meat stuff).@ > We dispatched the Lab driver up to Frankfurt, and he must have= > averaged about 200 Km/hour (perfectly legal on the Deutscheo? > Autobahnen :-).  This guy could talk his way out of anything,s< > and he managed to convince the authorities that this was a: > consignment of biological material for research purposesD > (remember, this was the European Molecular Biology Lab after all). > B > So the haggis was duly stamped on the outer skin "For Biological@ > Research Purposes Only, and driven at break-neck speed back to= > Heidelberg, where the canteen chef was eagerly awaiting it.r > B > The haggis was cooked just in time as the first whisky was beingB > consumed, and a truly wonderful time was had by all.  And no-oneC > got sick from eating this "for biological research purposes only"n
 > product :-)  > < > To this day, the lab still organises an annual Burns night< > celebration, and it is the only one of the "national" ones? > to survive, all the others (Swedish, Dutch, English, USofA'anf/ > etc. etc.) having long fallen by the wayside.T > A > And by the way, the Laboratory has just appointed a fellow Scot.> > as the new Director General, a most prestigious appointment;> > I'm *sure* it's got something to do with his performances as* > Holy Wullie at the many Burns nights :-) >  > Oh yeah, a joke: > Q: how do you make haggis ? D > A: You get a sheep, take all the wool off, stuff your hand and armD > down its throat, grab the inside of the stomach wall, pull as hardD > as you can, thus turning the sheep inside out, cook, and consume ! >p   More haggis facts at:V  1 http://haggishunt.scotsman.com/haggisclopedia.cfm    :-)a  < > Burns even wrote a most eloquent poem in honour of haggis.: > "To a Haggis" - "Fair fa' yer honest, sonsie face, great# > chieftain o' the pudden race ..."  >  > Enough ... now, stovies :-)n >  > Roy Omond  > Blue Bubble Ltd.     -- a
 Paul Sture  - OS X: "It's Unix, Jim, but not as we know it"r   ------------------------------  % Date: Wed, 21 Jul 2004 13:15:51 -0400 # From: "John Smith" <a@nonymous.com>cK Subject: [OT]: Microsoft Patent Cross License - Open Source Software Impact', Message-ID: <QOadnZF04uxXOWPdRVn-hQ@igs.net>  8 http://www.newsforge.com/article.pl?sid=04/07/19/2315200   ------------------------------   End of INFO-VAX 2004.401 ************************