1 INFO-VAX	Sun, 19 Aug 2001	Volume 2001 : Issue 460       Contents: Re: alphastion 233 & firmware  check this outE Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption) E Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption) E Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption) E Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption)  Re: DEC-C float & double Re: DEC-C float & double Re: tpcip upgrade  Re: tpcip upgrade  RE: tpcip upgrade  VAX: DEC-C float & double  ZLX2-E (TGA2) dial settings?  F ----------------------------------------------------------------------  % Date: Sat, 18 Aug 2001 19:11:22 -0500 / From: Chris Scheers <chris@applied-synergy.com> & Subject: Re: alphastion 233 & firmware3 Message-ID: <3B7F042A.43EC7E48@applied-synergy.com>    martin baechtel wrote: > L > I can't get my 4/233 to flash.  The firmware is v5.9 the system is set nowF > to ARC v5.7  and I am trying to set it to SRM.  J6 is set to enable.@ > It doses read the CD then a flash on the screen then it hangs.% > I can flash my 4/166 to ARC or SRM.   C Assuming you are referring to an AlphaStation 255/233, it is a half D flash system.  If you are referring to an Alphastation 200 4/233, itC could be either full flash (4 flash chips) or half flash (2 chips).   E A full flash system can hold both ARC and SRM.  A half flash can only  hold one at a time.   G To change a half flash system from one system to the other, boot up the E flash utility.  Then, instead of simply entering UPDATE, enter UPDATE ! ARC or UPDATE SRM as appropriate.   C If this does not answer your problem, please post more information.   G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  # Date: Sun, 19 Aug 2001 14:58:42 GMT  From: uucmqj@zwallet.com Subject: check this out 9 Message-ID: <CuQf7.17802$sv1.100393@wagner.videotron.net>   a DID YOU KNOW IF YOU DONT KNOW JESUS AS YOUR PERSONAL SAVIOR YOU WILL GO TO HELL. REPENT YOURSELF!  http://www.radiovie.qc.ca    ------------------------------  % Date: Sun, 19 Aug 2001 09:33:21 +0200 , From: "Bart Zorn" <B.Zorn@TrueBit.nospam.nl>N Subject: Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption); Message-ID: <3b7f6c10$0$58545$e4fe514c@newszilla.xs4all.nl>   7 "Simon Dodsley" <simon@sdinfotech.com> wrote in message 4 news:3b7efb56_1@mk-nntp-1.news.uk.worldonline.com..., > Yeah, we all want to upgrade to HSG80's !! > G > That wonderful bit of kit that, currently, can only have its firmware K > upgraded by shutting down every system connected to it.  What a wonderful I > way to keep your Wildfires running on 24x7 - Sorry Mr Business Critical F > Customer, your 24x7 solution you paid an arm and a leg for has to beK > shutdown to upgrade the disk arrays. What's that you say, can't you do it  onI > the fly like the old controllers? No, sorry, not yet, Compaq screwed up 
 > again... >  > K This must be a new feature of the HSG80's. I have upgraded pairs of HSG80's I without the application knowing it. Note: OpenVMS DID notice: some of the E disks failed over to another path. But the applications didn't blink.   	 Bart Zorn    ------------------------------  # Date: Sun, 19 Aug 2001 07:44:23 GMT  From: Dirk Munk <munk@home.nl>N Subject: Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption)' Message-ID: <3B7F6E56.34C5787C@home.nl>    Bart Zorn wrote:  9 > "Simon Dodsley" <simon@sdinfotech.com> wrote in message 6 > news:3b7efb56_1@mk-nntp-1.news.uk.worldonline.com.... > > Yeah, we all want to upgrade to HSG80's !! > > I > > That wonderful bit of kit that, currently, can only have its firmware M > > upgraded by shutting down every system connected to it.  What a wonderful K > > way to keep your Wildfires running on 24x7 - Sorry Mr Business Critical H > > Customer, your 24x7 solution you paid an arm and a leg for has to beM > > shutdown to upgrade the disk arrays. What's that you say, can't you do it  > onK > > the fly like the old controllers? No, sorry, not yet, Compaq screwed up  > > again... > >  > > M > This must be a new feature of the HSG80's. I have upgraded pairs of HSG80's K > without the application knowing it. Note: OpenVMS DID notice: some of the G > disks failed over to another path. But the applications didn't blink.   I We did an upgrade too abouth 2 months ago, without having to shutdown the J systems. But we didn't do the upgrade to 8.6 yet, so I can't comment about that.      >  >  > Bart Zorn    ------------------------------  % Date: Sun, 19 Aug 2001 12:54:58 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> N Subject: Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption), Message-ID: <3B7FEF62.8526D825@videotron.ca>   Mark Hemker wrote:G > I upgraded a pair of HSG80s two weeks ago from V8.5F to V8.6F without H > stopping anything.  VMS just failed over the paths from one controllerE > to the other and then failed them all back to the first controller.     L How is the failover implemented ? Are there two buses that go to the storage, array each going to a separate controller  ?  L If VMS becomes aware of the "failure" of the first controller and falls backK on the backup controller, that means that VMS has some smarts into it to do  that, correct ?   M Is it possible that Windows would not have such smarts, and hence the need to K shutdown and reboot Windows when you do that type of upgrade on the array ? J (which would explain such documentation that comes with the array which is; probably sold to more Windows systems than to VMS systems).    ------------------------------  # Date: Sun, 19 Aug 2001 17:39:58 GMT $ From: Scott Vieth <svieth@wi.rr.com>N Subject: Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption)) Message-ID: <3B7FFAAA.AFCCEE68@wi.rr.com>   M 1) When connecting a VMS system to a SAN, it is customary to install two HBAs O (minimum) in your AlphaServer.  Each HBA then goes to one of two switches.  The = host ports on the HSG80s are connected to these two switches.   P This connection method provides (at the OS level) four paths to each volume.  IfP you do a SHOW DEV DG/FULL, you'll be able to see all four paths for each volume.  N When one of the HSGs is shut down, two of the paths go away.  If VMS was usingK one of those paths, the device switches over to one of the remaining active  paths.  Q 2) In the NT world, this type of connectivity requires a piece of software called P SecurePath.  I've used it and it works very well.  The NT server sees four pathsH to any given disk volume (when two HBAs are installed in the NT server).  M If you connected an NT server into a SAN with only one HBA, then you probably ; have to shut down that system during HSG firmware upgrades.    -Scott   JF Mezei wrote:    > Mark Hemker wrote:I > > I upgraded a pair of HSG80s two weeks ago from V8.5F to V8.6F without J > > stopping anything.  VMS just failed over the paths from one controllerG > > to the other and then failed them all back to the first controller.  > N > How is the failover implemented ? Are there two buses that go to the storage. > array each going to a separate controller  ? > N > If VMS becomes aware of the "failure" of the first controller and falls backM > on the backup controller, that means that VMS has some smarts into it to do  > that, correct ?  > O > Is it possible that Windows would not have such smarts, and hence the need to M > shutdown and reboot Windows when you do that type of upgrade on the array ? L > (which would explain such documentation that comes with the array which is= > probably sold to more Windows systems than to VMS systems).    ------------------------------  % Date: Sun, 19 Aug 2001 16:29:52 +0200 . From: "Jesper Naur" <jesper.naur@post.tele.dk>! Subject: Re: DEC-C float & double = Message-ID: <3b7fcc2f$0$97077$edfadb0f@dspool01.news.tele.dk>   8 JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message& news:3B7F6C46.DBB2A94E@videotron.ca... > Ok, DEC-C on a VAX.  >  > 8 > can someone explain to me why/how the following works: >  > float sky = 25.9 > double water = 309.72673734 . > printf(" sky=%f  water=%f \n", sky, water ); > L > does printf have the smarts to analyse the contents of memory to determine ifF > it is stored as a normal float or as a double ? Or does the compilerD > automatically generate code that converts "water" to a float while building, > the argument list to be passed to printf ? > L It's the other way round, it will convert the float sky into a double before; printing, as illustrated by the machine code listing below:   4   674           void test( float sky, double water )5                                           0018  TEST: I                                     0000  0018          .entry  TEST,^m<> E                               5E 04 C2    001A          subl2   #4,sp      675           { ,   676              printf(" sky=%f ", sky );K                            7E 04 AC 56    001D          cvtfd   4(ap),-(sp)  ;*** pick up 'sky'E                               DC AF 9F    0021          pushab  $CODE =                     00000000* EF 03 FB    0024          calls  #3,DECC$DPRINTF   0   677              printf(" water=%f ", water );K                            7E 08 AC 70    002B          movd    8(ap),-(sp)  ;*** pick up 'water'G                               D7 AF 9F    002F          pushab  $CODE+9 =                     00000000* EF 03 FB    0032          calls  #3,DECC$DPRINTF      678           } ;                                     04    0039          ret     L It uses 'cvtfd' (convert floating to double) for the float 'sky', and 'movd'% (move double) for the double 'water'.     J > Arer there any gotcha's to be aware of when working with both types with some > interchanges going on ?  > K > (one part, which does fancy calculations, requires the added precision of I > double, while the other part was written separately and doesn't require  thatA > precision and just outputs the data (DXF output of a 3d model).  > G As long as you are sure, that whenever a double value is converted to a J float value, you can afford the loss of precision and range, you should beE OK. That would for example mean, that when the fancy calculations are J complete, and the (double) results fit in float variables, you can convert them without problems.       Best regards     Jesper Naur    ------------------------------  % Date: Sun, 19 Aug 2001 12:59:32 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> ! Subject: Re: DEC-C float & double , Message-ID: <3B7FF074.7C4CA77E@videotron.ca>   Jesper Naur wrote:N > It's the other way round, it will convert the float sky into a double before= > printing, as illustrated by the machine code listing below:   D Thanks ! I wouldn't have though of that. But it makes sense I guess.  J Now, how does the compiler know it should convert a float to a double in aK routine such as printf with undefined arguments ? Does "printf" get special A treatment because it is so closely associated with the language ?   L Or doers the compiler automatically convert any float to double when passing) to a routine whose argument are unknown ?    ------------------------------  % Date: Sat, 18 Aug 2001 19:12:11 -0400 ' From: Howard S Shubs <howard@shubs.net>  Subject: Re: tpcip upgrade< Message-ID: <howard-54F338.19121018082001@enews.newsguy.com>  = In article <34e80f93.0108181211.29c8426f@posting.google.com>, 2  techwebsite@netscape.net (Michael Angello) wrote:  G > Is it possible to upgrade from TCP/IP Services V5.0 (AXP) to a higher G > version (i.e. V5.1) without upgrading the OS (currently using V7.2)?, A > and if so, how may I accomplish that task, looking what kind of  > information?  J Not sure what you're asking.  You don't need to upgrade the OS to upgrade  TCPIP. --   Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  % Date: Sun, 19 Aug 2001 09:47:38 +0200 2 From: "Thomas H. Pauli" <thomaspauli@arcormail.de> Subject: Re: tpcip upgrade+ Message-ID: <3B7F6F1A.8060303@arcormail.de>   > The best way to find out tetse compatibility issues is to lookC into the release notes or the product announcements of TCP/IP V5.1! 3 -> http://www.openvms.compaq.com/network/tcpip.html 7 You'll find there that no OpenVMS upgrade is necessary.  Regards,   Thomas   Michael Angello wrote:   > Excuse me, > G > Is it possible to upgrade from TCP/IP Services V5.0 (AXP) to a higher G > version (i.e. V5.1) without upgrading the OS (currently using V7.2)?, A > and if so, how may I accomplish that task, looking what kind of  > information? > E > I'd like to upgrade those services because I had some problems with  > the systemF > due to It crashes sometimes due to INCONSTATE, Inconsistent I/O data > base. E > I was looking for issues about it on DejaNews, and accordinf what I , > read address me to upgrade such a service. > F > Please, really excuse me again for such matter, but I just found theG > Install docs regarinf OS and TCPIP Services, but I did not see such a ? > scenery or any comments about it are appreciated in advanced.  >  >  > Thanks >  > Mike >      --  9 Thomas H. Pauli, Hammersteinstr.19, 14199 Berlin, Germany    ------------------------------  % Date: Sun, 19 Aug 2001 04:06:45 -0500 * From: WILLIAM WEBB <WWEBB1@email.usps.gov> Subject: RE: tpcip upgrade- Message-ID: <0033000032673428000002L082*@MHS>   1 =0A(I'm answering this in the middle of the night 0 after getting paged into the office disclaimer),0  but off the top of my head, I see no reason why- you couldn't upgrade without an OS upgrade- I . haven't done that specific version-to-version,/ but I've done it with previous versions without # having to do an OS version upgrade.   . All you need is the right saveset- one usually$ gets this from the distribution CDs.   WWWebb   > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNET ) > Sent: Saturday, August 18, 2001 4:13 PM F > To: Webb, William W - Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET > Subject: tpcip upgrade >  >  > Excuse me, > H > Is it possible to upgrade from TCP/IP Services V5.0 (AXP) to a higher=  H > version (i.e. V5.1) without upgrading the OS (currently using V7.2)?,=  A > and if so, how may I accomplish that task, looking what kind of  > information? > E > I'd like to upgrade those services because I had some problems with  > the systemF > due to It crashes sometimes due to INCONSTATE, Inconsistent I/O data > base. E > I was looking for issues about it on DejaNews, and accordinf what I , > read address me to upgrade such a service. > F > Please, really excuse me again for such matter, but I just found theH > Install docs regarinf OS and TCPIP Services, but I did not see such a=  ? > scenery or any comments about it are appreciated in advanced.  >  >  > Thanks >  > Mike >=   ------------------------------  % Date: Sun, 19 Aug 2001 03:35:36 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> " Subject: VAX: DEC-C float & double, Message-ID: <3B7F6C46.DBB2A94E@videotron.ca>   Ok, DEC-C on a VAX.   M As I understand it, "float" generates one type of floating point number while T "double" generates a D floating point (by default, or optionally a G floting point).  6 can someone explain to me why/how the following works:   float sky = 25.9 double water = 309.72673734 , printf(" sky=%f  water=%f \n", sky, water );  M does printf have the smarts to analyse the contents of memory to determine if D it is stored as a normal float or as a double ? Or does the compilerK automatically generate code that converts "water" to a float while building * the argument list to be passed to printf ?  M Arer there any gotcha's to be aware of when working with both types with some  interchanges going on ?   I (one part, which does fancy calculations, requires the added precision of L double, while the other part was written separately and doesn't require that? precision and just outputs the data (DXF output of a 3d model).   E It all *seems* to work, but this is the type of stuff that is next toAK impossible to manually calculate to see if the program works. If the outputr. "looks" OK, I can only assume that it works...   ------------------------------  % Date: Sun, 19 Aug 2001 15:40:23 +0800s- From: David B Sneddon <dbsneddon@bigpond.com>p% Subject: ZLX2-E (TGA2) dial settings?yA Message-ID: <5.1.0.14.0.20010819153729.009f0ce0@mail.bigpond.com>.  	 Hi Folks,   F I have an Alphastation 255/333 with the above graphics card installed.E The card has a dial for various frequency settings.  Does anyone havedF the details of the various settings?  I have been unable to locate the! information via various searches.-     Regards, Dave.u -- 1I David B Sneddon (dbs)  OpenVMS Systems Programmer   dbsneddon@bigpond.com:I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/rI DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htmhI "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------   End of INFO-VAX 2001.460 ************************