1 INFO-VAX	Sat, 14 Jun 2003	Volume 2003 : Issue 328       Contents: 400 Mile Clustering?? / Re: DCPS or PCL and Printing Overlay / Underlay # Re: Rookie question for the experts # Re: Rookie question for the experts # Re: Rookie question for the experts ' Re: TK70 tape drive problem - Success ! ' Re: TK70 tape drive problem - Success !  Re: VMS Freeware CDs? Re: VMS V7.3-1 v. Exabyte EXB-8505XL and compression/compaction   F ----------------------------------------------------------------------  % Date: Sat, 14 Jun 2003 13:21:51 +0100 - From: Gerald Marsh <gerald@cyfer.demon.co.uk>  Subject: 400 Mile Clustering??8 Message-ID: <4e3mevoq0h0aabvr0nt098cja4t549hua4@4ax.com>  F We are currently running a two-site FDDI cluster over approx 5 miles -F An Alphaserver 8200 and DS20 on either end with HSZ50's and host-based# volume shadowing. (Based in the UK)   E The cluster runs business critical applications but only from one the D 8200's at any one time, although we do swap the live node frequentlyA for testing. All nodes have one vote except the usually live node F which has two. This would allow the cluster to continue without manual, intervention if the "stand-by" site is lost.  E We would like to extend the distance somewhat! The normal route would E be 200 miles but, if we needed to use a contingency network route, it  could be 400 miles.   D Most of the work of the live node is straight through processing butC there is some interactive load which would be sensitive to response C problems caused by network latency - In the past we looked at using E Gigaswitches with the FDDI interfaces to support longer distances but C the latency due to switching ruled it out. I  am therefore assuming ; that we would need ATM cards in the alphas to prevent this.   A Is anyone out there running a cluster over such a distance? If so C please let me know if we can contact someone to pick his/her brain.   D I intend to research the newer features of volume shadowing, such asD the device sites, to prevent extraneous IO over the WAN. Any info on' this would also be gratefully received.   E As we would be dependant on a network service provider for the comms, ? any information regarding supported protocols would be welcome.   D I'm pretty sure that there is no such cluster in the UK but I may be wrong.    F I suspect that I have not provided sufficient info in this request for* some, in which case I apologise profusely.  2 I would be grateful for any info on this, as ever.  F (BTW, I've recently set up CSWS on a test alpha and I'm very impressed+ with it, especially the DCL CGI interface.)   ' Bye for now and keep the flag a-flying!    Gerald.    Gerald Marsh  / gerald -at- cyfer -dot- demon -dot- co -dot- uk    ------------------------------  # Date: Sat, 14 Jun 2003 13:38:20 GMT 1 From: "Jerry Alan Braga" <jerry.braga@rogers.com> 8 Subject: Re: DCPS or PCL and Printing Overlay / UnderlayH Message-ID: <gzFGa.95881$3Sm.63567@news01.bloor.is.net.cable.rogers.com>  $ Sorry but I really don't understand.  J I look at the scanned image as PS document and it has multiple showpage(s) in it.  K Being that I do not know anything about PS what do you touch and not touch. H The postscript was created by a Win2K machine again Hp 5 PS drivers then transferred by my VMS box.  H With PCL (which macros work but the overlay just will not) it seems moreI simple but PS would be a route I would like to go as it could solve other  issues we have.   H Do you have a sample I could look at to compare and what steps you took.  G Again all I want is to take a standard text file and under/overlay this L graphic image on top.  ie the text document is not a PS file but as standard sequential print file.  8 "John E. Malmberg" <wb8tyw@qsl.network> wrote in message6 news:SVaGa.4453$Jw6.3084812@news1.news.adelphia.net... > Jerry Alan Braga wrote: ' > > Ok, here is what I am trying to do:  > > H > > Print an ascii print file created by an application and have a image file be   > > overlayed onto of this text. > > H > > I have installed DCPS for my hp9000 printer and created the queue. I haveG > > scanned in the image of my form and then printed to a file on my PC  using J > > PS (poform.ps) and PCL (poform.pcl) drivers.  I move the PS file to my Alpha J > > using pathworks and then print this file to my hp9000 DCPS printer and itL > > works just fine. Now the tricky part.  How do I make this ps file into a > > overlay/underlay.  > - > It has been a while since I have done this.  > J > What I did is convert the image in the file to be a macro that is loadedH > into the printer at the start of the print job.  Ideally this would beF > removing the "showpage" operator from it and making it into a macro. > E > Due to limitations in the printer that I was using, what I ended up I > doing is breaking up the image into a number of macros, and then making 4 > a macro that put called all the individual macros. > I > I put the resulting file in a printer text library, and had a form load H > the macro into the printer at the beginning of that type of print job. > 9 > Then I had the PostScript Document invoke it as a macro  > K > > I have seen the confidential.ps sample and tested it using my login.com  and K > > it prints ok. I know nothing of postscript so I have no way how to take  thisL > > 344 block file and covert the image postscript information to make it an > > overlay/underlay.  > J > Once you make the image into a macro, you can manipulate it the same way >   as the confidential.txt. > C > You can put it in a text library.  DCPS can access addtional text @ > libraries through a logical name that contains a list of them. > 9 > > Is there not any applications out there that would do H > > this sort of thing by parsing the postscript text and allowing these kinds  > > of changes.  > J > I used a custom written TPU program that did most of the conversion, but0 > some manual tweaks always seemed to be needed. > F > The people creating the PostScript document never seemed to rememberJ > what settings they used, and seemed to want to use a different tool each > time so it was interesting.  > L > > Does anyone know to take an ascii print file and overlay some other text or' > > in this case an image of some sort.  > H > See the above.  In addition, you may be able to store the image on theJ > printer and call it up as a macro.  I was looking into that when I had a; > change of job and stopped being concerned about printing.  >  > -John  > wb8tyw@qsl.network > Personal Opinion Only  >    ------------------------------  # Date: Sat, 14 Jun 2003 07:26:27 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER), Subject: Re: Rookie question for the experts5 Message-ID: <D6AGa.203073$lL2.1959215@news.chello.at>   S In article <bc941o$v7n$1@news3.tilbu1.nb.home.nl>, Dirk Munk <munk@home.nl> writes:  >JustMe wrote:L >> Just make sure you start up DECnet on the required line (EWA0) before youK >> start LAT.  My recollection from older (VAX) versions was that LAT would M >> "take over" the adapter if started first, and then DECnet wouldn't be able  >> to start the line.   G No. LAT has DECnet compatible addresses as default. So if you start LAT H before DECnet and do not prevent it using its default address mode, thenJ starting DECnet is no problem. (see LATCP> CREATE LINK/[NO]DECNET in Help)  O >Yep, when the interface is used for DECnet phase IV (compatible) traffic, the  O >physical address is changed to AA-00-04-00-aa-bb. This can only be done if no   >other protocol is active.  H This is/was true for DECnet Phase 4. But with DECnet Phase5, DECnet usesI Mac-Add in the form AA-00-04-00-aa-bb, while IP can use 08-00-2B-ab-cd-ef J (or 00-00-F8-ab-cd-ef) at the same time, so the ordering of startup is not so important anymore.   H In fact, if you use TCPIP DCHP Client on OpenVMS with DECnet-Plus, it isE important to start TCPIP before DECnet (and this one gets hard and/or  unsupported)  K >> (btw, I think the IP equivalent of MOP is BOOTP not TFTP, but I could be ( >> mistaken -- wouldn't be the 1st time) > I >If I'm not mistaken the 'call for help' is BOOTP, and the actual loading 	 >is TFTP.    Indeed. But to nitpick  @ BOOTP handles address information and this one MOP doesn't need,J because it is a Layer2 (hence LAN-only) protocol which uses (unchangeable)C MAC Addresses (or it other words, it is configured in the factory).   G And MOP can do "Remote Console", which neither BOOTP nor TFTP provides.    --   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: Sat, 14 Jun 2003 09:07:40 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER), Subject: Re: Rookie question for the experts5 Message-ID: <wBBGa.204303$lL2.1978857@news.chello.at>   N In article <3EE8CCD5.8D2AAD16@istop.com>, JF Mezei <jfmezei@istop.com> writes:N >What version of VMS are you running ? What software do you have that requires) >DECNET/OSI ? (or do you mean DECNET 5 ?)   , DECnet5 is DECnet/OSI. What was your point ?B The current official name is DECnet-Plus (with kitname DECNET_OSI)  M >My instinctive reaction would be to keep the real decnet (decnet 4) which si J >simpler (much simpler) and easier to configure, aswell as consuming fewerK >resources and use it only on the local lan to support the SNA gateway. And I >then configure the terminal servers to use LAT and have the MPP requests D                                                              ^^^ ;-) >handled by LANCP.  % Resist your instinctive reaction ;-). 5 I'm a happy Phase 5 customer since over a decade now. : I haven't seen a need for DECnet Phase 4 for a decade now.  4 This seems to become another religion war (4 vs. 5).E DECnet is DECnet, but Phase 5 has OSI, DECnet-over-IP and centralized D directory services (like BIND/DNS/DOMAIN or DECdns) in addition (andN resource consumption and NCP vs NCL aside is also much more logical/flexible).J It also have a graphical network management tool (SYS$SYSTEM:NET$MGMT.EXE)  K Only problem/annoyance I have with is the absence of multiwildcard support.   - 	NCL> SHOW NSP LOCAL NSAP * REMOTE NSAP * ALL  or4 	NCL> SHOW ROUTING CIRCUIT * REACHABLE ADDRESS * ALL  @ gives %NCL-E-MULTILEVELWILD, multi-level wildcarding not allowed> If this is fixed, then I'd say, DECnet-Plus is almost perfect.  K And I also see no need for LANCP if DECnet is installed/configured/running.    --   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: Sat, 14 Jun 2003 08:40:57 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER), Subject: Re: Rookie question for the experts5 Message-ID: <tcBGa.203808$lL2.1978857@news.chello.at>   h In article <9b72a38f.0306121005.6cb69b5d@posting.google.com>, rferrell@smcwv.com (Rusty Ferrell) writes:w >"Hank Vander Waal" <hvanderw@mansply.com> wrote in message news:<MDEGJFAOHGLNLHONIEBMEEEGCLAA.hvanderw@mansply.com>... # >> FWIW   decserver 300's do use IP   N Last I looked: Not for booting (no BOOTP/TFTP), only for connections (TELNET).  G >> I think LAT always wants to bind to the first ethernet it finds.  If G >> that's not the one you want you can define lat$link before you start  >> LAT.   A You can also add LATCP> CREATE LINK commands in LAT$SYSTARTUP.COM F but defining LAT$LINK is the preferred solution (not a DELETE LINK...) for a non-standard single link.   G >I'm thinking that I might have a port speed problem as well. The DS10L G >ethernet ports are 10/100 ports and the thinwire is only 10mb. I can't C >seem to locate in ALPHABIOS or VMS where I can set this speed to a < >fixed value, or even LOOK at the interface for that matter.  J VMS doesn't use AlphaBIOS (is for Win and Linux), it uses SRM (is for VMS,E Tru64 and Linux). And at the SRM's dead sergeant prompt you can enter   $ >>> SHOW EWAx_MODE	where x is A or BN >>> SET EWAx_MODE mode	where x is A or B and mode is TWISTED, FASTFD, FAST,...   And in VMS you can use   $ RUN SYS$SYSTEM:LANCP LANCP> SHOW DEVICE/PARAMETER( LANCP> SET DEVICE/SPEED=10/NOFULL_DUPLEX  H >                                                              I can seeF >the TCPIP interface which I set up on "WE1" (EWB0), but can't seem toE >get VMS to show me what DECnet, or anything else for that matter) is E >bound to. Also, when I do a "SHOW NETWORK DECNET" it show me thatthe E >address is 49.75 and I entered 1.10 when doing the NET$CONFIGURE.COM  >setup.   D This is because you haven't matched the SYSGEN parameter SCSSYSTEMID' to the DECnet address you like to have.   * SCSSYSTEMID = decnet-area*1024+decnet-node  D So, EDIT SYS$SYSTEM:MODPARAMS.DAT and replace SCSSYSTEMID=50251 withI SCSSYSTEMID=1034 and then run @SYS$UPDATE:AUTOGEN or change the parameter  in SYSGEN by hand    	$ RUN SYS$SYSTEM:SYSGEN 	SYSGEN> U C 	SYSGEN> SET SCSSYSTEMID 1034  	SYSGEN> W C or 	$ RUN SYS$SYSTEM:SYSMAN 	SYSMAN> PARAMETER USE CURRENT' 	SYSMAN> PARAMETER SET SCSSYSTEMID 1034   	SYSMAN> PARAMETER WRITE CURRENT   and reboot afterwards.  A >         I've removed DECnet and reinstalled DECnet/OSI from the ? >installation CD and it remembered all of the settings from the A >previous installation. Is there some place I can find where this C >information is stored so that I can correct the address? I've been ? >reading every manual that I can find but haven't had much luck C >locating the right information. IF you could point me in the right 7 >direction I'll be happy to RTFM...again and again.....   H As you found, the DECnet reinstallation was unneccessary. The info is in  % 		SYS$STARTUP:NET$ROUTING_STARTUP.NCL  and ' 		SYS$SYSTEM:NET$CHECKSUM_NCL_LOCAL.DAT    But your problem is in SYSGEN.   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  + Date: Sat, 14 Jun 2003 10:49:36 +0000 (UTC) A From: Adrian Stapley <adrian@semleystation.spammed-to-hell.co.uk> 0 Subject: Re: TK70 tape drive problem - Success !/ Message-ID: <bceujv$q13$1@titan.btinternet.com>   L Thanks to all for advice, it seems a combination of two suggestions did the  trick   L 1) cleaning the heads - they do take a lot of cleaning until the swabs stay  clear.  H 2) getting the device error count to zero. As one poster guessed, I was I trying to excercise a tape immediately after booting, and racking up the  I error count, then cleaning the heads and trying again, but with the high  B error count masking wether or not the cleaning had done the trick.  D I can also confirm that you can detach and re-attach a tk70 without G powering down, just disconnect the power plug, then the data lead, and  J re-attach in reverse order. However, it does not alter the error count in  the device driver area. :(  H The utility I had come across for resetting device error counts was for  modern open vms on alphas.   Adrian Stapley wrote:    > Adrian Stapley wrote:  > K > Thanks to all the replies - I have tried cleaning the heads, and also the 5 > small led? sensors either side of the rear capstan.  > L > Error count - does not get reset after a reboot, I have already tried thisL > and am currently looking for a utility I remember seeing on an ftp site to% > reset the error count for a device.  > J > The symptoms I am seeing would suggest to me that it is a head failure -L > after closing the handle on tape (either a fresh one for an init or a used0 > one for a mount), the following actions occur: > L > 1) The leader engages the tape and it is hauled through the capstans until5 > the hole in the tape is almost on the take up spool  > J > 2) Nothing else happens until an init or mount command, when the tape isJ > run back slightly until the hole has reached the heads, then the tape is: > forwarded until the hole is almost on the take up spool. > & > 3) The mount error message is given. > . > From 1 and 2 I assume that the LED's are OK. > K > Trying to manually wind the tape by turning the take up spool gives all 3 I > led's flashing (so movement sensors are also OK ?). Pressing the unload ; > button will rewind the tape and clear the flashing LED's.  > J > Of course, I am going to be totally wrong, and find it is the controller8 > board or error count, as has already been suggested :) >  > Adrian > " > (Why did I stop using my QL's ?)   --  ! Getting wight to the filthy stuff    ------------------------------  # Date: Sat, 14 Jun 2003 13:17:23 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network> 0 Subject: Re: TK70 tape drive problem - Success !< Message-ID: <DfFGa.6453$Jw6.3536717@news1.news.adelphia.net>   Adrian Stapley wrote: F > I can also confirm that you can detach and re-attach a tk70 without I > powering down, just disconnect the power plug, then the data lead, and  L > re-attach in reverse order. However, it does not alter the error count in  > the device driver area. :(  I Experience at a site I was at is that plugging in disks and tapes to the  H chassis would sometimes result in a power supply glitch that was enough  to start a reboot cycle.  G I would not recommend Hot plugging any device that does not explicitly   support it.    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  # Date: Sat, 14 Jun 2003 12:10:27 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: Re: VMS Freeware CDs 5 Message-ID: <TgEGa.207233$lL2.2023235@news.chello.at>   _ In article <20030611.4525900.702163943@imagnu.geo>, Andrew Balaam <abalaam@yahoo.co.uk> writes: E >Does anyone know if the images of the VMS freeware CDs are available E >somewhere on the Internet for downloading? If so, where would I find  >them?  7 Not from DEC^WCOMPAQ^WHPQ and not from DECUS^WENCOMPASS 8 but HPQ offers a collection of such freeware products at  3 	http://www.openvms.compaq.com/freeware/freeware50/   L But note, the webserver contains often more recent (if not current) versionsI of the products on the equally named CDs, so having the CDs is quite nice I to get an overview and insight. But to use the products I suggest finding < the current version in the internet (eg. on HPQs webserver).   -Peter  F PS: I think, the should have started a /freeware60/ tree and when in aF few months the VMS V7.3-2 comes out a snapshot of this tree should get8 to the freeware CDs and a new /freeware70/ should start.   --   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: Sat, 14 Jun 2003 11:49:20 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> H Subject: Re: VMS V7.3-1 v. Exabyte EXB-8505XL and compression/compaction' Message-ID: <3EEB5210.B97542BB@fsi.net>    sms@antinode.org wrote:  > [snip]J >    Back on 12-NOV-2002, I complained about a problem with compression on< > my Exabyte EXB-8505XL 8mm tape drive ("device type EXABYTEG > EXB-8505SMBANSH2", and revision level 0793, according to SCSI_INFO),    D Just curious - does this agree with SHOW DEVICE/FULL? I should think8 they'd both display the same string, but you never know.  I >    If anyone out there has one of these things actually compressing (on B > VMS -- It works fine on Solaris 9), or any useful advice, I'd be > interested in hearing it.    Agreed.   F As you might guess from another post, I'm currently locking horns withG my VAR over device compatibility issues in another arena (SDLT-320) and @ would be interested in any anecdotes anyone is willing to share.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------   End of INFO-VAX 2003.328 ************************