1 INFO-VAX	Sun, 24 Aug 2003	Volume 2003 : Issue 467       Contents:4 Re: (OT) PDP-8 CPU (was Re: PDP-11 OS Release Dates)4 Re: (OT) PDP-8 CPU (was Re: PDP-11 OS Release Dates) Re: Account creation info + connect FALCO 5220e to ALPHAstation 255/233 / Re: connect FALCO 5220e to ALPHAstation 255/233 > Re: Difficulties with DPW 500au and StorageWorks components... Re: DSSI problem Re: DSSI problemJ Re: I need 4 DEC/BASIC OpenVMS programmers for perm positions immediately!J Re: I need 4 DEC/BASIC OpenVMS programmers for perm positions immediately!8 Look for information on exercising third-party FC disks.< Re: Look for information on exercising third-party FC disks." Re: Looking for TK50s in the UK..." Re: Looking for TK50s in the UK..." Re: Looking for TK50s in the UK...$ M7940/DHV-11, TU58, and a VAX 11/750( Re: M7940/DHV-11, TU58, and a VAX 11/750$ Re: OpenVMS Jobs, OpenVMS.org etc... OpenVMS Professionals Directory  Re: OpenVMS SecurityE overhead with host-based volume shadowing for a one-member shadow-set 
 Re: Praxis. Question about a free shell provider - VisTech- Regarding previous post of mine about VisTech 2 Re: setting up a LAT service via a terminal server2 Re: setting up a LAT service via a terminal server3 Re: The Dell Effect (Was: Re: Will OpenVMS I64 ...) 1 Re: Will OpenVMS I64 run on a Dell PowerEdge 3250   F ----------------------------------------------------------------------  # Date: Sun, 24 Aug 2003 04:28:23 GMT - From: Stench Slobberwart <rivie@ridgenet.net> = Subject: Re: (OT) PDP-8 CPU (was Re: PDP-11 OS Release Dates) , Message-ID: <3F483EE6.BA0CD59A@ridgenet.net>   David M Smith wrote:  M > I learned about computers on PDP-8's. I was in college, studying electrical N > engineering, in 1970 when I got a job as a "coop" (work-based study program)& > with a small DEC OEM in Atlanta, GA.  H I first encountered -8s in the early 1980s. The EE department at USU hadB a professor who was an -8 fanatic. The EE BASIC programming courseF was taught timeshared on -8; one an -8/E and the other an -8/M. One ofD the machines had core, the other semiconductor. You could tell whichC was which in the fall when the thunderstorms came through; the core = machine kept on going while the semiconductor machine crashed  at each power flicker.  N It took me about ten years or so to realize the -8 was an interesting machine.L I hated it until one day when I suddently said "hey, wait a minute..."; I've been a fan ever since.  N > The PDP-8 used 12-bit words. The high-order 3 bits were the opcode, so there! > were only 8 basic instructions.   L The PDP-5 executed the same instruction set, but kept the program counter in8 core location zero. The PDP-5 just turned 40 this month. --
 Roger Ivie rivie@ridgenet.net   ------------------------------  # Date: Sun, 24 Aug 2003 05:28:10 GMT ; From: "John Gemignani, Jr." <jon-nope@thiswontworkossc.net> = Subject: Re: (OT) PDP-8 CPU (was Re: PDP-11 OS Release Dates) ; Message-ID: <K1Y1b.6819$Nc.4289372@news1.news.adelphia.net>   : "Stench Slobberwart" <rivie@ridgenet.net> wrote in message& news:3F483EE6.BA0CD59A@ridgenet.net... >  >  > David M Smith wrote: > D > > I learned about computers on PDP-8's. I was in college, studying
 electricalG > > engineering, in 1970 when I got a job as a "coop" (work-based study  program)( > > with a small DEC OEM in Atlanta, GA. > J > I first encountered -8s in the early 1980s. The EE department at USU hadD > a professor who was an -8 fanatic. The EE BASIC programming courseH > was taught timeshared on -8; one an -8/E and the other an -8/M. One ofF > the machines had core, the other semiconductor. You could tell whichE > was which in the fall when the thunderstorms came through; the core ? > machine kept on going while the semiconductor machine crashed  > at each power flicker. > G > It took me about ten years or so to realize the -8 was an interesting  machine.I > I hated it until one day when I suddently said "hey, wait a minute...";  I've > been a fan ever since. > J > > The PDP-8 used 12-bit words. The high-order 3 bits were the opcode, so there # > > were only 8 basic instructions.  > K > The PDP-5 executed the same instruction set, but kept the program counter  in: > core location zero. The PDP-5 just turned 40 this month. > -- > Roger Ivie > rivie@ridgenet.net >  >   H I learned programming on an 8E running TSS/8.22H in high school in 1974. Started K with Basic, then did FOCAL (who remembers that?), then Fortran-D (subset of 
 Fortran-III with a grand limit of 49 forward references in the compiler),  then PDP-8 
 assembler.  L I believe that I still have my manuals around the house somewhere, including the TSS/8 system< call manual.  Remember what the OLD DEC manuals looked like?   -John    ------------------------------  % Date: Sat, 23 Aug 2003 18:13:31 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>" Subject: Re: Account creation info( Message-ID: <3F47E701.9E58433@istop.com>   Phil wrote: B > How can I see who created a user account with VMS.  I was told aE > utility called Ferret existed for VMS that can get this infomation.   N I have logging of changes to SYSUAF turned on. so my operator log does contain, the information about changes to sysuaf.day.   Also:   K ANA/AUDIT/EVENT=AUTHORIZATION/FULL will give you those details (you can add ( /SINCE etc to further refine the search)   ------------------------------  + Date: Sat, 23 Aug 2003 23:55:23 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: connect FALCO 5220e to ALPHAstation 255/233$ Message-ID: <bi8utb$2f1$1@online.de>  F Several years ago, I used to have a FALCO 5220e terminal connected to I the serial port (OPA0 or TTA0) of an ALPHAstation 255/233.  Now, I can't   get it to work.   G The serial ports on the ALPHAstation work OK; a VT320 connects up with  I no problem.  Thus, I suspect that the problem is with the FALCO terminal   and/or with the cables.   F The cables are PHYSICALLY OK, since other connections using them work.  I When I had things working several years ago, I had a couple of very long  C serial cables connected together and used that as the connection.   G Presumably, the reason was that a standard serial cable wouldn't work,  = and the ones which did had some non-standard pin connections.   E Does the FALCO require a different sort of cable (apart from gender)  D than a VT320 (the FALCO has a female RS232 connector on the back (2 G actually, one for each port) whereas the VT320 has a male connector)?   H There are so many options in the setup that even if it were possible to F change the nature of the connection, I doubt I could find it.  Anyone  know if this is possible?   E What should the setup values be for the FALCO to use it as a console   terminal with the ALPHAstation?   G I have a serial cable where I can flip switches to turn the individual  F pins on and off at each end.  What should the settings be for a VT320? For the FALCO?   ------------------------------  % Date: Sat, 23 Aug 2003 19:07:10 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 8 Subject: Re: connect FALCO 5220e to ALPHAstation 255/233' Message-ID: <3F4801AE.34AB8423@fsi.net>   / Phillip Helbig---remove CLOTHES to reply wrote:  > [snip]F > What should the setup values be for the FALCO to use it as a console! > terminal with the ALPHAstation?  > H > I have a serial cable where I can flip switches to turn the individualH > pins on and off at each end.  What should the settings be for a VT320? > For the FALCO?  & I'm thinking the usual should suffice:   9600, 8-N-1, data-leads only.   G If the Falco can't be set up to ignore the "hardware flow control", try H (on the terminal end) jumpering DSR to DTR (pins 6+20), CTS to RTS (pins< 4+6), and if necessary, DTR+DSR to carrier detect (6+8?+20).   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Sun, 24 Aug 2003 03:54:00 GMT " From: "ohm62" <ohm62@coldmail.com>G Subject: Re: Difficulties with DPW 500au and StorageWorks components... 8 Message-ID: <sFW1b.13897$kF1.12593@nwrdny02.gnilink.net>  ' Thanks again for the suggestion, Tom...   J Unfortunately, setting the first 3 address dip switches didn't do it in my
 case...   :-(   J I was not using INIT, and power-cycling the PWS completely, as well as theK BA. prior to each test, to make sure the bus was fully reinitialized, so it B is not a problem with INIT either (although I was not aware of the8 limitation of INIT either, so the warning is welcome :-)  K I am wondering if I have the right QLA1040, now...  It looks like there are I many different models out there, and according to the document William W. I Webb was kind enough to point out to me (TCR_HWCFG.pdf, easily found with H google search) I would need the KZBPA-CB, which is capable of ultra wideE differential (and single ended?).  The SRM let me select differential F terminaison on mine, but I am not sure that proves anything...  It wasK merely advertised as QLA1040 Ultra-wide single ended when I bought it, so I K am not sure what it really is.  I'll probably have to pull it out again and $ look at the part & serial numbers...  D Tom, did you get a second QLA1040, PCI based (which port, btw?), forK external SCSI or do you use the internal adaptor with an extension cable to J provide the external port?  I am curious to know whether a second QLA is a> viable solution, as Dirk mentionned he read somewhere about anI incompatibility, and I do remember coming across such a warning in one of J the news groups...  As I am running SRM v7.2-1 and the QLA firmware v5.57,H and the versions the problem description referred to were ancient, I hadI some hopes my more recent versions would help getting around that part of  the issue...  H Thanks to all of you who gave a hand so far, I am very grateful for your time and advices!    Cheers,   
   -- Olivier.    ------------------------------   Date: 23 Aug 2003 18:56:24 GMT2 From: Thierry Dussuet <thierry@squeeeez.no-ip.com> Subject: Re: DSSI problem 1 Message-ID: <slrnbkfe6o.1cf.thierry@VENUS.Family>   8 In article <bi53pu$2do$1@lore.csc.com>, Nic Clews wrote: > Thierry Dussuet wrote: >>  H >> From these 3, number 7 has the terminator and is on Bus 0. I've takenF >> out the hard disk with the ID 3 to clean it, put it back in, and itG >> stopped working.  SHOW DSSI shows the Busses, with ID's 6 and 7, but D >> nothing more.  So... if it isn't the terminators and not the fuse2 >> either... Do you have any ide what it could be? > H > The DSSI backplanes are notoriously hard to seat a drive into, you mayH > have to really push to make sure the back connector seats properly. IfG > you noticed, the connector on the back of the drive has some movement G > (or play) so it aligns. Just also make sure that the ribbon connector 1 > hasn't got displaced out the back of the drive.   C OK - I've just taken the drive out to check if the ribbon cable was C seated in correctly and put it back.  It seemed quite ok to me, and H pushing it back in is really hard, but may I assume that the drive seatsF properly if the screws fit in without problems and I can't fasten them any tighter?  D > I'd also try powering up with the drive with ID 3 pulled out. It'sE > possible that a fault on the DSSI is stopping the other drives from 
 > being seen.   C When I pull out an ID (I've tried with all of them, 1, 2 and 3 at a A time), when I power the system up, the red fault light lights up. 1 But still, SHOW DSSI sees only the terminators... F I've now also tried to pull the hard disk 3 totally out, but it didn't help either.   Thierry    ------------------------------   Date: 23 Aug 2003 23:30:03 GMT2 From: Thierry Dussuet <thierry@squeeeez.no-ip.com> Subject: Re: DSSI problem 1 Message-ID: <slrnbkfu7r.1cf.thierry@VENUS.Family>   B In article <erx1b.3165$Oo7.731@news.cpqcorp.net>, Bob Blunt wrote: > Thierry Dussuet wrote:F >> Hello!  I have a VAX 4000/300 here with 3 RF72 disks in it, as wellG >> as a TK70 tape drive.  Everything went well until someday I realised F >> it was quite dusty, so I took one drive out to see if it was reallyG >> that dusty - but it wasn't, so I put it back in.  And now it doesn't E >> see the disks anymore! All vanished, although the disks themselves C >> have the green light on, and I hear them from time to time.  The D >> startup tests show no errors, either, and trying to boot from the >> system disk does not work.  >>   >>   >>>>>sh dev  >>   >> DSSI Bus 0 Node 6 (*) >>   >> DSSI Bus 1 Node 7 (*) >>  0 >> UQSSP Tape Controller 0 (774500) -MUA0 (TK70) >>  - >> Ethernet Adapter -EZA0 (08-00-2B-16-E7-33)  >>   >>  C >> They were on Bus 0.  Does anybody have a hint?  Something that I 8 >> could have done wrong while putting the disk back in? >>  
 >> Thierry > H > Thierry, the fuse for the one of the DSSI busses is on the back of theE > system console panel.  IIRC, it was a fairly large and obvious item E > that was soldered onto the circuit board, visible without taking it  > apart.  C What I have on the back of the back of the control panel is a white , plastic which covers the circuit board, with - 2 jumpers on the top rightE - a big black box with "LAN PAC" written onto it and a cable going to    a card near the CPU cardF - a small green electronic component with a white "O" written below it   on the circuit board. 1 Is it one of these 3? Or is it on the other side?   F >  From your last response, I gather that you have other drives in theC >  drive bays, so I doubt that this unit got in upside down (Again,   C Putting it upside down doesn't work, the screws on the front aren't < exactly in the middle, such that you see it rather easily...  H >  IIRC, you have both DSSI AND SCSI connectors on the backplane but theF >  alignment of the edge connector on the back of the drive determines   unfortunately no SCSI...  F >  which slot gets used; if by chance it got worked into the SCSI slotG >  and you have no SCSI adapter, it definitely would NOT be seen on the E >  DSSI).  I presume that you've reseated the drive.  Does it spin up F >  and go through it's calibration passes when power is applied to the >  system?  Any fault lights?   F I don't see any difference between the drives... they all spin up, andH then start to make noises like a steam train :) They behave normally (or; just like I've always seen them before). No fault lights...   B > If you pull the drive out, also check the appearance of the edgeH > connector.  Make sure that it isn't binding or sitting at an angle andF > that it looks basically "square" back there.  There is some room forE > movement of that connector, but not a whole lot.  I can't recall if B > the "slop" at that connector to align with the backplane slot isC > single- or double-axis.  Also make sure that the ribbon cable and E > power feed from that connector are both seated onto the ISE for the  > drive.  ; Uh... I'll make a checklist - not sure of everything yet...   G > Stupid time to ask this question, but you "cleaned" this disk...  Did F > you mechanically disassemble it in any way to do so, or did you just > brush or vacuum it off?   D I not even brushed it.  Just vacuum a bit, since there was really no) dust (or not much), and put it back in...    Thierry    ------------------------------  % Date: Sun, 24 Aug 2003 02:21:16 +0200 + From: Milan Jercic <milan.jercic@access.si> S Subject: Re: I need 4 DEC/BASIC OpenVMS programmers for perm positions immediately! ( Message-ID: <3F4804FC.7010008@access.si>  	 Hi Jason,   H > Currently interviewing and looking to fill ASAP.  If you fit the skillG > set listed above or know of someone that does please send your resume   > to jcarl@paradigm-wa.com ASAP. >    Um, how about outsourcing?     Regards,           J.M    ------------------------------  % Date: Sat, 23 Aug 2003 18:23:10 -0700 * From: "Jack Peacock" <peacock@simconv.com>S Subject: Re: I need 4 DEC/BASIC OpenVMS programmers for perm positions immediately! 2 Message-ID: <akmdnWjqTcrjjtWiU-KYvw@mpowercom.net>  7 Milan Jercic" <milan.jercic@access.si> wrote in message " news:3F4804FC.7010008@access.si... > Hi Jason,  >  > Um, how about outsourcing? > K Must really be a bad job if they can't even get one of the Indian companies  to take it on.    Jack Peacock    ------------------------------  # Date: Sat, 23 Aug 2003 18:43:14 GMT " From: Lee <lytmah@telusplanet.net>A Subject: Look for information on exercising third-party FC disks. / Message-ID: <3F47B5BB.3B528F98@telusplanet.net>   A Our VMScluster is comprised of 3 sites and 5 nodes (4 production, ; 1 quorum) employing HBVS and connected by GigE and a 10/100 ? ISP cloud for cluster heartbeat.  Storage is HSJ40/HSJ50 disks. @ There are 2 ES45's at site 1 and 2 ES45's at site 2.  The quorumC node is at site 3.  I'm currently testing some third party FC disks 6 $1$DGA) to see how they would function in the cluster.  D                                        Site 1                 Site 20 Shadow set DSA10        $1$DGA11        $1$DGA120 Shadow set DSA20        $2$DUA21        $2$DUA220 Shadow set DSA30        $1$DGA31        $2$DUA320 Shadow set DSA40        $2$DUA41        $1$DGA42  = There are another 6 shadow sets of various combinations of FC = and HSJ disks.  The ten shadow sets are mounted cluster-wide.   ; My current exercise procedure performs the following tasks: ;     1. Use BACKUP to copy files from a muti-level directory 9         source and write to a directory on the test disk. 2     2. Run CONVERT to read an ISAM file and create5         a new ISAM file in [000000] on the test disk. 1     3. Run CONVERT to read a second ISAM file and <         create a new ISAM file in [000000] on the test disk.9 These 3 tasks are executed on all 5 nodes of the cluster. 8 The result is that when the exercise begins, there would3 be 15 concurrent batch jobs running.  On the shadow 6 set with both FC members, the elapsed time ranges from8 5 minutes to 30 minutes.  The elapsed time on an HSJ/HSJ* shadow set takes much longer, as expected.= In addition to the above tasks, I also run delayed batch jobs  to do the following:4     1. Do a "$ DIRECT [000000...]" of the test disk.<     2. Check out the fragmentation of the disk using a third         party defragger.3     3. Do an "$ ANALYZE/DISK/READ" of the test disk   C So far, I have run the exerciser on no more than 4 shadow sets at a D time, mainly because only one of the 4 production nodes is connectedA to the FC disks at this time.  The FC disks have exhibited normal > behavior just like the HSJ disks so far.  When all 4 nodes are= connected to the third-party FC platform next week, I plan on @ running the exerciser concurrently on more than 4 disks.  I wantA to see what happens when a nodes exits and re-enters the cluster. ; Also on how the FC shadow sets behave when a nodes crashes.   > Is my exerciser extensive enough?  Suggestions/hints welcomed.4 What other tests can I perform on the foreign disks?7 Is a canned procedure available to perform other tests?      -- Lee    lytmah@telusplanet.net   ------------------------------    Date: 23 Aug 2003 23:29:16 -0500+ From: young_r@encompasserve.org (Rob Young) E Subject: Re: Look for information on exercising third-party FC disks. 3 Message-ID: <2hF0K83lfcFv@eisner.encompasserve.org>   T In article <3F47B5BB.3B528F98@telusplanet.net>, Lee <lytmah@telusplanet.net> writes:C > Our VMScluster is comprised of 3 sites and 5 nodes (4 production, = > 1 quorum) employing HBVS and connected by GigE and a 10/100 A > ISP cloud for cluster heartbeat.  Storage is HSJ40/HSJ50 disks. B > There are 2 ES45's at site 1 and 2 ES45's at site 2.  The quorumE > node is at site 3.  I'm currently testing some third party FC disks 8 > $1$DGA) to see how they would function in the cluster. > F >                                        Site 1                 Site 22 > Shadow set DSA10        $1$DGA11        $1$DGA122 > Shadow set DSA20        $2$DUA21        $2$DUA222 > Shadow set DSA30        $1$DGA31        $2$DUA322 > Shadow set DSA40        $2$DUA41        $1$DGA42 > ? > There are another 6 shadow sets of various combinations of FC ? > and HSJ disks.  The ten shadow sets are mounted cluster-wide.  > = > My current exercise procedure performs the following tasks: = >     1. Use BACKUP to copy files from a muti-level directory ; >         source and write to a directory on the test disk. 4 >     2. Run CONVERT to read an ISAM file and create7 >         a new ISAM file in [000000] on the test disk. 3 >     3. Run CONVERT to read a second ISAM file and > >         create a new ISAM file in [000000] on the test disk.; > These 3 tasks are executed on all 5 nodes of the cluster. : > The result is that when the exercise begins, there would5 > be 15 concurrent batch jobs running.  On the shadow 8 > set with both FC members, the elapsed time ranges from: > 5 minutes to 30 minutes.  The elapsed time on an HSJ/HSJ, > shadow set takes much longer, as expected.? > In addition to the above tasks, I also run delayed batch jobs  > to do the following:6 >     1. Do a "$ DIRECT [000000...]" of the test disk.> >     2. Check out the fragmentation of the disk using a third >         party defragger.5 >     3. Do an "$ ANALYZE/DISK/READ" of the test disk  > E > So far, I have run the exerciser on no more than 4 shadow sets at a F > time, mainly because only one of the 4 production nodes is connectedC > to the FC disks at this time.  The FC disks have exhibited normal @ > behavior just like the HSJ disks so far.  When all 4 nodes are? > connected to the third-party FC platform next week, I plan on B > running the exerciser concurrently on more than 4 disks.  I wantC > to see what happens when a nodes exits and re-enters the cluster. = > Also on how the FC shadow sets behave when a nodes crashes.  > @ > Is my exerciser extensive enough?  Suggestions/hints welcomed.6 > What other tests can I perform on the foreign disks?9 > Is a canned procedure available to perform other tests?     7 	The HSJ's don't support sequential pre-fetch, your DGA : 	storage does - so anything sequential in nature (assuming? 	files opened for shared write in a cluster such that XFC won't 5 	come into play) the DGA storage will do much better.     > http://h18002.www1.hp.com/products/quickspecs/SOC/QB0074PF.PDF  D 	Secondly, the HSJ50 is rated at 4600 IOPS , 12.7 MByte/sec internalB 	(click on PDF above) but is hanging off CI so the CI bus is your G 	choke point at 2 * 70 MBit/sec or what .. 17 MByte/sec?  You will get  F 	40+ MByte/sec out of one of those hypers - much more out of a Meta.   	 A 	Smallish Random IO it is a toss-up.  You could create nice sized 9 	RAID 0+1 on those HSJs that will do nicely on random io.   D 	However, it is easy to see for backups, full table scans, month-end3 	runs, the more modern DGA stuff is a big time win.     F [Aside:	You can direct your IO to one shadow member or another.  Trick= 	here is to flip read costs.  See help set device/read_cost.]   ( 	Testing a hyper a couple of years back:    E iot /count=4000 /pattern=random/queue=32/size=2 $1$dga222:/show=histo N _$1$DGA222: (DGX00 DG)  4000 xfer,     9.530 sec,419.7 xfer/sec,    420 Kb/secD iot /count=4000 /pattern=random/queue=4/size=2 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,    13.170 sec,303.7 xfer/sec,    304 Kb/secD iot /count=4000 /pattern=random/queue=1/size=2 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,    35.940 sec,111.3 xfer/sec,    111 Kb/secF iot /count=4000 /pattern=random/queue=32/size=16 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,    10.690 sec,374.2 xfer/sec,   2993 Kb/secE iot /count=4000 /pattern=random/queue=4/size=16 $1$dga222:/show=histo N _$1$DGA222: (DGX00 DG)  4000 xfer,    14.140 sec,282.9 xfer/sec,   2263 Kb/secE iot /count=4000 /pattern=random/queue=1/size=16 $1$dga222:/show=histo N _$1$DGA222: (DGX00 DG)  4000 xfer,    36.190 sec,110.5 xfer/sec,    884 Kb/secG iot /count=4000 /pattern=random/queue=32/size=127 $1$dga222:/show=histo N _$1$DGA222: (DGX00 DG)  4000 xfer,    19.020 sec,210.3 xfer/sec,  13354 Kb/secF iot /count=4000 /pattern=random/queue=4/size=127 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,    20.070 sec,199.3 xfer/sec,  12656 Kb/secF iot /count=4000 /pattern=random/queue=1/size=127 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,    46.890 sec, 85.3 xfer/sec,   5417 Kb/secI iot /count=4000 /pattern=sequential/queue=32/size=2 $1$dga222:/show=histo O _$1$DGA222: (DGX00 DG)  4000 xfer,     0.920 sec,4347.8 xfer/sec,   4348 Kb/sec H iot /count=4000 /pattern=sequential/queue=4/size=2 $1$dga222:/show=histoO _$1$DGA222: (DGX00 DG)  4000 xfer,     0.930 sec,4301.1 xfer/sec,   4301 Kb/sec H iot /count=4000 /pattern=sequential/queue=1/size=2 $1$dga222:/show=histoO _$1$DGA222: (DGX00 DG)  4000 xfer,     1.250 sec,3200.0 xfer/sec,   3200 Kb/sec J iot /count=4000 /pattern=sequential/queue=32/size=16 $1$dga222:/show=histoO _$1$DGA222: (DGX00 DG)  4000 xfer,     1.560 sec,2564.1 xfer/sec,  20513 Kb/sec I iot /count=4000 /pattern=sequential/queue=4/size=16 $1$dga222:/show=histo O _$1$DGA222: (DGX00 DG)  4000 xfer,     1.570 sec,2547.8 xfer/sec,  20382 Kb/sec I iot /count=4000 /pattern=sequential/queue=1/size=16 $1$dga222:/show=histo O _$1$DGA222: (DGX00 DG)  4000 xfer,     2.120 sec,1886.8 xfer/sec,  15094 Kb/sec K iot /count=4000 /pattern=sequential/queue=32/size=127 $1$dga222:/show=histo N _$1$DGA222: (DGX00 DG)  4000 xfer,     5.380 sec,743.5 xfer/sec,  47212 Kb/secJ iot /count=4000 /pattern=sequential/queue=4/size=127 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,     5.320 sec,751.9 xfer/sec,  47744 Kb/secJ iot /count=4000 /pattern=sequential/queue=1/size=127 $1$dga222:/show=histoN _$1$DGA222: (DGX00 DG)  4000 xfer,     6.040 sec,662.3 xfer/sec,  42053 Kb/sec  @ 	First - note one interesting fact.  4347 IOPS out of one mirrorD 	pair (hyper).  Queue depth 32 - IO sequential - IO size 2 blocks.  E 	Obviously, all IO cache hits.  Prefetched to cache in 64K chunks no  @ 	doubt.  Kinda hard to get 4300 IOPS out of two disks otherwise.  C 	I interspersed random runs between sequential runs to trash cache. ; 	Prior to that insight (trashing cache), I was getting some ; 	truly wonderful numbers on subsequent sequential runs ;-).   D 	What you can also see above is 127 block IO with queue length of 1 F 	results in 42 MByte/sec throughput.  You won't get that out of a pairE 	of HSJ50s.  That is about how fast you could stream a backup of that C 	hyper (assuming you can write it somewhere that fast).  Also, note = 	that queue length of 4, IO size of 16 blocks (8K) results in & 	20 MByte/sec (maybe a month-end run). 	    				Rob    ------------------------------  % Date: Sat, 23 Aug 2003 13:37:02 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> + Subject: Re: Looking for TK50s in the UK... ' Message-ID: <3F47B44E.F4B6B8F0@fsi.net>    Rod Chapman wrote: > 3 > Does anyone know where I can buy blank TK50 tapes 0 > (also known as DLT type I) in the UK?  Believe8 > it or not, we still have to supply our biggest VAX/VMS > user using TK50 media...  G Somehow, that's not surprising. I expect the need will exist long after " Alpha has gone the way of the VAX.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 23 Aug 2003 14:02:26 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) + Subject: Re: Looking for TK50s in the UK... 3 Message-ID: <dEY0phvoxDwl@eisner.encompasserve.org>   o In article <cf2c6063.0308230900.38e6a6b3@posting.google.com>, rod.chapman@praxis-cs.co.uk (Rod Chapman) writes: 3 > Does anyone know where I can buy blank TK50 tapes 0 > (also known as DLT type I) in the UK?  Believe8 > it or not, we still have to supply our biggest VAX/VMS > user using TK50 media... > 
 >  Thanks,8 >   Rod Chapman, Praxis (yes..that's _the_ Praxis! :-) )  H VMS itself is no longer shipped on TK50 tapes, because Digital EquipmentF could not get blank media in sufficient quantity.  But for the size ofE demand you describe (i.e., not all customers), eBay would be a viable  source.   E Rod, is Spark Inspector available for VAX VMS ?   What VMS versions ?    ------------------------------  # Date: Sat, 23 Aug 2003 19:11:28 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>+ Subject: Re: Looking for TK50s in the UK... 0 Message-ID: <3F47BA80.3E30FC8A@blueyonder.co.uk>   Rod Chapman wrote: > 3 > Does anyone know where I can buy blank TK50 tapes 0 > (also known as DLT type I) in the UK?  Believe8 > it or not, we still have to supply our biggest VAX/VMS > user using TK50 media...  9 You may be lucky and find one of the office suppliers who   carries them, but I'm not sure.   9 I am wondering just what the maintenance charges for your : customer's TK50 drives  are, and how many CDrom drives one9 could buy  out of that. OK, there are bootability issues, : but there are ways round that like booting VMS on a Wintel? running a VAX emulator and booting enough of a cluster to copy   data from CD to a local disk.       > 
 >  Thanks,8 >   Rod Chapman, Praxis (yes..that's _the_ Praxis! :-) )   --   tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Sat, 23 Aug 2003 22:18:21 -0500  From: msell <msell@pdq.net> - Subject: M7940/DHV-11, TU58, and a VAX 11/750 - Message-ID: <bi9anp$qc1@library2.airnews.net>    Hello everyone,     G I was wondering if anyone on this newsgroup has a suggestion on how to  - connect a dual-TU58 tape drive to a QBus VAX?   G I found a document describing how to configure an M7940 (which I have)  H for operation with a TU58, and my QBus VAX (MicroVAX II) recognizes the C drives (the interface, actually) - so the card at least appears to   partially operate.  D However - when I try to "mount" a tape, there is no activity on the H interface (M7940) as tested with a logic probe. This board came from an G LSI-11/03 in a VAX 11/780, and I believe this board to be operational.  I The point of using the logic probe is to make a cable to attach the dual   TU58 to the M7940.  B The goal is to create a set of boot tapes for my 11/750. I have a H working VAXCluster with 4000's, and two MicroVAXes, and I'd like to get  this 11/750 booting.  < Has anyone here been successful in using a M7940 with TU58s?  & Am I barking up the wrong tree?    : )   Thanks for the insight!      	- Matt              --     -------------------------  Matthew Sell
 Programmer UNIX System Administrator  On Time Support, Inc.  www.ontimesupport.com   4 Join the Metrology Software Discussion List METLIST!Z http://www.ontimesupport.com/index.php?module=pagemaster&PAGE_user_op=view_page&PAGE_id=24    ( AIX - Linux - NetBSD - OpenVMS - Solaris  3 Expand Your Horizons - Stop Looking Through Windows    ------------------------------  # Date: Sun, 24 Aug 2003 05:34:09 GMT ; From: "John Gemignani, Jr." <jon-nope@thiswontworkossc.net> 1 Subject: Re: M7940/DHV-11, TU58, and a VAX 11/750 ; Message-ID: <l7Y1b.6823$Nc.4290541@news1.news.adelphia.net>   ( "msell" <msell@pdq.net> wrote in message' news:bi9anp$qc1@library2.airnews.net...  >  > Hello everyone,  >  > H > I was wondering if anyone on this newsgroup has a suggestion on how to/ > connect a dual-TU58 tape drive to a QBus VAX?  > H > I found a document describing how to configure an M7940 (which I have)I > for operation with a TU58, and my QBus VAX (MicroVAX II) recognizes the D > drives (the interface, actually) - so the card at least appears to > partially operate. > E > However - when I try to "mount" a tape, there is no activity on the I > interface (M7940) as tested with a logic probe. This board came from an H > LSI-11/03 in a VAX 11/780, and I believe this board to be operational.J > The point of using the logic probe is to make a cable to attach the dual > TU58 to the M7940. > C > The goal is to create a set of boot tapes for my 11/750. I have a I > working VAXCluster with 4000's, and two MicroVAXes, and I'd like to get  > this 11/750 booting. > > > Has anyone here been successful in using a M7940 with TU58s? > ( > Am I barking up the wrong tree?    : ) >  > Thanks for the insight!  >  >  > - Matt >   H If it's from an 11/03 then is it possible that it's limited to an 18-bit address space?L You may need to do up a special driver to copy data within the 22-bit (QBUS) space L into an 18-bit space for the operation to work.  The device probably doesn't doL scatter/gather, so that would mandate that you pull all of the data together	 manually.   I I'm trying to remember what the device was that had this issue years ago.A Was itF the TK50 or its predecessor?  It was a tape driver from what I recall.   -Johnw   ------------------------------  # Date: Sat, 23 Aug 2003 18:36:32 GMTp2 From: "Ken Farmer" <kfarmer@NOSPAM.SpyderByte.com>- Subject: Re: OpenVMS Jobs, OpenVMS.org etc...C= Message-ID: <QuO1b.25323$5H4.667171@twister.southeast.rr.com>k  ; "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in messagef8 news:20030823030707.638.qmail@web20209.mail.yahoo.com...
 > Well People  >t< > I belive that the best way to survive in the OpenVMS World; > is we stay united. I suggested to K.Farmer to OpenVMS.org 9 > become the central repository of all OVMS professionalsn; > worldwide. Would be an association ? Paying 25 dollars... = > I dont  know ! But would be the perfect place for us to pute; > our resumes and to be the central hub for the hiring guysv: > contact us ! As I said to him we would have our profiles3 > like www.openvms.org/profile/fcardoso (etc...)  !r< > May be installing a Instanting Message server software for4 > our communty to have online contact worlwide  !!!!    H I also think it's a great idea.  I've just been too busy.  Now is a good  time to start a formal dialogue.  L I've created a mailing list to discuss an "OpenVMS Professionals Directory."" openvms-pros-subscribe@openvms.org  0 OpenVMS Professionals Directory Project webpage:/ http://www.openvms.org/pages.php?page=ovms-prost  > Join if you wish to discuss this issue or want to participate.   Kend   -- Kenneth Farmer  <><r OpenVMS.org  |  dcl.OpenVMS.orga   ------------------------------  # Date: Sun, 24 Aug 2003 01:21:37 GMT.2 From: "Ken Farmer" <kfarmer@NOSPAM.SpyderByte.com>( Subject: OpenVMS Professionals Directory= Message-ID: <BqU1b.25959$5H4.705528@twister.southeast.rr.com>t  C Started a new thread to get that subject line in front of everyone.s    ; "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in messagec8 news:20030823030707.638.qmail@web20209.mail.yahoo.com...
 > Well Peoplet > < > I belive that the best way to survive in the OpenVMS World; > is we stay united. I suggested to K.Farmer to OpenVMS.org89 > become the central repository of all OVMS professionals ; > worldwide. Would be an association ? Paying 25 dollars...e= > I dont  know ! But would be the perfect place for us to puts; > our resumes and to be the central hub for the hiring guys : > contact us ! As I said to him we would have our profiles3 > like www.openvms.org/profile/fcardoso (etc...)  !u< > May be installing a Instanting Message server software for4 > our communty to have online contact worlwide  !!!!    H I also think it's a great idea.  I've just been too busy.  Now is a good  time to start a formal dialogue.  L I've created a mailing list to discuss an "OpenVMS Professionals Directory."" openvms-pros-subscribe@openvms.org  0 OpenVMS Professionals Directory Project webpage:/ http://www.openvms.org/pages.php?page=ovms-prosm  > Join if you wish to discuss this issue or want to participate.   Kene   -- Kenneth Farmer  <><  OpenVMS.org  |  dcl.OpenVMS.org    ------------------------------  # Date: Sat, 23 Aug 2003 21:29:47 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network>t Subject: Re: OpenVMS Securityn; Message-ID: <f1R1b.6669$Nc.4189820@news1.news.adelphia.net>I   Bob Ceculski wrote:   E > one more great place to get a really good idea is the cert site ...  > @ > OpenVMS the last time I checked had 31 certs in 13 years while <snip>E Those certs in the search statistics also include were the operating nE system was listed as "Not  Vulnerable", and some of them are summary e- lists, which duplicate all previous listings.   L So a search for any platform will reveal more certs than there actually are.  A You have to look at each cert document and manually weed out the e0 duplicates and summaries to get the true totals.  < Statistics are useful, but raw data is sometimes misleading.   -Johnt wb8tyw@qsl.network Personal Opinion Onlym   ------------------------------  + Date: Sat, 23 Aug 2003 20:44:49 +0000 (UTC)DP From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)N Subject: overhead with host-based volume shadowing for a one-member shadow-set$ Message-ID: <bi8jo1$pss$1@online.de>  I My hobbyist cluster has simple SCSI disks, a mixture of "internal" disks oE in computers and expansion boxes (disks and boxes: some DEC and some  F third-party) and SBBs in BA35x boxes (and one DSSI disk inside a VAX) F which have a SCSI connection to only one node and are MSCP-served and $ mounted by all nodes in the cluster.  E Some of these are shadow sets; members of system-disk shadow sets areiD connected to the same node (and controller and box) while members ofC non-system-disk shadow sets are connected to different nodes (and,  I though theoretically not necessary with a split-bus box, different boxes  D and of course different controllers).  Some are non-shadowed single  disks.  D My mid-term goal is to have every logical disk be a two-membered (orD even three-membered) shadow set.  To make things easier when working< towards this goal (i.e. less to think about when editing theE disk-mounting procedure which is run at startup), I'm thinking about ,G making ALL disks a shadow set, even if at the moment there is only one   member.H  * How much additional overhead will this be?  H All communication (MSCP, cluster communication, "normal" network trafficD (LAT, TCPIP, no DECnet at the moment) is via a 10 Mb/s LAN, and eachC node has only ethernet card.  While I'm sure that the network is a oE bottleneck in some cases, at the moment it's not a problem and other wI bottlenecks (such as memory in an ALPHA which has only 64 MB) are higher r  on my list of problems to solve.  I Also, I now mount all disks on all nodes, even if these disks are rarely cF accessed from a node other than the one they are connected to via the E SCSI bus (for example, system disks).  It can be convenient, though,  G for example when running a procedure to show information about all the rC disks.  Again, how much additional overhead is this (i.e. the MSCP a	 serving).e   ------------------------------  % Date: Sat, 23 Aug 2003 23:31:43 -0400n. From: Glenn Everhart <Everhart-nospam@gce.com> Subject: Re: Praxisa+ Message-ID: <bi9alp$lrs$1@bob.news.rcn.net>t  9 Praxis appeared on the old Structured Languages SIG tapes 9 and might have existed on some vms tapes. Suggest lookingn6 at sigtape archives and also see if Tim Shoppa's site ) (trailing-edge.com) has a copy somewhere.l  B As I recall it was submitted in binary (vax, natch) and the source was not complete.i Glenn Everhart   Tom Linden wrote: E > Why do you want it?  I ahven't heard of anybody using for ~20 years. >  >  >>-----Original Message-----2 >>From: Barry Treahy, Jr. [mailto:Treahy@MMaz.com]' >>Sent: Friday, August 22, 2003 9:08 AM0 >>To: Info-VAX@Mvb.Saic.Com  >>Subject: Re: Praxis  >> >>* >>I see the list still has their Trekkie's >>& >>Wasn't what I had in mind though :-) >> >>Barryi >> >>Tom Crabtree wrote:o >> >>G >>>Actually, it is rich in dylithium crystals, necessary for warp driveiG >>>engines.  The mines are said to be hazardous and unhealthy places to  >>>work in.  >>>l
 >>>Engage! >>>r >>>T >>>i >>>Jeff Cameron wrote: >>>c >>>.L >>>>On 8/21/03 3:00 PM, in article 3F454100.5090400@MMaz.com, "Barry Treahy,! >>>>Jr." <Treahy@MMaz.com> wrote:- >>>> >>>> >>>>8 >>>>>Anyone recall Praxis and where that can be located? >>>>> 
 >>>>>Barry >>>>>  >>>>>m >>>>F >>>>Yes, it is a moon rich in deuterium, and can be located around the >>>>KlingonS >>>>homeworld of Kronos. >>>> >>>  >>>i >>-- >>@ >>Barry Treahy, Jr                       E-mail: Treahy@MMaz.com@ >>Midwest Microwave                          Phone: 480/314-1320@ >>Vice President & CIO                         FAX: 480/661-7028 >> >> >> >> >>---c( >>Incoming mail is certified Virus Free.< >>Checked by AVG anti-virus system (http://www.grisoft.com).B >>Version: 6.0.511 / Virus Database: 308 - Release Date: 8/18/2003 >> >  > --- ( > Outgoing mail is certified Virus Free.< > Checked by AVG anti-virus system (http://www.grisoft.com).B > Version: 6.0.511 / Virus Database: 308 - Release Date: 8/18/2003 >    ------------------------------    Date: 23 Aug 2003 17:00:26 -0700- From: djcameron60616@yahoo.com (stg-delfuego)g7 Subject: Question about a free shell provider - VisTechh= Message-ID: <6b28e43f.0308231600.2ba19449@posting.google.com>Q  C While attempting over the past five days to vistech.net to set up arE free account, their banner states that they are down (since the 15th)oA for maintenance, and should be back up shortly.  The demo accountw? works fine, and I tried to see if a few utilities could show meiE whether or not my old account was still in the sysuaf.dat (not that IoE would have rights to see it, but worth a shot).  So I guess what I'mnv asking is :A  4 1.  Anyone here know when their site may be back up?D 2.  Anyone know if their are commands I can run in the system to seeF if my account (signed up a few months back up and got too busy to keep; up with it), only withing good taste and legal constraints?dF 3.  Anyone know of a good site offering free user accounts on VMS that
 is available?   $ I'll google in the meantime, but TIA   Delfuego   ------------------------------    Date: 23 Aug 2003 17:03:00 -0700- From: djcameron60616@yahoo.com (stg-delfuego)a6 Subject: Regarding previous post of mine about VisTech= Message-ID: <6b28e43f.0308231602.2db8a422@posting.google.com>l  E 1.  Please forgive the horrible spelling/grammar on the first post, I @ wasn't paying attention to the keyboard while typing it. (or the screen for that matter)gF 2.  I was attempting to telnet to the site where I recieved the banner about NEWUSER being down.    HTH-   Delfuego   ------------------------------  % Date: Sat, 23 Aug 2003 13:35:36 -0500y1 From: "David J. Dachtera" <djesys.nospam@fsi.net>e; Subject: Re: setting up a LAT service via a terminal servere' Message-ID: <3F47B3F8.4B4B7AAC@fsi.net>r  / Phillip Helbig---remove CLOTHES to reply wrote:  > [snip]H > I think I DID install the terminal-server software on at least one VMSC > machine; do I have to "do anything" to get it to serve a MOP bootS
 > request?  . If you have DECnet started, try commands like:   $ dir mom$system:pr*.sys   $ dir mom$load:pr*.sys  H ...and see if the load file(s) is(are) in there somewhere. If so, enableH line service on the DECnet circuit associated with the LAN segment whereH he terminal server(s) is(are) connected. Run SYS$SYSTEM:NCP and see HELP@ SET CIRCUIT. Remember to DEFINE the CIRCUIT to have line service enabled.  G As long as the machine has line service enabled on that DECnet circuit,lE it should serve up whatever load file is requested to whatever servert- requested it, if the VMS system has the file.r  H You shouldn't need to define the server as a DECnet node unless you want* to be able to use the LOAD command in NCP.  H The syntax is a bit different in LANCP. I've never set up LANCP that wayE (never had the need); so, perhaps someone else will chime in with the  command sequence...    -- e David J. Dachterae dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/d   ------------------------------  # Date: Sun, 24 Aug 2003 03:43:13 GMTp2 From: rdeininger@mindspring.com (Robert Deininger); Subject: Re: setting up a LAT service via a terminal serverbL Message-ID: <rdeininger-2308032350540001@user-105n96l.dialup.mindspring.com>  D In article <bi81j0$8hd$2@online.de>, helbig@astro.multiCLOTHESvax.de1 (Phillip Helbig---remove CLOTHES to reply) wrote:s   >In articlehB ><rdeininger-2308031116490001@user-105n8c2.dialup.mindspring.com>,6 >rdeininger@mindspring.com (Robert Deininger) writes:  > L >> I think the 200/MC needs a MOP server to load its software when it powersM >> on.  I believe it is/was on the layered product CDs.  I don't remember the,J >> file name; if you configure your MOP server (via LANCP or either DECnetJ >> flavor) and watch OPCOM, you'll see a MOP request when you power on the
 >> server. >hH >I think I DID install the terminal-server software on at least one VMS C >machine; do I have to "do anything" to get it to serve a MOP boot t	 >request?o  H I think I have always done this by running cluster_config_lan.com.  That! lets you set up the MOP server.  D  J MOP service for terminal servers is "anonymous".  The terminal server asksF for a file by name, and any available MOP server that has the file canG supply it.  This is simpler than MOP load for VMS satellites, where thehE MOP server needs to know the satellite's ethernet address in advance.   G The crucial steps are 1) setting up MOP service on 1 or more VMS nodes,yG and 2) putting the load image for the terminal server in the MOM$SYSTEMt
 directory.   ------------------------------  % Date: Sat, 23 Aug 2003 18:09:33 -0400v* From: JF Mezei <jfmezei.spamnot@istop.com>< Subject: Re: The Dell Effect (Was: Re: Will OpenVMS I64 ...)) Message-ID: <3F47E614.167F79B6@istop.com>    Charlie Hammond wrote:H > The cost of Dell harwware qualified to run OpenVMS may be more or lessK > than the cost of similar HP hardware.  Similarly, the cost of maintaininga9 > the system (hardware and software) may be more or less.   L Dell doesn't have the dinausaur distribution system called "the channel". HPG still does.  So even of Dell were to have equivalent margins to HP, itst9 products would be cheaper because Dell is more efficient.   M In the wintel business, both Dell and HP and assemblers/distributors. Dell is E JIT, whereas HP is still with the old inventory-channel-retail setup.   K Also, because Dell builds to order, you will often find circumstances where.L the product you need will be much cheaper simply because you can get the oneF high end feature you want without the rest of a collection of high endL features you'd be forced to get on an HP model where your desired feature is< only available on high end models loaded with many features.  J Another advantage of the Dell model is that dell can move much faster withH changes since it doesn't have any inventory to get rid of before the new) models can be sold or new prices offered.    ------------------------------  % Date: Sat, 23 Aug 2003 17:35:08 -0400v( From: David Froble <davef@tsoft-inc.com>: Subject: Re: Will OpenVMS I64 run on a Dell PowerEdge 3250, Message-ID: <3F47DE0C.5050301@tsoft-inc.com>   healyzh@aracnet.com wrote:   >  Will it even run OpenVMS?    P Totally useless thread.  With the exception of a few and VMS engineering, there L are no IA-64 systems that will run VMS, because there isn't a VMS for IA-64  available yet.     Dave   -- e4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------   End of INFO-VAX 2003.467 ************************=+v-uc_Q!++4SSKu\C:3䒭I2d(	C6i}RC;-RbK'H(#^V);V.IKVawegɦ\KaFӓ	ƉPE!3=nQr8C&VeQJ˩HqOknd8RtIԉ_YDX!ԩ
	
L x
I>QƋJ/#yu.etU[<̘, |9%.rU ^-0{+I
jKI^<
35~xcբbR΂ʧ1FJf\x܉#[f8]b&.eX
Rtbg$H3[hthWXR'Ɓbӓ> ';&lhjTCݝVahP.l0}Pa6t,.7*I_jᕖQzE˷m*W
tQL46,nlje7olR@;7mt@76CGr̕IK73j]kI1mںfc^%bO[/mb(UFcV)IeQV(*lkcbE Jwa{AZS	Q"gk1[#)E9=C.SZgXmme]c( Nnpp/7^wsH~LK<wt#ÿp?`o6o%^~B)Q/bey؜GQZ֩H
%>okypx,&9)b8|%e?v}i¨֩a)?ӣorODp=Sv)hT 'bz]	.xZoDiN
P(FR8>gɋUEaoeUzzvBtI	kUz1BѪ3@mF֑H/Ѷ`w!<a=!RWFgGM