1 INFO-VAX	Sat, 08 Nov 2003	Volume 2003 : Issue 619       Contents:+ Re: AlphaServer ES47/ES80/GS1280 User Guide  Re: An update on 'CPU WARS' . Re: Daylight to Standard Time Change via Batch. Re: Daylight to Standard Time Change via Batch. Re: Daylight to Standard Time Change via Batch# Re: DECwindows screen on an XP box? # Re: DECwindows screen on an XP box? # Re: DECwindows screen on an XP box? # Re: DECwindows screen on an XP box? # Re: DECwindows screen on an XP box? 9 How to determine how much TAPE-GB is required for DISK-GB = Re: How to determine how much TAPE-GB is required for DISK-GB = Re: How to determine how much TAPE-GB is required for DISK-GB  Re: IA64 TPC results Re: IA64 TPC results Re: IA64 TPC results Re: IA64 TPC results Re: mounting tape drive? Re: mounting tape drive?9 Newbie problem, act 2: recovering an old VMS installation = Re: Newbie problem, act 2: recovering an old VMS installation = Re: Newbie problem, act 2: recovering an old VMS installation  Re: Performance problem , RE: SEDT text editor for MS-DOS for download, RE: Where's the full list of ports to IA-64?, Re: Where's the full list of ports to IA-64? [OT] p*o*e*t*s of the day  Re: [OT] p*o*e*t*s of the day  Re: [OT] p*o*e*t*s of the day  Re: [OT] p*o*e*t*s of the day  [very OT] Matrix Revolutions  F ----------------------------------------------------------------------  # Date: Fri, 07 Nov 2003 22:49:55 GMT A From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> 4 Subject: Re: AlphaServer ES47/ES80/GS1280 User Guide; Message-ID: <nkVqb.1154$Sc.8524984@news-text.cableinet.net>   K Yes. Please use PDF in preference to HTML. Way better for off-line reading,  even on an iPAQ handheld.    Thanks, Colin.   --     Hope this helps, Colin. ) colin DOT butcher AT xdelta DOT co DOT uk L Systems Archaeologist - Investigation & troubleshooting of older systems and	 networks.    ------------------------------   Date: 7 Nov 2003 14:42:53 -0600 4 From: kuhrt@nospammy.encompasserve.org (Marty Kuhrt)$ Subject: Re: An update on 'CPU WARS'3 Message-ID: <a$KjWtWtfoYI@eisner.encompasserve.org>   W In article <3FAB26A0.3010305@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:  > Larry Kilgallen wrote: > X >> In article <3FA9EBFE.30604@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes: >>   >>  E >>>There is a rumor that Microsoft told Intel that another 64 bit x86 R >>>implementation would not be supported by windows, thus implying that any 64 bitO >>>x86 CPU produced by Intel had better be compatable with HAMMER.  While Intel S >>>would prefer to have uncompatable CPUs and use it's muscle to do in AMD, if it's O >>>CPU would not run windoz, then that duck is so dead that it was never alive.  >>>  >>  J >> Thus it behooves Intel to support Itanic to the hilt in order to retain >> their proprietary advantage.  >>   > R > Come on Larry.  The issue here is market share.  Even Intel seems to be backing Q > off of IA-64 being a commodity.  If Intel conceeds the desktop, and much else,  R > to AMD, well, I know that you know just how big DEC was at one time.  AMD could P > become the new gorilla, and Intel lucky to be the chimp.  Hey, maybe HP could 3 > buy them also, after they're run into the ground.  > S > Intel appears to be trying to prop up IA-64, but, if AMD sells thousands of CPUs  P > for each IA-64 sold, Intel will not have anything with which to prop up IA-64. > P > Recent events point to Opteron and Athlon/64 being poised to do to Intel what Q > IA-32 did to so many of the competition.  Intel can't stand for this to happen  P > without some kind of a fight.  They're not as stupid as Palmer/Capellas/et al. >   I It's Palmer on the board of AMD?  You'd think that would make the success / of Opteron and Athlon/64 questionable, at best.    ------------------------------  % Date: Fri, 07 Nov 2003 14:15:04 -0500 * From: JF Mezei <jfmezei.spamnot@istop.com>7 Subject: Re: Daylight to Standard Time Change via Batch ) Message-ID: <3FABEF37.E370BDD2@istop.com>    Charlie Hammond wrote:= > I regard my posting as more of a comment than a suggestion. 4 > I am neither a lawyer nor a medical practitioner,   ' But did you stay at a Holiday Inn ?????    :-)    ------------------------------  % Date: Sat, 08 Nov 2003 00:20:30 +0800 , From: Paul Repacholi <prep@prep.synonet.com>7 Subject: Re: Daylight to Standard Time Change via Batch - Message-ID: <87y8us1775.fsf@prep.synonet.com>   , JF Mezei <jfmezei.spamnot@istop.com> writes:  D > Having one fixed time helps a lot. And that is GMT. (England is at > GMT+1 during summer BTW).   ? I could do a Stan about the difference between GMT and the UTCs  but will refrain I think...   D (If you use TAI you don't even have to worry about leap seconds :) )   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Fri, 07 Nov 2003 19:42:52 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) 7 Subject: Re: Daylight to Standard Time Change via Batch 1 Message-ID: <0BSqb.8671$55.2231@news.cpqcorp.net>   V In article <3FABEF37.E370BDD2@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: >Charlie Hammond wrote: > >> I regard my posting as more of a comment than a suggestion.5 >> I am neither a lawyer nor a medical practitioner,   > ( >But did you stay at a Holiday Inn ?????   No.   @ The worst hotel/motel experience of my life was at a Holiday InnD in Altoona, PA.  I'll spare you the details, but I ended up leaving E at 3:00 AM in disgust.  My complaint to the Holiday Inn orgainzation  B recieved a more-or-less form letter saying that I had to deal withB the local innkeeper.  The local inkeeper's only responce was that D I should have been more polite to the clerk when I left.  No attemptG to address or acknowledge the issues that caused my premature dparture.   3 So, when I have a choice, I now avoid Holdiay Inns.  (I usually do have a choice.)    --  J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  * Date: Fri, 7 Nov 2003 22:54:57 +0000 (UTC), From: lewis@mazda.mitre.org (Keith A. Lewis), Subject: Re: DECwindows screen on an XP box?. Message-ID: <boh7s1$f07$2@newslocal.mitre.org>   Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> writes in article <3FABD71C.1D5C8412@blueyonder.co.uk> dated Fri, 07 Nov 2003 17:36:55 GMT: >Didier Morandi wrote: >>  1 >> You talked to Keith Lewis before posting this?  >> :-) > 5 >Nope, his post popped up while I was composing mine. 4 >I went thru freeware Xserver angst last winter :-). > 6 >So, two of us independently suggest VNC. You do need  >a 3rd box though.  + Only until somebody ports Xvnc to VMS!  :^)   A Not a small job though.  And I've played with Xvnc on Solaris and L Alpha/Linux, they aren't anywhere near as reliable as Xvnc on x86/Linux.  SoH even after the port is done, shaking out the bugs would be a significant task.     0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  * Date: Fri, 7 Nov 2003 23:04:32 +0000 (UTC), From: lewis@mazda.mitre.org (Keith A. Lewis), Subject: Re: DECwindows screen on an XP box?. Message-ID: <boh8e0$f07$3@newslocal.mitre.org>   kuhrt@nospammy.encompasserve.org (Marty Kuhrt) writes in article <gCesZCeqOhao@eisner.encompasserve.org> dated 7 Nov 2003 15:02:44 -0600: h >In article <3FABD71C.1D5C8412@blueyonder.co.uk>, Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> writes:8 >> So, two of us independently suggest VNC. You do need  >> a 3rd box though. > B >Is the addition of the third box because the XP box won't do VNC B >right?  I have VNC 3.3.6 on my NT box with the desktop displayed + >on my V7.3 Alpha without the intermediary.   H No, it's because neither VMS nor XP will run Xvnc.  Xvnc is an X displayL server which is also a VNC server.  It serves a virtual X display to VMS (orJ any X-windows based computer) and then makes that available to VNC clients via the VNC protocol.     K The Windows version of VNC is interesting too, it serves the actual Windows K screen to any VNC client, so you can control your PC from your VMS console!   8 realvnc.org can tell you much more about VNC than I can.  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------   Date: 7 Nov 2003 15:02:44 -0600 4 From: kuhrt@nospammy.encompasserve.org (Marty Kuhrt), Subject: Re: DECwindows screen on an XP box?3 Message-ID: <gCesZCeqOhao@eisner.encompasserve.org>   g In article <3FABD71C.1D5C8412@blueyonder.co.uk>, Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> writes:  > Didier Morandi wrote:  >>  1 >> You talked to Keith Lewis before posting this?  >> :-) >>   > 6 > Nope, his post popped up while I was composing mine.5 > I went thru freeware Xserver angst last winter :-).  > 7 > So, two of us independently suggest VNC. You do need   > a 3rd box though.   A Is the addition of the third box because the XP box won't do VNC  A right?  I have VNC 3.3.6 on my NT box with the desktop displayed  * on my V7.3 Alpha without the intermediary.   >  >> D.  >>   >> Tim Llewellyn wrote:  >> >M >> > What I do, configue Xdm on VMS, run a a VNC server process on Linux that R >> > connects to VMS via Xdm, run Vnc client on Windows. Works great and you don'tQ >> > loose your work when Windows crashes/reboots or when you come back next day.  >  > --    > tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Fri, 07 Nov 2003 22:23:56 +0100 " From: Didier Morandi <no@spam.com>, Subject: Re: DECwindows screen on an XP box?2 Message-ID: <3fac0d7c$0$248$636a55ce@news.free.fr>   Thanks Martin,  P That was the problem. I was using a program which was not an X Windows terminal 	 emulator.   9 Now I have my Alpha screen at home on my Presario. Funny!    D.   Martin Kirby wrote:   D > As various others have said. This means that you cannot connect toB > port 6000. If there are access control issues you will get other! > messages that ICO will display.  >  > Likely reasons: 1 >     You don't have an X Server on the XP system    ------------------------------   Date: 7 Nov 2003 19:00:54 -0600 4 From: kuhrt@nospammy.encompasserve.org (Marty Kuhrt), Subject: Re: DECwindows screen on an XP box?3 Message-ID: <sSpBeQ1wTOmb@eisner.encompasserve.org>   ] In article <boh8e0$f07$3@newslocal.mitre.org>, lewis@mazda.mitre.org (Keith A. Lewis) writes:  > kuhrt@nospammy.encompasserve.org (Marty Kuhrt) writes in article <gCesZCeqOhao@eisner.encompasserve.org> dated 7 Nov 2003 15:02:44 -0600: i >>In article <3FABD71C.1D5C8412@blueyonder.co.uk>, Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> writes: 9 >>> So, two of us independently suggest VNC. You do need   >>> a 3rd box though.  >>C >>Is the addition of the third box because the XP box won't do VNC  C >>right?  I have VNC 3.3.6 on my NT box with the desktop displayed  , >>on my V7.3 Alpha without the intermediary. > J > No, it's because neither VMS nor XP will run Xvnc.  Xvnc is an X displayN > server which is also a VNC server.  It serves a virtual X display to VMS (orL > any X-windows based computer) and then makes that available to VNC clients > via the VNC protocol.    > M > The Windows version of VNC is interesting too, it serves the actual Windows M > screen to any VNC client, so you can control your PC from your VMS console!  > : > realvnc.org can tell you much more about VNC than I can.  > I guess I'm not clear on the distinction between Xvnc and the > the windows desktop displayed via Xwindows (DECWindows) on my  VMS box with vncclient.   - The vncclient info for VMS can be found at...   = http://www.uk.research.att.com/archive/vnc/platforms.html#vms    ------------------------------  % Date: Fri, 07 Nov 2003 15:30:05 -0600 ( From: brandon@dalsemi.com (John Brandon)B Subject: How to determine how much TAPE-GB is required for DISK-GB1 Message-ID: <03110715300574@dscis6-0.dalsemi.com>   P A bit back I asked how to determine the amount of MB that was being put to tape N and the answer was a combination of f$getdvi(device,"OPCNT"), /BLOCKS=, and so on.   O Now I was wondering if there is a way to determine how much TAPE will be needed K for DISK, using the variances of number of files, file size, etc.  Is there  such a methodology out there?      For example.  < I have two devices, roughly 34-GB each for a total of 68-GB.H I backup these two devices to a 35/70 (TL89 series) with compression on.D I would expect that the 68-GB of disk would fit onto the 70-GB tape. However I overflow the tape.  O Disk # 1 is a collection of small number of large files.  I do not beleive that # this disk is causing the over-flow.   K Disk # 2 is a collection of a large number of small files.  Roughly 160,000  files on this disk.   K I beleive that the problem is due to the large number of small files on the  disk.   J Is there a way to determine the amount of tape space used by file headers?  J I would like to create a utility that I can run against a disk in order to; calculate the amount of tape space that would be required.    E Yes I know, compression is not always a 2:1 science... just hoping...    TIA      J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 07 Nov 2003 15:45:52 -0600 ( From: brandon@dalsemi.com (John Brandon)F Subject: Re: How to determine how much TAPE-GB is required for DISK-GB1 Message-ID: <03110715455254@dscis6-0.dalsemi.com>    Barry Treahy wrote: I > Binary files are not highly compressible, and if you have ZIP files on  E > the drives, you can actually see negative compression.  That said,  K > blocking with IRGs can contribute to 'tape-waste', so be sure you have a  H > very large /BLOCK.  The number of files should not be a factor though K > the size might be if they are too small for quality compression...  Lets  I > face it though, you are pressing the limit with the drive capacity vs,  2 > disk usage, so why don't you just use two tapes?  G I should have mentioned that I kind of expect this to be an exercise in H futility...  however I kind of would like to do it - ego thing you know.    * I have the block size set to /BLOCK=64000 J Most of the files on the disk are 1 to 15 blocks (EOF) and are ASCII text.    6 Now then, if I had management that agreed with you...          J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 07 Nov 2003 14:55:17 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> F Subject: Re: How to determine how much TAPE-GB is required for DISK-GB' Message-ID: <3FAC14C5.6050602@MMaz.com>    John Brandon wrote:    >Barry Treahy wrote: >    > I >>Binary files are not highly compressible, and if you have ZIP files on  E >>the drives, you can actually see negative compression.  That said,  K >>blocking with IRGs can contribute to 'tape-waste', so be sure you have a  H >>very large /BLOCK.  The number of files should not be a factor though K >>the size might be if they are too small for quality compression...  Lets  I >>face it though, you are pressing the limit with the drive capacity vs,  2 >>disk usage, so why don't you just use two tapes? >>     >> > H >I should have mentioned that I kind of expect this to be an exercise inI >futility...  however I kind of would like to do it - ego thing you know.  >  > + >I have the block size set to /BLOCK=64000  K >Most of the files on the disk are 1 to 15 blocks (EOF) and are ASCII text.  >  > 7 >Now then, if I had management that agreed with you...   >  >  >    > H It is my understanding that single-pass compression used in tape drives F cannot compress as much as a multipass compression like ZIP might.  I G suppose you could try to compress everything that is possible, eat any  H negative compression loss, and see if the net gain is enough to squeeze E everyone on a single tape.  Also, you could try turning off XOR with  E /GROUP=0, though I have no idea how much that will really save you...    Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  # Date: Fri, 07 Nov 2003 18:53:38 GMT & From: Rick Jones <foo@bar.baz.invalid> Subject: Re: IA64 TPC results 2 Message-ID: <SSRqb.8666$YQ7.2569@news.cpqcorp.net>  5 Norman Lastovica <norman.lastovica@oracle.com> wrote:  [posting snipped]   ? I'm not completely up on the "rules and regs" for TPC-C, but my E understanding was that one of them was that one cannot make (public?) # estimates of TPC benchmark results.   E And while the snipped post is unlikely to be construed as any sort of B marketing pitch or whatnot, it would probably be prudent for folksB from companies which are members of the TPC to avoid making public estimates regardless of nature.   ' perhaps I'm just being overly paranoid,   
 rick jones --  G oxymoron n, Hummer H2 with California Save Our Coasts and Oceans plates F these opinions are mine, all mine; HP might not want them anyway... :)A feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...    ------------------------------  % Date: Fri, 07 Nov 2003 13:04:51 -0700 4 From: Norman Lastovica <norman.lastovica@oracle.com> Subject: Re: IA64 TPC results * Message-ID: <3FABFAE3.F260D700@oracle.com>  ; you're right.  I was mostly looking at this as a kinda mind ? game.  technically, one can not publish estimates or that sort  > of thing.  I was just trying to "guess" what the machine could= do.  nothing more, nothing less.  Speaking for me and not for : oracle, and all that (I suppose I should make a sig with a disclaimer just in case).      Rick Jones wrote:  > 7 > Norman Lastovica <norman.lastovica@oracle.com> wrote:  > [posting snipped]  > A > I'm not completely up on the "rules and regs" for TPC-C, but my G > understanding was that one of them was that one cannot make (public?) % > estimates of TPC benchmark results.  > G > And while the snipped post is unlikely to be construed as any sort of D > marketing pitch or whatnot, it would probably be prudent for folksD > from companies which are members of the TPC to avoid making public! > estimates regardless of nature.  > ) > perhaps I'm just being overly paranoid,  >  > rick jones > --I > oxymoron n, Hummer H2 with California Save Our Coasts and Oceans plates H > these opinions are mine, all mine; HP might not want them anyway... :)C > feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...    ------------------------------  $ Date: Fri, 7 Nov 2003 18:20:39 -0500* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: IA64 TPC results 2 Message-ID: <2pSdnSEijpJVtTGiRVn-jg@metrocast.net>  A "Norman Lastovica" <norman.lastovica@oracle.com> wrote in message $ news:3FABD92F.7E737EA1@oracle.com...H > interesting question regarding what could a 64p GS1280 do.  The "best"G > published Alpha number was 230,533 tmpC on a GS320 at 1001MHz.  If we H > look at compaq/hp's estimated performance for the GS320 vs. the GS1280D > at http://h18002.www1.hp.com/alphaserver/performance/perf_tps.html@ > we might be lead to believe that a GS1280 with 64p at 1150 mhzF > is around 2.76 times faster than a GS320 32p at 1000mhz (my personalE > experience leads me to believe that the difference would be quite a  > bit greater).   H Er, yes.  To start with, does anyone *really* think that a Marvel systemJ scales *less* linearly than the GS80/160/320 did (according to those 'TPS'I numbers)?  Well, maybe if you turned off all NUMA optimization and made a ! real effort not to tune things...   J One could instead compare the 8-processor GS1280 ('TPS' score 36,000) withK the 4-processor ES45 1.25 GHz ('TPS' score 18,600).  Expecting nearly twice L the performance with twice the processors doesn't seem unreasonable:  EV7 isD comparable to EV68 in SPECint, for example, and Marvel scales, well,0 marvelously linearly in both SPECint and SPECfp.  K So instead of SWAGging that Marvel might generate something like 636271 for J a tpmC score (based on the 'TPS' comparison with the GS320), consider thatJ scaling linearly up from the ES45's score of 56375 would yield 902000 tpmCG at 64 processors - not bad for an aging, previous-generation core whose  clock rate has been hobbled.  L Assuming linear scaling is too generous, you say?  Maybe - but I've wonderedJ why with a 25% faster processor and twice as much fast off-chip SRAM cacheG the 1.25 GHz ES45 only managed a 12.5% increase in TPC-C score over the D previous 1 GHz ES45 system.  It's not as if the ES45 was starved forI bandwidth, and the 32 GB of memory it had is sufficient to generate TPC-C L scores nearly twice as high for today's high-end Xeons so it doesn't seem to have been starved there either.   G It could, of course, have had something to do with having less than 300 J disks to work with, vs. nearly 500 for the previous ES45 tested.  And bothF systems used 10Krpm disks rather than the 15Krpm disks that had becomeF standard for serious submissions by the time the 1.25 GHz platform wasI tested:  while that doesn't explain why the delta was only 12.5%, it does E help explain why the 1.25 GHz platform's score didn't stand up better  against the other competition.  I For that matter, those 'TPS' numbers HP itself provides indicate that the K 1.25 GHz ES45 has more like 23% higher performance than the 1 GHz ES45.  So ? I've suspected that someone *seriously* interested in getting a I significantly better TPC-C score out of the ES45 probably could have done K so - which would allow one to get close to 1M tpmC in a 64-processor Marvel 7 system even if scaling was noticeably less than linear.   H Of course, that still wouldn't look as good as the POWER4+ TPC-C scores:G with only 32 processors they managed to post a score of 768839 - and it I might be rash to believe that they've really hit their limit yet (even if K they have, POWER5 next year should raise it considerably).  You'd need EV79 F to compete processor-to-processor with POWER4+ (and EV8 to beat POWER5 convincingly).   - bill   ------------------------------  % Date: Fri, 07 Nov 2003 23:41:24 -0500 $ From: Hein <hein_cov@eps.zk.dec.com> Subject: Re: IA64 TPC results . Message-ID: <3FAC73F4.10239FAB@eps.zk.dec.com>   Bill Todd wrote:  J > Er, yes.  To start with, does anyone *really* think that a Marvel systemL > scales *less* linearly than the GS80/160/320 did (according to those 'TPS'K > numbers)?  Well, maybe if you turned off all NUMA optimization and made a # > real effort not to tune things...   > Er, yes. It's the software stupid! Not the hardware (alone). '$ You need Oracle 10g for this result.  ; Look carefully are the full disclosure for the 230K result. E (http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=101061401) P You will notice that this benchmark was run as an 8 instance 'cluster in a box'.  " Look for: p_run_1.ora its friends.B The biggest reason for that was redo log buffer  latch contention.N Not something your average app has to worry about as we are speaking many tens
 of MB/sec.J The cluster in the box was a workaround for that, with a realitively hefty (performance) price tag.H The NUMA attributes of the GS320 were largely/completely avoided in this
 benchmark.  M I think it is safe to say that HP would have published a 32p 400K+ result  if O they could have achieved that 1 1/2 years ago. It wasn't there, but I'd like to L think it was more a software problem than a hardware performance deficiency.O Oracle 10g would have made a difference. That GS1280 is a mighty impressive box   N You correctly point out that the ES45 result is a better base, but not the 56K sybasse,N but rather the 50K Oracle with 1.0Ghz cpus. Now scale that to 1.25 and compare with theP rx5670 4p 1.5Ghz itanium at 136K. The IA64 performs pretty good, better than the5 best Alpha and supposedly this is just the beginning.    fwiw,  Hein.    ------------------------------  # Date: Fri, 07 Nov 2003 18:40:53 GMT # From: hoff@hp.nospam (Hoff Hoffman) ! Subject: Re: mounting tape drive? 1 Message-ID: <VGRqb.8664$pY7.821@news.cpqcorp.net>   X In article <2bRqb.36152$0M6.54780@news1.mts.net>, "Garion" <rwenden@hotmail.com> writes: :I need to do a restore.' :The mka1500 is online but not mounted. K :I don't want to give it a label or log, just want to see whats on the tape * :and possibly copy some of the files off!!     $ HELP MOUNT  I   There are two ways that tapes can be processed, with the native OpenVMS D   file system and with a so-called foreign file system.  Most BACKUPG   savesets are somewhat confusingly created with foreign-mounted tapes, E   using MOUNT/FOREIGN, and BACKUP is the most common means that files    are stored on tape.   L   For details on MOUNT and BACKUP commands, please see the system management<   documentation, and the system management utilities manual.  I   BACKUP can mount tape devices itself, so you may be able to perform the )   restore with the following DCL command:   -     BACKUP MKA1500:/SAVE ddcu:[dir.subdir...]   G   This will mount the device and restore the files of the first saveset F   found on the tape into the specified ddcu:[dir.subdir...] directory.   	--   K   Please remember that multiple exclamation points are often a signature of J   spam, and please remember that spam has created a world-wide shortage ofI   these precious ASCII characters.  Before the advent of spam, there were I   vast herds of these characters, stretching as far as the eye could see. F   (The former portion of this paragraph actually is serious, while the#   latter part obviously isn't.  :-)   ( :Help would be great, i'm a newb to vms.  I   There is little need to identify yourself as a newbie, we have already  F   inferred that based on your question and based on our own individual+   memories of being newbies ourselves.  :-)   I   I would certainly encourage you to read the available documentation for G   OpenVMS -- newbies are appreciated here of course (again, we were all G   newbies once), and newbies that have made a reasonable effort to read F   and understand the manuals are also obviously very much appreciated.8   (I mean no offense here.)  The manuals are located at:  %     http://www.hp.com/go/openvms/doc/    G   The OpenVMS FAQ will also be of interest, as it addresses many of the )   more common questions newbies can have:   %     http://www.hp.com/go/openvms/faq/   N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  $ Date: Fri, 7 Nov 2003 13:51:18 -0600$ From: "Garion" <rwenden@hotmail.com>! Subject: Re: mounting tape drive? 1 Message-ID: <WISqb.36275$0M6.55498@news1.mts.net>   0 "Hoff Hoffman" <hoff@hp.nospam> wrote in message+ news:VGRqb.8664$pY7.821@news.cpqcorp.net... < > In article <2bRqb.36152$0M6.54780@news1.mts.net>, "Garion" <rwenden@hotmail.com> writes:  > :I need to do a restore.) > :The mka1500 is online but not mounted. H > :I don't want to give it a label or log, just want to see whats on the tape, > :and possibly copy some of the files off!! >  >   $ HELP MOUNT > K >   There are two ways that tapes can be processed, with the native OpenVMS F >   file system and with a so-called foreign file system.  Most BACKUPI >   savesets are somewhat confusingly created with foreign-mounted tapes, G >   using MOUNT/FOREIGN, and BACKUP is the most common means that files  >   are stored on tape.  > C >   For details on MOUNT and BACKUP commands, please see the system 
 management> >   documentation, and the system management utilities manual. > K >   BACKUP can mount tape devices itself, so you may be able to perform the + >   restore with the following DCL command:  > / >     BACKUP MKA1500:/SAVE ddcu:[dir.subdir...]  > I >   This will mount the device and restore the files of the first saveset H >   found on the tape into the specified ddcu:[dir.subdir...] directory. >  > -- > J >   Please remember that multiple exclamation points are often a signature ofL >   spam, and please remember that spam has created a world-wide shortage ofK >   these precious ASCII characters.  Before the advent of spam, there were K >   vast herds of these characters, stretching as far as the eye could see. H >   (The former portion of this paragraph actually is serious, while the% >   latter part obviously isn't.  :-)  > * > :Help would be great, i'm a newb to vms. > J >   There is little need to identify yourself as a newbie, we have alreadyH >   inferred that based on your question and based on our own individual- >   memories of being newbies ourselves.  :-)  > K >   I would certainly encourage you to read the available documentation for I >   OpenVMS -- newbies are appreciated here of course (again, we were all I >   newbies once), and newbies that have made a reasonable effort to read H >   and understand the manuals are also obviously very much appreciated.: >   (I mean no offense here.)  The manuals are located at: > ' >     http://www.hp.com/go/openvms/doc/  > I >   The OpenVMS FAQ will also be of interest, as it addresses many of the + >   more common questions newbies can have:  > ' >     http://www.hp.com/go/openvms/faq/  > ( >  ---------------------------- #include' <rtfaq.h> ----------------------------- 3 >     For additional, please see the OpenVMS FAQ --  www.hp.com/go/openvms/faq , >  --------------------------- pure personal# opinion --------------------------- G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    Thank you very much gentlemen,+  the mount dev /override=id seemed to work. ! Now I need perm to mount...doh :)  thanks for the links.    ------------------------------  % Date: Sat, 08 Nov 2003 00:21:20 +0100 - From: WiseLYNX <wiselynx.naima@fastwebnet.it> B Subject: Newbie problem, act 2: recovering an old VMS installation5 Message-ID: <uOVqb.751$ZH3.312@tornado.fastwebnet.it>   
 Hi everybody,   I having for now given up installing a new version of VMS on my old VAX at  I least until I can't find a suitable CDROM drive (see the "Newbie problem  H installing VMS" posted on 03-nov), I was wondering if there is some way A of making it work with the current installation. Let's explain :)   I This machine (a VAXStation 3100/30) belonged once to the university I do  I attend. I tryied to gather other info about it's previous use and setup,  F but I could not find anything.. anyway, I doubt it once was part of a  cluster or something.   6 Now, here is what happens if I ask the system to boot:    >>> boot dka200   -DKA200 F     VAX/VMS Version V5.4-OP  Major version id = 1 Minor version id = 0  I $!  Copyright (c) 1990 Digital Equipment Corporation.  All right reserver   - %DCL-W-ACTIMAGE, error activating image SETP0  -CLI-E-IMGNAME, image file  2 POLCA1$DKA200:[SYS0.SYSCOMMON.][SYSEXE]SETPO.EXE;13 -SYSTEM-F-VA_IN_USE, virtual address already in use    The system hangs here.  H ..and, being I a newbie to both VAXes and VMS.. I can't figure out what D this can mean :( Anyone has a clue? Do you think I can recover this H installation? (I don't need the lastest version of the OS, I was just a ; lot curious of seeing both a VAX at work and VMS at work..)    Thanks in advance :)   --      Wise	   O  LYNX   8 GPG key: http://wiselynx.interfree.it/wiselynx.naima.ascA MD5 checksum: http://wiselynx.interfree.it/wiselynx.naima.asc.md5    -----BEGIN GEEK CODE BLOCK-----  Version: 3.1I GE>$/IT/S d+ s: !a>? C++(+++)>$ U P>+ L++>++++ E--- W++ N ?o ?K w-- ?O ?M G ?V>+ PS+ !PE Y+ PGP+ t- ?5 X- R+(*) tv- b+(++)>+++ DI(++) !D G e++ h!*@  r%>++ y-->+  ------END GEEK CODE BLOCK------    ------------------------------  % Date: Sat, 08 Nov 2003 06:43:38 +0100 " From: Didier Morandi <no@spam.com>F Subject: Re: Newbie problem, act 2: recovering an old VMS installation2 Message-ID: <3fac829a$0$257$636a55ce@news.free.fr>   WiseLYNX wrote:o  / > %DCL-W-ACTIMAGE, error activating image SETP0h > -CLI-E-IMGNAME, image file a4 > POLCA1$DKA200:[SYS0.SYSCOMMON.][SYSEXE]SETPO.EXE;15 > -SYSTEM-F-VA_IN_USE, virtual address already in usem >  > The system hangs here. > J > ..and, being I a newbie to both VAXes and VMS.. I can't figure out what F > this can mean :( Anyone has a clue? Do you think I can recover this  > installation?   G Wise, VMS is not Windoze. VMS is the fruit of an exact science (French cN expression). It has a binary logic (no pun). It works or it doesn't. You will K *never* see in your VMS life such messages like "if your computer does not  M restart after mumblefratz, unplug the power supply, remove the battery, then  G reinstall the battery, then plug the power supply then restart and the CE installation "should" continue" (laptop version :-). Either it had a nO standard/valid/normal installation (choose your favorite) and it will work for fP ages, or it hasn't and either there is a bug in the params and we'll fix it, or O there is a mix of software (which seems the case) and the only wise action, to jC me, is to format the system disk then install a new version of VMS.   O So, as Nov the 3rd thread suggested to me, go and purchase another SCSI CD-ROM  M drive with a new flat cable (one never knows) and do perform the hobbyist CD - installation. Then enjoy.   P Something else noone noted, which has nothing to do with your problem. You have N two SCSI controllers. I would suggest that you remove one and plug your disks K from address 0, to have DKA0, DKA100 and DKA400 (the CD-ROM generally uses tN address 4). Then ask http://www.digital-equipment.com/ to send you a good old Q RZ29 to put as a system disk and to store all your Hobbyist layered products and g
 much more.  * When you have the CD-ROM drive, come back.   My 2 euros.    D. ToulouseS (PS: do you know that none of your site's pages are displayable but the home page?)o   ------------------------------  % Date: Sat, 08 Nov 2003 01:10:54 -0500e* From: JF Mezei <jfmezei.spamnot@istop.com>F Subject: Re: Newbie problem, act 2: recovering an old VMS installation) Message-ID: <3FAC88BC.6CD20CBD@istop.com>n   WiseLYNX wrote: K > $!  Copyright (c) 1990 Digital Equipment Corporation.  All right reservert > / > %DCL-W-ACTIMAGE, error activating image SETP0a > -CLI-E-IMGNAME, image file4 > POLCA1$DKA200:[SYS0.SYSCOMMON.][SYSEXE]SETPO.EXE;15 > -SYSTEM-F-VA_IN_USE, virtual address already in use     Q I would suspect your disk is corrupt, or your hardware config has not too kosher.c  I Try b/1 to get to the SYSBOOT> prompt, if that succeeds, use the command wF SYSBOOT> SET STARTUP_P1 = "MIN"     (not sure of = is required or not) SYSBOOT> SET VAXCLUSTER = 0- SYSBOOT> SET MSCP = 0, SYSBOOT> SET/STARTUP=OPA0: SYSBOOT> CONTINUEb  J This invokes a minimal boot with minimal configuration of devices. It thisN gets you further, it may be a hint. If you get to the $ sign, you will be in a extremely limited environment.  ) You can then try ANA/DISK/REPAIR DKA200: i  O If you can't do any of this. I would consider a low level format of your drive.2M At the >>> prompt, use TEST 75 and follow the prompts. Once the disk has beenfR reformatted, you can then try again to load the B saveset in the conventional way.   ------------------------------  % Date: Fri, 07 Nov 2003 14:23:06 -0500p* From: JF Mezei <jfmezei.spamnot@istop.com>  Subject: Re: Performance problem) Message-ID: <3FABF118.301B9930@istop.com>a   Robert Trawinski wrote:uJ > When I start Motif application from decterm it works fine. When the same > application starts fromcG > CDE front panel the performance is very bad. Account is the same. The " > application is graphic intensive > and memory expensive.n    D Not sure about CDE itself, but on VAX, one has the choice to have anK application spawned as a subprocess of the session manager or as a detached2N process. I believe that detached processes inherit more the PQL process quotas than the UAF process quotas.   ------------------------------  $ Date: Fri, 7 Nov 2003 12:58:05 -06001 From: "Grealy, Patrick" <PGrealy@sph.uth.tmc.edu>i5 Subject: RE: SEDT text editor for MS-DOS for downloadgL Message-ID: <EEC575D39D864C4BBAE8CD309982B0F2071523@sphnt33.sph.uth.tmc.edu>  D Thanks Dave. But I later found a version 5.1 SEDT hidden away in a =6 [PCSA] directory on one of our dormant disks. - Pat G.   > -----Original Message-----9 > From: Dave Weatherall [mailto:djw-nothere@nospam.nohow] ) > Sent: Friday, November 07, 2003 1:32 AMu > To: Info-VAX@Mvb.Saic.Comr7 > Subject: Re: SEDT text editor for MS-DOS for downloadt >=20 >=207 > On Thu, 6 Nov 2003 22:11:03 UTC, "Grealy, Patrick"=20 " > <PGrealy@sph.uth.tmc.edu> wrote: >=20 > > Hi,m? > > Does anyone know where I can download a copy of SEDT for=20 B > MS-DOS. It used to come with Pathworks but I can't find it on=20/ > the CDs we have. The home page for SEDT at=20eB > http://users.rcn.com/anker/sedt/sedt.htm has an online manual=20 > but the download link=20D ftp://ftp.ultranet.com/pub/anker/sedt/sedwin.exe results in a page = unavailable error.	 > Thanks,h > Pat G.  D If nobody else can help drop me a note. I've got a copy of a  DOS=20 version (3.3?) here.   --=20  Cheers - Dave.  ! Real address is something like=20o* djweathATTHEDOMAINattglobalDOTCHARACTERnet   ------------------------------  $ Date: Fri, 7 Nov 2003 10:56:53 -0800# From: "Tom Linden" <tom@kednos.com>.5 Subject: RE: Where's the full list of ports to IA-64?i9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEODIHAA.tom@kednos.com>h  D To be precise, it is a rewrite, not a port.  I hope they know how toB protect against buffer overruns.  Rewrites, of course, always takeH longer to stabilize compared to recompiles.  I think they need to update their roadmaps.p   >-----Original Message-----y- >From: Main, Kerry [mailto:kerry.main@hp.com]i) >Sent: Friday, November 07, 2003 10:55 AMo >To: Info-VAX@Mvb.Saic.Com6 >Subject: RE: Where's the full list of ports to IA-64? >s >e >  >> -----Original Message----- , >> From: Tom Linden [mailto:tom@kednos.com] " >> Sent: November 7, 2003 10:26 AM >> To: Info-VAX@Mvb.Saic.Com8 >> Subject: RE: Where's the full list of ports to IA-64? >> iG >> Some of these are written in PL/I which could explain their absence.-
 >> >MQ Series- >> >DECmessageQ (BEA >> 0 >2 >Tom,  >0 >Re: BEA - reference:v0 >http://h71000.www7.hp.com/news/beamessageq.htmlH >"BEA MessageQ is at the core of tens of thousands of enterprise OpenVMSE >applications and systems, enabling them to communicate reliably with D >each other. To help customers integrate with, or port their OpenVMSG >applications to, the recently announced EV7-based AlphaServer systems,aB >BEA has certified MessageQ on OpenVMS v7.3.1 and will continue toC >support all future AlphaServer platforms. As a result of importantoF >customer feedback that has brought about a concerted effort to fosterI >closer ties with its enormous installed base, BEA has recently committed1H >to porting MessageQ to OpenVMS on Itanium(r)-based systems.....[see url >for rest of release]. >u >Regards >r >Kerry Maini >Senior Consultant >HP Services Canada  >Voice: 613-592-4660 >Fax: 613-591-4477 >Email: kerryDOTmainAThpDOTcom- >(remove the DOT's and AT for email address) e >n >e >x >---' >Incoming mail is certified Virus Free.h; >Checked by AVG anti-virus system (http://www.grisoft.com).3A >Version: 6.0.536 / Virus Database: 331 - Release Date: 11/3/2003t >  ---t& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.536 / Virus Database: 331 - Release Date: 11/3/2003   ------------------------------  % Date: Fri, 07 Nov 2003 14:20:32 -0500 * From: JF Mezei <jfmezei.spamnot@istop.com>5 Subject: Re: Where's the full list of ports to IA-64? ) Message-ID: <3FABF07E.F67C5ECA@istop.com>y  ! VAXman-, @SendSpamHere.ORG wrote:hL > An interesting link.  I found my products listed with (Status in progress)L > whatever the hell that means and I honestly don't know where they obtainedL > the rudimentary information that does appear there.  I do know that nobody4 > from HP has contacted me about future IPF support.    I Well, if they artificially inflated the list, it is in a way a good thingsL since they are at least making an effor to make VMS appear more popular thanJ it is in reality. It doesn't matter whether their primary motivation is toA justify the move to IA64 or whether they want to really push VMS.   H The danger however is that the people who artificially inflated the listK haven't told the rest of VMS management who will then rest on their laurels  thinking all is fine and dandy.n   ------------------------------  % Date: Fri, 07 Nov 2003 22:42:12 +0100s" From: Didier Morandi <no@spam.com>" Subject: [OT] p*o*e*t*s of the day2 Message-ID: <3fac11c4$0$257$636a55ce@news.free.fr>  2 http://h18000.www1.hp.com/info/SP3343/SP3343PF.PDF   D.   ------------------------------  # Date: Fri, 07 Nov 2003 22:31:32 GMTt# From: "John Smith" <a@nonymous.com>a& Subject: Re: [OT] p*o*e*t*s of the dayL Message-ID: <83Vqb.208475$3f.127825@twister01.bloor.is.net.cable.rogers.com>   Didier Morandi wrote:d4 > http://h18000.www1.hp.com/info/SP3343/SP3343PF.PDF    L HP is really missing the boat on this....with an Alpha/VMS on a big cluster,I they'd have enough horsepower to control hundreds of thousands of cameras B simultaneously at ports of entry, bus/rail/subway stations and keyF infrastructure locations. Then they'd just have to wash all the imagesH through facial and license plate capture software to get the bad guys in
 real-time.   [grassy knoll]J Ashcroft, Bush, & Poindexter (ok..so he's out...but be certain he's on theA periphery someplace) would buy all they could get their hands on.p [/grassy knoll]'   ------------------------------  # Date: Fri, 07 Nov 2003 22:50:12 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>& Subject: Re: [OT] p*o*e*t*s of the day0 Message-ID: <3FAC20F0.8295CDED@blueyonder.co.uk>   John Smith wrote:e >  > Didier Morandi wrote:l6 > > http://h18000.www1.hp.com/info/SP3343/SP3343PF.PDF > N > HP is really missing the boat on this....with an Alpha/VMS on a big cluster,K > they'd have enough horsepower to control hundreds of thousands of camerassD > simultaneously at ports of entry, bus/rail/subway stations and keyH > infrastructure locations. Then they'd just have to wash all the imagesJ > through facial and license plate capture software to get the bad guys in > real-time. >   I What makes you so sure some of the ahem less effective implementations ofsG such systems are not on VMS? Just because you have VMS doesn't mean you-  can't implement a crap solution.  M In no way am I implying that the London congestion charging system is on VMS,-R but it definitely doesn't work well and both London Transport and the contractors 0 have been hiring VMS people in london this year.  N btw you probably noticed anyway but the SPD dates from VMS 5.4 era (late 80's)       > [grassy knoll]L > Ashcroft, Bush, & Poindexter (ok..so he's out...but be certain he's on theC > periphery someplace) would buy all they could get their hands on.V > [/grassy knoll]e   -- o tim.llewellyn@blueyonder.co.uk   ------------------------------  # Date: Sat, 08 Nov 2003 01:32:22 GMT # From: "John Smith" <a@nonymous.com>a& Subject: Re: [OT] p*o*e*t*s of the dayF Message-ID: <GIXqb.4647$XDB.1022@news01.bloor.is.net.cable.rogers.com>   Tim Llewellyn wrote: > John Smith wrote:! >> >> Didier Morandi wrote:6 >>> http://h18000.www1.hp.com/info/SP3343/SP3343PF.PDF >>F >> HP is really missing the boat on this....with an Alpha/VMS on a big@ >> cluster, they'd have enough horsepower to control hundreds of9 >> thousands of cameras simultaneously at ports of entry,UB >> bus/rail/subway stations and key infrastructure locations. ThenE >> they'd just have to wash all the images through facial and license ; >> plate capture software to get the bad guys in real-time.i >> >n8 > What makes you so sure some of the ahem less effective > implementations ofE > such systems are not on VMS? Just because you have VMS doesn't meanM > you/" > can't implement a crap solution. >sG > In no way am I implying that the London congestion charging system isoG > on VMS, but it definitely doesn't work well and both London Transport F > and the contractors have been hiring VMS people in london this year. >gD > btw you probably noticed anyway but the SPD dates from VMS 5.4 era
 > (late 80's)   K Of course - I have about 1200 SPD's on a CD from that timeframe and shortly - thereafter. I was being extremely fascetious.g  @ Any o/s can lend itself to foul & odious applications, and flaky8 implementations thereon if not well designed and tested.   ------------------------------  % Date: Fri, 07 Nov 2003 22:49:25 +0100 " From: Didier Morandi <no@spam.com>% Subject: [very OT] Matrix Revolutions 3 Message-ID: <3fac1385$0$8203$636a55ce@news.free.fr>    Hmmm... just out of the cinema.nL I think that Nemo is Smith's brother and the Oracle is Matrix's wife, which 
 would explainu   why Nemo has his power! why the Oracle can see the futuref' why Smith does not succeed to kill Nemo ' why Nemo does not succeed to kill Smithn= why Smith calls the Oracle his "mom" before "terminating" herbN why the Oracle has a deal with that old man at the end of the movie ("what do  you think? I'm not an human")w" why the Matrix accepts Nemo's deal  " (any other forum to discuss that?)   D.   ------------------------------   End of INFO-VAX 2003.619 ************************ww.hp.com/go/openvms/faq/  > ( >  ---------------------------- #include' <rtfaq.h> ----------------------------- 3 >     For additional, please see the OpenVMS FAQ --  www.hp.com/go/openvms/faq , >  --------------------------- pure personal# opinion --------------------------- G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    Thank you very much gentlemen,+  the mount dev /override=id seemed to work. ! Now I need perm to mount...doh :)  thanks f:!9FUh!AKMQrjDf2[6{@4@ClΈa8W>1TA (6*4ޕͭMnn	͊=6ra*[I.CdTt>1"qlϙ:a0JeX\Z"~)n WwxP_&M%xc,)%=XDܵiFA |vOğT%]21mK
~mBuqD=2TD	<yEK8T=ӂMC
^Y[oS2QT'}-s-N^V:+eI0-&镝3kb$lr_'MQ)+L|Z	6fVLw}vGbkı=Nor|}qph~$ot4W(gucko7}]ƛ^#
4;)tǡF<'t#Gj}y"EݛatFj8VMlq#(sZH@7_,o/y.~%Ed[j>al=RhJ,XaFlc)J x
J~9vWk5r3޿/ߺ %UD޸u	sƭו'?%R

Z4))ϯa:	`
IS1q)TƼ9VbQLXzNN?]!TӜ
E?oJMSu_jbJ0=M1J%hz2}ft$Dܯ<%FƁ"'EBϿ>~M&yt:V9prj#_h-.CN9'*dYhx%d>pjE
^	
4/w$b. V+R7Ɍ㉉|;-Mdy>:,#al$
)ĚI1u"u&MDO&f49hrzp01
uh4r<HIp*ya%k#]AAZƌ0Ix#0̠#
"+8,E20l9,nۏuz~BI@z5B^:a_M
34Z~
a*lgmT>F]wl~m2޼ew;wG3 /V*ƍֻfYT-hmoz0ݜ2:2sR_,--}1,kǱ,4\+aRe+:aIvNOUAOFxMMf#
G|EP;5; gp>̣<eР@o݃bX)I
Aed>
B1)?-Xǯ,v_Hmꉷ
w{yߡ*A{T6v~}tXMluEKtST;`H`l hb(?,NJ\\ho>n?^wgFyxb/
@vI:2XE5˓{]OgqTT{2~6]M3}'{Fm?}okk_,|Wn=¯<Γگ~j?>kMxgz6wjo>Vm^tR|a^{=|Z^|>wǵy+ϟGPj[Բ?_輽J/,<mxw
\;޹xKN/'B<G kz
xo{GkЏ>'{-G?ǵ0ZmQ>noøjOZ?־ZϿSO/?s:Ϡ_?+NBZ?^\}[s?}o