INFO-VAX Mon, 09 Apr 2007 Volume 2007 : Issue 196 Contents: CSWB on EV4 (21064) Re: CSWB on EV4 (21064) Re: OPCOM and TCPIP Services Re: OPCOM and TCPIP Services Re: OPCOM and TCPIP Services Re: OPCOM and TCPIP Services Re: OPCOM and TCPIP Services Re: OPCOM and TCPIP Services Re: OT: Global Warming script to FTP file Re: vmsbackup 4.1.1 problem X25 packet tracing in VMS ---------------------------------------------------------------------- Date: Mon, 9 Apr 2007 11:04:11 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: CSWB on EV4 (21064) Message-ID: I'm putting together an inventory of my machines (11 ALPHAs and 15 VAXes) since I now have too many to keep all the details in my head. In particular, I am hoping that a PW 433 AU might be powerful enough to run CSWB. Now, I have a 5305 (AS 1200) as a satellite which I boot just when I need CSWB. Thus, in the inventory I'll note how much RAM, what CPU etc and test CSWB. Just to see what happens, I fired up CSWB on a Digital 3000/600, i.e. a 21064 (EV4) first-generation ALPHA. 192 MB RAM. CSWB runs and is actually usable, although it is slow. (On the 1200, the CSWB images are installed, which was NOT the case on the 3000.) CPU usage was high but not at maximum, there was no paging and memory use was high but there was still some left (both in the modified list and in the free list). Quite frankly, I was surprised, especially since performance might be even better if the images are installed. It is still too slow to use normally, but for a quick look at a page which actually needs features CSWB has and my other browsers (LYNX, Mosaic, Netscape) don't, it is faster to fire up CSWB on the 3000 than booting the 1200. (The 1200 has its own monitor; for the 3000, I set the display to a VAXstation 4000/90 with a huge Digital colour monitor which has a better picture than the much newer "PC-style connector" monitor I bought in 1997 with a 255/233.) ------------------------------ Date: 9 Apr 2007 09:24:20 -0700 From: "Rich Jordan" Subject: Re: CSWB on EV4 (21064) Message-ID: <1176135860.730715.39020@q75g2000hsh.googlegroups.com> On Apr 9, 6:04 am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig--- remove CLOTHES to reply) wrote: > I'm putting together an inventory of my machines (11 ALPHAs and 15 > VAXes) since I now have too many to keep all the details in my head. In > particular, I am hoping that a PW 433 AU might be powerful enough to run > CSWB. Now, I have a 5305 (AS 1200) as a satellite which I boot just > when I need CSWB. Thus, in the inventory I'll note how much RAM, what > CPU etc and test CSWB. > > Just to see what happens, I fired up CSWB on a Digital 3000/600, i.e. a > 21064 (EV4) first-generation ALPHA. 192 MB RAM. CSWB runs and is > actually usable, although it is slow. (On the 1200, the CSWB images are > installed, which was NOT the case on the 3000.) CPU usage was high but > not at maximum, there was no paging and memory use was high but there > was still some left (both in the modified list and in the free list). > > Quite frankly, I was surprised, especially since performance might be > even better if the images are installed. It is still too slow to use > normally, but for a quick look at a page which actually needs features > CSWB has and my other browsers (LYNX, Mosaic, Netscape) don't, it is > faster to fire up CSWB on the 3000 than booting the 1200. (The 1200 has > its own monitor; for the 3000, I set the display to a VAXstation 4000/90 > with a huge Digital colour monitor which has a better picture than the > much newer "PC-style connector" monitor I bought in 1997 with a > 255/233.) I use CSWB on my PWS600au regularly. I do have 1GB memory and the 4MB cache module; I think that cache will make a substantial difference in performance. That system is down right now so I can't get the relevant UAF settings but they are set based on the Java recommendation (big). Your 433au is obviously slower but I'd imagine it will do well enough, especially if you have the cache and enough memory. I think I remember Hoff recommending an EV56 or better for decent CSWB performance; I'm pretty sure even the older Intel-bridge-chip based PWS boxes had the EV56 processors in them. Java is pokey on my 600au, only slightly faster with the "Fast VM" than the regular one. Java apps, even complex ones, do run but its pretty leisurely. MOdestly usable, but not ideal. One thing; if you look at larger pages that require scrolling, a decent video card can make a big difference. When i upgraded from a 4D10T (Elsa SYnergy) to a 4D20, scrolling performance and video quality improved markedly. ------------------------------ Date: Mon, 9 Apr 2007 01:05:05 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: OPCOM and TCPIP Services Message-ID: <07040901050507_202002DA@antinode.org> From: JF Mezei > Remote nodename: hn.kd.dhcp > Remote node id: 1026966587 > At the time of an attack, OPCOM messages should provide usable information right > off the bat instead of providing untrustable or unusable information. In the > above example, there is no usuable information. 1026966587 (decimal) = 3D36443B (hexadecimal) = 61.54.68.59 Write yourself a program? I'll admit that a format more easy to interpret would be an improvement, but let's not get too excited about the complete lack of "usuable information". ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Mon, 09 Apr 2007 02:08:19 -0500 From: Dan Foster Subject: Re: OPCOM and TCPIP Services Message-ID: In article <07040901050507_202002DA@antinode.org>, Steven M. Schweda wrote: > From: JF Mezei > >> Remote nodename: hn.kd.dhcp >> Remote node id: 1026966587 > >> At the time of an attack, OPCOM messages should provide usable information right >> off the bat instead of providing untrustable or unusable information. In the >> above example, there is no usuable information. > > 1026966587 (decimal) = 3D36443B (hexadecimal) = 61.54.68.59 > > Write yourself a program? Easy enough. $ create convert-dec-ipaddr.pl ($a,$b,$c,$d)=unpack("C4",pack("N", $ARGV[0])); print "$a.$b.$c.$d \n"; $ perl convert-dec-ipaddr.pl 1026966587 61.54.68.59 I'm a little rusty with my DCL now. Something like this: $ hexip=f$fao("!XL", f$integer(p1)) $ a=f$extract(0,2,hexip) $ b=f$extract(2,2,hexip) $ c=f$extract(4,2,hexip) $ d=f$extract(6,2,hexip) $ write sys$output "Hex is: ''a' ''b' ''c' ''d'" $ write sys$output "Converted IP is: ''f$fao("!ZB.!ZB.!ZB.!ZB",a,b,c,d)'" ...but the last line is wrong somehow since it doesn't do proper conversion. Hmm. Also, probably should be specified as !2(ZB) to zero-pad it. -Dan ------------------------------ Date: Mon, 09 Apr 2007 05:10:47 -0400 From: JF Mezei Subject: Re: OPCOM and TCPIP Services Message-ID: <15014$461a033e$cef8887a$23070@TEKSAVVY.COM> Steven M. Schweda wrote: > 1026966587 (decimal) = 3D36443B (hexadecimal) = 61.54.68.59 > > Write yourself a program? When you are under an IP attack and need to add a filter on your router, do you REALLY want to be bothered with having to run some program and transcribe the unusable opcom address to get a usable one when OPCOM should/could really provide you with a immediatly usable IP address ? ------------------------------ Date: Mon, 9 Apr 2007 09:35:02 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: OPCOM and TCPIP Services Message-ID: <07040909350271_202002DA@antinode.org> From: JF Mezei > Steven M. Schweda wrote: > > 1026966587 (decimal) = 3D36443B (hexadecimal) = 61.54.68.59 > > > > Write yourself a program? > > When you are under an IP attack and need to add a filter on your router, do you > REALLY want to be bothered with having to run some program and transcribe the > unusable opcom address to get a usable one when OPCOM should/could really > provide you with a immediatly usable IP address ? As I recall, I didn't say that it was convenient, but I also didn't say anything like "no usable information" (principally because _I_ can use the information there). It wasn't intended for publication, so I don't claim that it's efficient, or free from spelling errors, or that I actually use it much, but I did find this procedure lying around: $! 10 April 2004. SMS. $! $! Translate decimal IP address (32-bit) into dotted octets. $! $! $ ip = p1 $! $! Arrange compensation for sign bit if address >= 128.0.0.0. $! $ inc3 = 0 $ if (ip .lt. 0) $ then $ inc3 = 128 $ ip = ip+ 2147483647 + 1 $ endif $! $! Separate octet values. $! $ oc0 = ip $ oc1 = oc0/ 256 $ oc0 = oc0- oc1* 256 $ oc2 = oc1/ 256 $ oc1 = oc1- oc2* 256 $ oc3 = oc2/ 256 $ oc2 = oc2- oc3* 256 $ oc3 = oc3+ inc3 $! $! Reverse byte order for proper display. $! $ write sys$output f$fao( "!ZB.!ZB.!ZB.!ZB", oc0, oc1, oc2, oc3) $! alp $ @ DECIP.COM 1026966587 59.68.54.61 It's also handy for translating the addresses in some junk e-mail ("PayPal Security Check", and so on) where the evil Web server is disguised this way. But if you think that throwing a tantrum here will help more than a polite complaint to the free non-support Web page, I wouldn't dream of stopping you. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode.org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: 9 Apr 2007 10:02:15 -0700 From: "Peter Weaver" Subject: Re: OPCOM and TCPIP Services Message-ID: <1176138135.063566.270280@y80g2000hsf.googlegroups.com> On Apr 9, 5:10 am, JF Mezei wrote: > Steven M. Schweda wrote: > > 1026966587 (decimal) = 3D36443B (hexadecimal) = 61.54.68.59 > > > Write yourself a program? > > When you are under an IP attack and need to add a filter on your router, do you > REALLY want to be bothered with having to run some program and transcribe the > unusable opcom address to get a usable one when OPCOM should/could really > provide you with a immediatly usable IP address ? Rather than having to remember which utility you wrote to run, you could just do a ping 1026966587 PING 1026966587 (61.54.68.59): 56 data bytes 64 bytes from 61.54.68.59: icmp_seq=0 ttl=43 time=302 ms But when I find someone attacking my system with FTP I just let them carry on since they are not ever going to find a password that would let them log into an "Administrato" account on my system :) Usually they give up after 2,833 attempts. On February 14th I had someone try 40,084 attempts to log into Administrato before I blocked the IP since I figured the report I would fire off to the ISP would be too large if I let the idiot carry on much longer. Anyone know what the 2,833 most popular passwords for Adminsitrator actually are? -- www.weaverconsulting.ca CHARON-VAX CHARON-AXP DataStream Reflection PreciseMail HP Commercial Hardware ------------------------------ Date: Mon, 09 Apr 2007 17:42:10 GMT From: Tad Winters Subject: Re: OPCOM and TCPIP Services Message-ID: Dan Foster wrote in news:slrnf1kkv1.32q.usenet@zappy.catbert.org: > In article <07040909350271_202002DA@antinode.org>, Steven M. Schweda > wrote: >> $! >> $ ip = p1 >> $! >> $! Arrange compensation for sign bit if address >= 128.0.0.0. >> $! >> $ inc3 = 0 >> $ if (ip .lt. 0) >> $ then >> $ inc3 = 128 >> $ ip = ip+ 2147483647 + 1 >> $ endif >> $! >> $! Separate octet values. >> $! >> $ oc0 = ip >> $ oc1 = oc0/ 256 >> $ oc0 = oc0- oc1* 256 >> $ oc2 = oc1/ 256 >> $ oc1 = oc1- oc2* 256 >> $ oc3 = oc2/ 256 >> $ oc2 = oc2- oc3* 256 >> $ oc3 = oc3+ inc3 > > Wouldn't it be easier to do: > > $ hexip=f$fao("!XL", f$integer(p1)) > $ oc1=f$extract(0,2,hexip) > $ oc2=f$extract(2,2,hexip) > $ oc3=f$extract(4,2,hexip) > $ oc4=f$extract(6,2,hexip) > > ? > > (Was the !XL directive new to a relatively recent version of OpenVMS?) > > -Dan No, I'm sure it's been around since at least V5.5-2. The below works for me. (In fact, tested with a negative value.) $ hexed_ip = f$fao("!XL", f$integer(p1)) $ hexed_octet_1 = "%x" + f$extract(0, 2, hexed_ip) $ hexed_octet_2 = "%x" + f$extract(2, 2, hexed_ip) $ hexed_octet_3 = "%x" + f$extract(4, 2, hexed_ip) $ hexed_octet_4 = "%x" + f$extract(6, 2, hexed_ip) $ octet_1 = 'hexed_octet_1' $ octet_2 = 'hexed_octet_2' $ octet_3 = 'hexed_octet_3' $ octet_4 = 'hexed_octet_4' $ write sys$output octet_1, ".", octet_2, ".", octet_3, ".", octet_4 $ exit ------------------------------ Date: 9 Apr 2007 08:38:42 -0700 From: "roger" Subject: Re: OT: Global Warming Message-ID: <1176133122.174855.127930@l77g2000hsb.googlegroups.com> Please take this discussions somewhere else. Global warming may not be as bad as scientists think, because the world may have far less fossil fuels (including oil) than previously reported. Do the research. The global warming predictions assume business as usual, which will not be the case if the world as already peaked in global oil production. There are many newsgroups and sites devoted to theses topics so please don't fill up comp.os.vms with this. Watch http://www.youtube.com/watch?v=kaKaH2UTsNE or read www.theoildurm.com Also don't forget to "Ask Dr. Bartlett what time it is"... ------------------------------ Date: 9 Apr 2007 10:26:27 -0700 From: "zoohornrollo@gmail.com" Subject: script to FTP file Message-ID: <1176139587.689896.79560@q75g2000hsh.googlegroups.com> I am new to VMS and need a simple callable script (like unix script) to ftp a file to a windows box on our network. I could write this if I were on Unix but nor VMS. If I get a base I think I can modify it to my own uses. Thank you for any help given. ------------------------------ Date: 9 Apr 2007 08:55:35 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: vmsbackup 4.1.1 problem Message-ID: In article <4617dcb4$0$90272$14726298@news.sunsite.dk>, =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: > I believe a VMS BACKUP tape is in standard ANSI labeled tape format, > so any tape program that are capable of reading a tape should > be able to get the blocks of the save set of the tape. VMS BACKUP uses standard ANSI labels, but does not always use standard error handling. I've seen many instances where I could not COPY a saveset from tape and had to use TCOPY instead. ------------------------------ Date: 9 Apr 2007 06:12:43 -0700 From: harikrishna.hari@gmail.com Subject: X25 packet tracing in VMS Message-ID: <1176124363.436376.108780@n76g2000hsh.googlegroups.com> Can any any one help me how to get the X25 Packet format in VMS 6 Version? I mean what are the commands to be given to get the Packet format coming into VMS Alpha system How to trace the Packet. Regards, Harry ------------------------------ End of INFO-VAX 2007.196 ************************