jbuck@epimass.UUCP (Joe Buck) (05/12/86)
Sometime over the weekend, it seems that site 'hpitg' shipped
everything in its spool back out to the net, 10-20 Mbytes
worth. You can recognize these articles on your machine by
their "Path:" headers, which have the form
Path: ...!hpitg!site!user@site
Normally, Path headers have no @'s in them, and "site" can be
any site on the net.
Fortunately, the full 20 Mb won't go everywhere, because spool
directories should fill up all over the net first, and the same
message-IDs were used, so I hope some of the backbone sites keep
news around for a long time (so the old Message-IDs will still
be in the history file). Expect a larger phone bill this month.
--
- Joe Buck {ihnp4!pesnta,oliveb}!epimass!jbuck
Entropic Processing, Inc., Cupertino, California
Better living through entropy!Tim@lll-lcc.UUcp (Tim Kehres) (05/13/86)
Cc: In article <221@epimass.UUCP> you write: >Sometime over the weekend, it seems that site 'hpitg' shipped >everything in its spool back out to the net, 10-20 Mbytes >worth. You can recognize these articles on your machine by >their "Path:" headers, which have the form > > Path: ...!hpitg!site!user@site > >Normally, Path headers have no @'s in them, and "site" can be >any site on the net. > >Fortunately, the full 20 Mb won't go everywhere, because spool >directories should fill up all over the net first, and the same >message-IDs were used, so I hope some of the backbone sites keep >news around for a long time (so the old Message-IDs will still >be in the history file). Expect a larger phone bill this month. > >-- >- Joe Buck {ihnp4!pesnta,oliveb}!epimass!jbuck > Entropic Processing, Inc., Cupertino, California > >Better living through entropy! If the message ID's were left the same, most sites should drop most of the articles as "duplicate". Please let me know if this is not the case. Tim Kehres Control Data Corporaton / Lawrence Livermore National Laboratory ---------------------------------------------------------------- UUCP: {idi,ihnp4!lll-lcc}!styx!kehres ARPA: kehres@lll-tis-b.ARPA AT&T: (415) 423-6252
wex@milano.UUCP (05/13/86)
In article <221@epimass.UUCP>, jbuck@epimass.UUCP (Joe Buck) writes: > [...] it seems that site 'hpitg' shipped everything in its spool back out to > the net [...]. You can recognize these articles on your machine by > their "Path:" headers, which have the form > > Path: ...!hpitg!site!user@site > > Normally, Path headers have no @'s in them, and "site" can be > any site on the net. This is not strictly true, expecially for sites like ours that talk to both UUCP and ARPA. All Usenet originated mail (and much news) appears here as Path: ...!site!user@im4u.uucp (im4u is a UT machine which generously feeds us UUCP news and mail (bless'em!)) In addition, when vnews shows news articles here, it often shows From: user@site.uucp (thus making the `r' function useless). Anyway, the point is that the simple presence of the @ may not be a good indicator and Joe's advice may mislead some people. -- Alan Wexelblat ARPA: WEX@MCC.ARPA UUCP: {ihnp4, seismo, harvard, gatech, pyramid}!ut-sally!im4u!milano!wex "We do not act as a result of consideration, but as a way of being."
joel@gould9.UUCP (Joel West) (05/15/86)
In article <221@epimass.UUCP>, jbuck@epimass.UUCP (Joe Buck) writes: > Sometime over the weekend, it seems that site 'hpitg' shipped > everything in its spool back out to the net, 10-20 Mbytes > worth. You can recognize these articles on your machine by > their "Path:" headers, which have the form > > Path: ...!hpitg!site!user@site I executed the following shell script to count and pull these: #!/bin/sh L=$HOME/HPITG cd /usr/spool/news/net find . -exec grep -s '^Path:.*!hpitg' {} \; -exec echo {} \; >$L wc $L df rm `cat $L` df By my estimate, I found 1,054 articles totalling ca. 2.5 mb that reached here. Some came hplabs->sdcrdcf, others hplabs->hao->seismo etc. I'd previously heard claims that Notes was superior in some aspects, but it seems apparent that it is unable to keep up with the demands of the current net traffic. -- Joel West (619) 457-9681 CACI, Inc. Federal, 3344 N. Torrey Pines Ct., La Jolla, CA 92037 {cbosgd, ihnp4, pyramid, sdcsvax, ucla-cs} !gould9!joel joel%gould9.uucp@NOSC.ARPA
jbs@mit-eddie.MIT.EDU (Jeff Siegal) (05/15/86)
In article <8605130111.AA01357@lll-tis-b.ARPA> Tim@lll-lcc.UUcp (Tim Kehres) writes: >>[...] >>Fortunately, the full 20 Mb won't go everywhere, because spool >>directories should fill up all over the net first, and the same >>message-IDs were used, so I hope some of the backbone sites keep >>news around for a long time (so the old Message-IDs will still >>be in the history file). >>[...] > >If the message ID's were left the same, most sites should drop most >of the articles as "duplicate". Please let me know if this is not the >case. It is certainly not the case. Many, many duplicate messages made it here, and beyond. Jeff Siegal
krossen@bbnccp.ARPA (Ken Rossen) (05/15/86)
In article <553@gould9.UUCP> joel@gould9.UUCP (Joel West) writes: > I'd previously heard claims that Notes was superior in some aspects, It is. > but it seems apparent that it is unable to keep up with the demands of > the current net traffic. I'm as annoyed with this ridiculous behaviour on the part of HP as almost anyone (except perhaps the backbone site administrators, bless 'em), but the latest round of notesfiles-bashing it has prompted is misguided. Many sites run notesfiles on USENET quite transparently and without incident. Notably, University of Illinois, NYU, Intermetrics and Convex. The extremely poor citizenship of some notesfiles sites shouldn't lead clear-thinking people to make generalizations about the program. You don't see Orphaned Responses from U of I any more. I suppose it's time to remind people that, with notesfiles, you're ALWAYS shown your article right after it's submitted, that you generally have to be looking a group before you post to it, that cross-posting to a ton of groups is much more complicated in notesfiles. These are benefits to the net, not to mention the additional benefits to the notesfiles reader, which are many. When people bash notesfiles by whining that they gave us "Orphaned Response," I wonder why nobody recalls who gave us entire articles of quoted text and "REPLACE THIS LINE WITH YOUR MESSAGE," or postings to net.general about lost earrings in the West Parking Lot. I have to use news now, and there are certainly things I like about it, notably support for regional distribution in non-regional newsgroups. Rn makes things much more palatable, of course. But brain-damage on the part of HP (and Gould Ft. Lauderdale -- guys, clean up those articles from "daemon@houligan"!) nonwithstanding, I think we'd be suffering less from information overload if we were all using notesfiles. -- Ken Rossen ...!{ihnp4,harvard,seismo}!bbnccv!krossen ____or____ krossen@ccp.bbn.com -or- krossen@bbnccp.arpa
zben@umd5.UUCP (Ben Cranston) (05/16/86)
This brings up an interesting point... Suppose I am a commercial entity,
providing a service that competes directly with a service of USENET. Many
potential customers may refuse my services based on the fact that they
already receive the services in question from USENET.
If my research lab gets a machine and connects it to USENET, then sabotages
the performance of the network under the guise of system testing, then
perhaps some of those customers could be persuaded to buy my services
after all.
This is *NOT* an accusation against HP or ATT or anybody else who may
periodically accidently flood the net with random strangeness. It is just
a random speculation triggered by the recent stuff from HP...
--
"We're taught to cherish what we have | Ben Cranston
by what we have no longer..." | zben@umd2.umd.edu
...{seismo!umcp-cs,ihnp4!rlgvax}!cvl!umd5!zben earle@smeagol.UUCP (Greg Earle) (05/17/86)
In article <1447@milano.UUCP>, wex@milano.UUCP writes: > > Normally, Path headers have no @'s in them, and "site" can be > > any site on the net. > This is not strictly true, expecially for sites like ours that talk to both > UUCP and ARPA. All Usenet originated mail (and much news) appears here as > Path: ...!site!user@im4u.uucp That's not a very good addressing scheme for replying to mail addressed like that! > In addition, when vnews shows news articles here, it often shows > From: user@site.uucp > (thus making the `r' function useless). Uh, sorry, that's the fault of your Netnews installer. Vnews has an option to use either the From: line or the Path: line as the address to send to for return mail. `uumail' V 3.0 was just posted to mod.sources. I suggest you have it installed. If you do, and can use pathalias generated paths, then the From: field is not "useless" at all; your smart mailer can handle the translation for you. That is what my machine does right now. I saw your article as "From: wex@milano.UUCP", and if I had replied to you directly via `r', the mail would have been sent via uucp mailer to wex@milano.UUCP. Since `uumail' is my uucp mailer *it* gets the wex@milano line and has to do something with it - which it does; it looks up milano in the pathalias database and substitutes the optimal path to your machine instead of the random-walk path that is probably present in the Path: header. It then calls uux to do the queuing. Since most sites that post are in the UUCP maps, this scheme works flawlessly 90% of the time - and it saves $$$$. Now, if only everyone else would get on the stick and implement the pathalias - uumail - uuhosts combo, we'd be getting somewhere ... -- Greg Earle UUCP: sdcrdcf!smeagol!earle; (new!!) attmail!earle JPL ARPA: elroy!smeagol!earle@csvax.caltech.edu Were these parsnips CORRECTLY MARINATED in TACO SAUCE?
henry@utzoo.UUCP (Henry Spencer) (05/20/86)
> Now, if only everyone else would get on the stick and implement the > pathalias - uumail - uuhosts combo, we'd be getting somewhere ... Do let us know how to implement pathalias on a 16-bit machine. It can't be done. (We have local >16-bit sites we could lean on, but others don't.) -- Join STRAW: the Society To Henry Spencer @ U of Toronto Zoology Revile Ada Wholeheartedly {allegra,ihnp4,decvax,pyramid}!utzoo!henry
ronc@fai.UUCP (Ronald O. Christian) (05/21/86)
>When people bash notesfiles by whining that they gave us "Orphaned >Response," I wonder why nobody recalls who gave us entire articles of >quoted text and "REPLACE THIS LINE WITH YOUR MESSAGE," or postings to >net.general about lost earrings in the West Parking Lot. ** Now now, those are caused by uneducated users, not by bugs in the software, as are the Orphans. You have some good other points, (even though they're surrounded by fightin' words like 'bashing' and 'whining') but this is comparing apples to waffles. Ron -- -- Ronald O. Christian (Fujitsu America Inc., San Jose, Calif.) seismo!amdahl!fai!ronc -or- ihnp4!pesnta!fai!ronc Oliver's law of assumed responsibility: "If you are seen fixing it, you will be blamed for breaking it."
jbs@mit-eddie.MIT.EDU (Jeff Siegal) (05/23/86)
In article <6709@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes: >[...] >Do let us know how to implement pathalias on a 16-bit machine. It can't >be done. (We have local >16-bit sites we could lean on, but others don't.) >[...] Yes, but you can use the -l option to compile the database on a large machine and transfer it to the 16bitter. If there isn't a larger machine nearby, perhaps someone on the net would be willing to compile pathalias databases, as a public service (obviously no more than one or two per day). Jeff Siegal
mark@cbosgd.UUCP (Mark Horton) (05/28/86)
In article <2070@mit-eddie.MIT.EDU> jbs@mit-eddie.UUCP (Jeff Siegal) writes: >In article <6709@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes: >>Do let us know how to implement pathalias on a 16-bit machine. It can't >>be done. (We have local >16-bit sites we could lean on, but others don't.) > >Yes, but you can use the -l option to compile the database on a large >machine and transfer it to the 16bitter. If there isn't a larger >machine nearby, perhaps someone on the net would be willing to compile >pathalias databases, as a public service (obviously no more than one >or two per day). Lauren Weinstein supports RFC 976 on a floppy based IBM PC with MS DOS. If he can do it, surely an 11/70 running UNIX can. The basic trick is that if you don't have room on your machine for a complete map of the world (or if your machine is a PC that goes through a gateway for everything anyway, and you don't want to maintain a copy of a world map on every PC you have) you use an incomplete map, and take advantage of the domain structure. Your software has a table showing some local configuration information (your PC's name, your neighboring PC's names, your gateway's name) and a default entry that says "anything I don't recognize, I send to the gateway." Then you put the full table (or at least a fuller table) on the gateway. Pathalias isn't strictly necessary, although you can still run pathalias on a subset of the world map to get the same results. pathalias itself will run on an IBM PC (I know of one PC XT running PC/IX with pathalias), as long as the data it's fed is fairly small. Mark
ben@catnip.UUCP (Bennett Broder) (05/28/86)
In article <2070@mit-eddie.MIT.EDU> jbs@mit-eddie.UUCP (Jeff Siegal) writes: >In article <6709@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes: >>[...] >>Do let us know how to implement pathalias on a 16-bit machine. It can't >>be done. (We have local >16-bit sites we could lean on, but others don't.) >>[...] Can it? I thought I caught a reference in an article a couple of weeks ago that noted that someone was able to hack the source. If so, and anyone has a copy I would greatly appreciate it if you could pass it on. Thanks Ben Broder {ihnp4,decvax} !hjuxa!catnip!ben {houxm,topaz}/ P.S. If there is really some limitation that prevents it from working on 16bit machines, couldn't it be hacked to use temporary disk files?