dave@lsuc.uucp (David Sherman) (09/09/88)
Does anyone read tor.news.stats on a regular basis?
This is a reminder that it's there to examine when your feed
dries up. Sites in Toronto are supposed to forward it to other
uucp sites, specifically including those that they don't have
news connections to. That way, if site x feeds y, and y's
news users wonder why there's suddenly no news, they can examine
recent tor.news.stats postings (arriving from site z) and
quickly determine whether x is getting news and not sending it
on, or not getting it at all. The major news relay sites should
be posting there. It's useful for tracking adequacy and frequency
of people's mail connections, too.
Someone's suggested that the date be included in the Subject: header
for lsuc's postings. I prefer not to, since if I do find myself
reading t.n.s., I can more easily skip what I don't want with rn's
k command. Anyone care either way about this major burning issue?
(Certainly seems worth starting a major flame war about.)
David Sherman
--
{ uunet!attcan att pyramid!utai utzoo } !lsuc!davehenry@utzoo.uucp (Henry Spencer) (09/11/88)
In article <1988Sep9.155024.4475@lsuc.uucp> dave@lsuc.uucp (David Sherman) writes: >...The major news relay sites should >be posting there... Blush. My apologies to all for utzoo not doing its duty in this regard. I haven't yet had a chance to get uucp/news administration fully automated since our Sun conversion. -- NASA is into artificial | Henry Spencer at U of Toronto Zoology stupidity. - Jerry Pournelle | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
cks@ziebmef.uucp (Chris Siebenmann) (09/12/88)
In article <1988Sep9.155024.4475@lsuc.uucp> dave@lsuc.uucp (David Sherman) writes: >Does anyone read tor.news.stats on a regular basis? Ayup, I certainly do. And very helpful indeed it is, if only to see how much more news I need to expire to catch up to my news feed. >Someone's suggested that the date be included in the Subject: header >for lsuc's postings. I prefer not to, since if I do find myself >reading t.n.s., I can more easily skip what I don't want with rn's >k command. Anyone care either way about this major burning issue? >(Certainly seems worth starting a major flame war about.) There's no need to put it in the Subject: header, since it's already in the first line. What I'd like to have is the software that produces these reports for Cnews sites, so I can monitor the newsflow through the Ziebmef. I looked at this at one point, and was stumped as to how to pull the name of the system that sent me a particular article out of the available logs. tor.news.tools, anyone? :-) [Btw, Cnews is great; it's small, fast, works well, and is easily modifiable. Great stuff.] -- "Oh BLESS you, sir! The ANGEL OF DEATH was after me just as SURE as you're standing there, yes he WAS!" Chris Siebenmann uunet!utgpu!{ontmoh!moore,ncrcan}!ziebmef!cks cks@ziebmef.UUCP or .....!utgpu!{,ontmoh!,ncrcan!brambo!}cks
molnar@gpu.utcs.toronto.edu (Tom Molnar) (09/13/88)
In article <1988Sep9.155024.4475@lsuc.uucp> dave@lsuc.uucp (David Sherman) writes:
# Someone's suggested that the date be included in the Subject: header
# for lsuc's postings. I prefer not to, since if I do find myself
# reading t.n.s., I can more easily skip what I don't want with rn's
# k command. Anyone care either way about this major burning issue?
# (Certainly seems worth starting a major flame war about.)
My comment: No! I am only interested in specific site postings, the
rest are in a kill file. The date is in the appropriate header.
Keep it there.
Tom
--
Tom Molnar
Unix Systems Group, University of Toronto Computing Services.msb@sq.uucp (Mark Brader) (09/13/88)
> > Someone [that was me, actually] > > 's suggested that the date be included in the Subject: header > > for lsuc's postings. I prefer not to, since if I do find myself > > reading t.n.s., I can more easily skip what I don't want with rn's > > k command. > My comment: No! I am only interested in specific site postings, the > rest are in a kill file. This is an argument for "yes"; *adding* the date to the Subject line does not make it any harder to use kill files to kill on a site basis, as long as the sitename stays there too. Now consider the = command. This becomes much more valuable if the date is there, because you can see when things are out of sequence. On the other hand, as dave said, the k command becomes harder to use. So does the ^N command. I don't use either; I use =, pick specific messages of interest, and c the rest. But then, I'm not a newsadmin anyway. Mark Brader Summary of issue: Fix FORTRAN-8x. SoftQuad Inc., Toronto Committee Response: This proposal contains utzoo!sq!msb insurmountable technical errors. msb@sq.com -- X3J11 responses to 2nd public review
henry@utzoo.uucp (Henry Spencer) (09/14/88)
In article <1988Sep12.011926.2711@ziebmef.uucp> cks@ziebmef.UUCP (Chris Siebenmann) writes: > What I'd like to have is the software that produces these reports for >Cnews sites, so I can monitor the newsflow through the Ziebmef. I >looked at this at one point, and was stumped as to how to pull the >name of the system that sent me a particular article out of the >available logs. "Fixed in the next release." I know, that's not too helpful right now. :-( -- NASA is into artificial | Henry Spencer at U of Toronto Zoology stupidity. - Jerry Pournelle | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
dave@lsuc.uucp (David Sherman) (09/15/88)
henry@utzoo.uucp (Henry Spencer) writes: > cks@ziebmef.UUCP (Chris Siebenmann) writes: > > What I'd like to have is the software that produces these reports for > >Cnews sites, so I can monitor the newsflow through the Ziebmef. I > >looked at this at one point, and was stumped as to how to pull the > >name of the system that sent me a particular article out of the > >available logs. > > "Fixed in the next release." I know, that's not too helpful right now. :-( We're running C news, so I did this ages ago (with's Geoff's OK as to the change in log file format). Here's the opening part of the history() routine in rnews/history.c on lsuc: history(hdrs) /* generate history entries */ register struct headers *hdrs; { register char *lcmsgid; int status = 0; time_t now; long pos; char msgid[MAXLINE]; /* Message-ID sans \t & \n */ char expiry[MAXLINE]; /* Expires sans \t & \n */ datum msgidkey, posdatum; char fromwhere[64]; /* David Sherman, Feb/88 */ register char *p; /* David Sherman, Feb/88 */ /* strip \n & \t to keep history file format sane */ sanitise(hdrs->h_msgid, msgid, sizeof msgid); sanitise(hdrs->h_expiry, expiry, sizeof expiry); /* TODO: is the 3rd parameter needed anymore? */ timestamp(stdout, &now, (char **)NULL); /* David Sherman, Feb 13/88: show got/utzoo or whencever */ if((p=index(hdrs->h_path, '!')) == NULL) strcpy(fromwhere, "local"); else /* copy up to the '!', not including */ { strncpy(fromwhere, hdrs->h_path, p-hdrs->h_path); fromwhere[p-hdrs->h_path] = '\0'; } (void) printf(" got/%s %s", fromwhere, msgid); /* NB: no newline */ And here's a tail -2 of our log file at this very moment: Sep 15 07:50:51 got/utmanitou <1695@wyse.wyse.com> ncrcan sickkids Sep 15 08:01:41 got/local <1988Sep15.075928.26959@lsuc.uucp> utzoo hcr utgpu ncrcan oscvax attcan tmsoft ryesone forgen maccs utcsri utmanitou sq sickkids David Sherman P.S. if you want a version of scanlog which recognizes this format, ask me. -- { uunet!attcan att pyramid!utai utzoo } !lsuc!dave