home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!CUNYVM.BITNET!BIGCU
- Message-ID: <NETNWS-L%92121517133878@VM1.NODAK.EDU>
- Newsgroups: bit.listserv.netnws-l
- Date: Tue, 15 Dec 1992 18:03:07 EST
- Sender: NETNWS-L Netnews List <NETNWS-L@NDSUVM1.BITNET>
- From: Bill Gruber <BIGCU@CUNYVM.BITNET>
- Subject: Re: VLIB's getting corrupted...
- In-Reply-To: Message of Tue, 15 Dec 1992 11:03:48 EST from <CJS@psuvm.psu.edu>
- Lines: 20
-
- On Tue, 15 Dec 1992 11:03:48 EST Christopher Sacksteder said:
- >I am changing VLIB EXEC to fail on an ADD request when the member name is
- >longer than 222. By experimentation, this seems to be where VLIBCOMP
- >fails in parsing a member header (not sure why). A limit that is easier
- >to identify is 256, which is the maximum sort field length for RXQSORT
- >(called whenever a member is added).
-
- The funny thing about all this is that in the NETNEWS MSGIDS file the
- msgid is truncated to around 133 anyhow. Is there any point to putting
- the long msgid into the VLIB in the first place if it is truncated in
- NETNEWS MSGIDS file?
-
- Also need to point out that even though the msgid is truncated, the LRECL
- of that RECFM=F file continues to grow. Our msgid file is close to
- 400,000 records. Up the lrecl by 10 or so and all of a sudden we're
- talking another 4 meg of disk space for all that extra unused white space.
- (OK, it's unused black space if you are running reverse video).
-
- Bill Gruber
- City University of New York Computer Center
-