Path: utzoo!attcan!uunet!husc6!mailrus!tut.cis.ohio-state.edu!bloom-beacon!bu-cs!purdue!decwrl!hplabs!hp-sdd!ncr-sd!se-sd!rns
From: rns@se-sd.sandiego.NCR.COM (Rick Schubert)
Newsgroups: news.admin
Subject: How do I do "sendme"?
Summary: Need to have lost news articles resent
Keywords: sendme inews control
Message-ID: <1523@se-sd.sandiego.NCR.COM>
Date: 20 May 88 17:37:51 GMT
Reply-To: rns@se-sd.sandiego.NCR.COM (Rick Schubert)
Organization: NCR Corp. Systems Engineering, San Diego
Lines: 40
References:

(I hope this is the right newsgroup for this plea for help)

We have lost a large (~1000) news articles because we ran out of
file system space (we have had the same problem in the past from
running out of inodes due to an OS bug, but the file system exhausted
was real -- we received a large volume of backlogged news articles).

The symptom is that the file for the article has size 0.  By looking for
files of size 0 and comparing this with the history file, I have been
able to determine the message-ids of the lost articles (also the newgroups,
if these are necessary).

I was told that the way to have these articles re-sent is to send a "sendme"
control message.  I have found practically no information on "sendme", but
by piecing things together, I have come up with the following approximation:

	/usr/lib/news/inews -c sendme <<EOF
	162@checkers.UUCP
	375@cloud9.UUCP
	377@cloud9.UUCP
		.
		.
		.
	543@utacs.UTA.FI
	EOF

When I execute this, I get the error message:

	inews: sendme: Can't find sys record for

What am I doing wrong?  Am I even close?  Thanks in advance.

P.S.  Is there a way in the future to avoid the problem to begin with?
i.e. can the news software be set up to detect that the articles are
not being received properly (open fails for inode problem; write fails
for disk-space problem) and have them retransmitted them later?

Thanks again.

-- Rick Schubert (rns@se-sd.sandiego.NCR.COM)