Subject: bin/7733: ftp retry option doesn't retry when it probably should
To: None <gnats-bugs@gnats.netbsd.org>
From: Scott Reynolds <scottr@og.org>
List: netbsd-bugs
Date: 06/07/1999 21:20:58
>Number:         7733
>Category:       bin
>Synopsis:       ftp retry option doesn't retry when it probably should
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    bin-bug-people (Utility Bug People)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Jun  7 21:20:01 1999
>Last-Modified:
>Originator:     Scott Reynolds
>Organization:
>Release:        NetBSD 1.4
>Environment:
	NetBSD 1.4, HP 9000/425T, NAT environment w/dynamic IP
System: NetBSD polka 1.4 NetBSD 1.4 (POLKA) #29: Sat May 8 16:55:15 CDT 1999 scottr@polka:/a/src/sys/arch/hp300/compile/POLKA hp300


>Description:
	When dropped from (at least some) ftp servers because the number
	of users has reached an administrative limit, the retry option
	doesn't always retry.

>How-To-Repeat:
	Find a site using NcFTPd that has reached the login limit.  Try to
	fetch a file using a URL on the command line, specifying a retry
	time with the `-r' option.  Notice that it doesn't retry.

	The feature works fine if you simply specify the host to connect
	to rather than a file to retrieve.

>Fix:
	None given.  The problem is probably that the code checks only to
	see if we're connected; we don't track login success or failure
	separately.  This seems simple enough to fix.  However, someone
	more familiar with the code than I am should consider the
	potential problems with any solution.
>Audit-Trail:
>Unformatted: