Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 10 Oct 1996 18:47:04 -0500 (CDT)
From:      Karl Denninger  <karl@Mcs.Net>
To:        fenner@parc.xerox.com (Bill Fenner)
Cc:        fenner@parc.xerox.com, karl@Mcs.Net, current@freebsd.org, hackers@freebsd.org, pst@jnx.com
Subject:   Crash in -current (and fix) - plus NEW issue!
Message-ID:  <199610102347.SAA16950@Jupiter.Mcs.Net>
In-Reply-To: <96Oct10.150729pdt.177476@crevenia.parc.xerox.com> from "Bill Fenner" at Oct 10, 96 03:07:17 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> 
> Of course, tcp_attach() is static so can't be called from tcp_input().
> Either we need to call (*so2->so_proto->pr_usrreqs->pru_attach)(so2),
> or we need to sofree(so2) and then do another so2=sonewconn(so, 0).
> (or make tcp_attach() not static).
> 
>   Bill
> 

>From what I can see of the code the following fragment looked safe:

sofree(so2);
so2=sonewconn(so, 0);

So I inserted that in the appropriate place...  We'll see what happens; I'm
running that test kernel now on the machine which was blowing up.

Since the time to crash is variable, I probably won't know for a day or so
if it fixes the problem -- if it does, I'll send in a "send-pr" for the
commit with a code diff.

On to other things:

	Has anyone seen THIS oddity?

Environment:
	2.2-current NFS client
	BSDI NFS server

SOMETIMES, a "pwd" to a mounted directory will fail.  If it does, OTHER
directories on the same disk (mount point) also may fail -- but NOT all of 
them.  Order is important; ie: if one user fails, another always does, but
if the second does, the first may or may not (!)

This shows up most frequently either in the shell (complaints at login time,
although the directory contents are accessible) or in FTP ("pwd" command 
returns a failure).

It comes and goes without obvious reason or intervention, and when its 
"gone" everything works as expected.

Anyone else seen this?  This one is really, really strange!  

The disks are mounted "rw,bg,intr,nodev,nosuid".

Again, this is new in recent -CURRENTs (since about 8/20), and does NOT 
appear to happen for directories permitted as 755 (which should be readable 
by anyone anyway).

If we can fix *this* one, and the crash fix works, then I think we have
a stable build here which we could roll out.

--
--
Karl Denninger (karl@MCS.Net)| MCSNet - The Finest Internet Connectivity
http://www.mcs.net/~karl     | T1 from $600 monthly; speeds to DS-3 available
			     | 23 Chicagoland Prefixes, 13 ISDN, much more
Voice: [+1 312 803-MCS1 x219]| Email to "info@mcs.net" WWW: http://www.mcs.net/
Fax:   [+1 312 248-9865]     | Home of Chicago's only FULL Clarinet feed!



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610102347.SAA16950>