NetBSD Problem Report #13119

Received: (qmail 15913 invoked from network); 5 Jun 2001 21:54:51 -0000
Message-Id: <am8zj6mvz5.fsf@stevens.e.kth.se>
Date: 05 Jun 2001 23:56:46 +0200
From: Love <lha@stacken.kth.se>
Sender: lha@e.kth.se
To: gnats-bugs@gnats.netbsd.org
Subject: umapfs serverly broken
X-Send-Pr-Version: 3.95

>Number:         13119
>Category:       kern
>Synopsis:       umapfs serverly broken
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Jun 05 21:55:00 +0000 2001
>Closed-Date:    
>Last-Modified:  
>Originator:     Love
>Release:        NetBSD-1.5V (by 26 May)
>Organization:
        Stacken Computer Club 
>Environment:
        NetBSD-1.5V/i386

>Description:

        umapfs keep making my kernel panic in deadlocks

        stacktrace isn't provided due to that savecore doesn't work for me
        (see kern/13076), and the ddb output keep sending

>How-To-Repeat:

        nfs-mount and umap mount your netbsd build tree (I map one uid, not
        a gid).

        cd ${umapS}/arch/i386/stand
        while true ; do make -j12 && make clean ; done

        wait ~2-30 sec and se the kernel reclaim the first vnode in
        getnewvnode() and then die in a deadlock.

>Fix:

        Not a clue. I'm not a friend of ddb and depend on that I have kgdb
        or coredumps to diagnose my bugs. And since I don't have a build
        enviroment for this computer I haven't tried yet.

        I might have a chans to diagnose the bug when I get back to my
        debug computer.

>Release-Note:
>Audit-Trail:
>Unformatted:

Home	PR Database Search