Re: kern/51377: fss(4) panic if snapshot mounted read/write

[Paul Goyette <paul%whooppee.com@localhost>]

The following reply was made to PR kern/51377; it has been noted by GNATS.

From: Paul Goyette <paul%whooppee.com@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/51377: fss(4) panic if snapshot mounted read/write
Date: Sun, 31 Jul 2016 12:48:48 +0800 (PHT)

 The traceback here is somewhat misleading.  The frame which references 
 do_sys_waitid()+0x960 is really in proc_exit() at line 1193
 
 (gdb) list *do_sys_waitid+0x960
 0xffffffff80819231 is in do_sys_waitid 
 (/build/localcount/src/sys/kern/kern_exit.c:1193).
 1188             * Release reference to text vnode
 1189             */
 1190            if (p->p_textvp)
 1191                    vrele(p->p_textvp);
 1192
 1193            mutex_destroy(&p->p_auxlock);
 1194            mutex_obj_free(p->p_lock);
 1195            mutex_destroy(&p->p_stmutex);
 1196            cv_destroy(&p->p_waitcv);
 1197            cv_destroy(&p->p_lwpcv);
 
 
 It would appear that mount_ffs(8) detected an inconsistency with the 
 snapshot (file-system dirty), and unmounted it (which generated the 
 "snapshot invalid: forced unmount" message).  then, when mount_ffs tried 
 to exit, it ran into a problem with its iamge/text file, resulting in 
 the panic.
 
 It is unclear to me how the fss unmount hook could have caused the 
 corruption of the process's text vnode.
 
 
 
 
 
 +------------------+--------------------------+------------------------+
 | Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:      |
 | (Retired)        | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com   |
 | Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd.org |
 +------------------+--------------------------+------------------------+