fsck_ufs locked in snaplk

Dmitry Morozovsky marck at rinet.ru
Sat Apr 29 16:09:33 UTC 2006


On Mon, 24 Apr 2006, Kris Kennaway wrote:

KK> > I'll try to build DDB kernel tomorrow evening to check. Which commands should I 
KK> > issue in ddb ?
KK> 
KK> 'show lockedvnods', 'ps' and 'alltrace' are important.

Well, common usage pattern does not lead to lock today. I made some 
snapshots on  quota'ed file system and delete them successfully. 

Then, I made one snapshot (finished well), and simultaneously issue 

rm -f /st/.snap/temp_snapshot.1
mksnap_ffs /st /st/.snap/temp_snapshot.2
mksnap_ffs /st /st/.snap/q_snapshot

This leads to

marck at office:~> pidd snap
    0    32     0   0  -4  0     0     8 snaplk DL    ??    0:32.45 [bufdaemon]      root    0.0  0.0 Thu10PM
    0 23279  8868   6  -4  0  1260   708 snaplk D+    p1    0:00.69 mksnap_ffs /st / root    0.0  0.1  7:57PM
    0 23325 23302   0  -4  0  1220   492 ufs    D+    p3    0:00.00 mksnap_ffs /st / root    0.0  0.1  7:59PM
    0 23301 23284   0  -4  0  1228   568 snaplk D+    p4    0:00.00 rm -f /st/.snap/ root    0.0  0.1  7:58PM

DDB output (BTW, it seems BREAK_TO_DEBUGGER together with KDB_UNATTENDED is 
*not* workable in RELENG_6) is available at 
http://woozle.net/FreeBSD/debug/office-ddb.20060429.txt.gz (35k)


Sincerely,
D.Marck                                     [DM5020, MCK-RIPE, DM3-RIPN]
------------------------------------------------------------------------
*** Dmitry Morozovsky --- D.Marck --- Wild Woozle --- marck at rinet.ru ***
------------------------------------------------------------------------


More information about the freebsd-stable mailing list