> http://www.opensolaris.org/jive/thread.jspa?messageID=36229#36229
The problem is back, on a different system: a laptop running on-20060605 bits.
Compared to snv_29, the error message has improved, though:
# zfs snapshot hdd/[EMAIL PROTECTED]
cannot snapshot 'hdd/[EMAIL PROTECTED]': dataset is busy
# zfs snapshot hdd/[EMAIL PROTECTED]
cannot snapshot 'hdd/[EMAIL PROTECTED]': dataset is busy
Both the hdd/usr-base and hdd/export zfs are using a legacy mountpoint and are
not mounted.
zdb reports this:
# zdb -ivvvv hdd/export
Dataset hdd/export [ZPL], ID 313, cr_txg 378637, last_txg 598821, 1.71G, 147592
objects
ZIL header: claim_txg 598821, seq 0
first block: [L0 ZIL intent log] 2000L/2000P DVA[0]=<0:51506000:2000>
zilog uncompressed LE contiguous birth=519227 fill=0
cksum=ed322d1e7ca741fc:194d26d6f2582165:139:1
Block seqno 1, already claimed, [L0 ZIL intent log] 2000L/2000P
DVA[0]=<0:51506000:2000> zilog uncompressed LE contiguous birth=519227 fill=0
cksum=ed322d1e7ca741fc:194d26d6f2582165:139:1
Note the "claim_txg" value != 0.
For the hdd/usr-base zfs the output was similar, claim_txg != 0. By manually
mounting and unmounting that filesystem the claim_txg was reset to 0 and
the zfs snapshot command didn't fail with "dataset is busy" any more.
This message posted from opensolaris.org
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
