-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, It seems that with a lot of file on ZFS (used as / but I don't think this would make change) w/ compression configuration, the nightly locate(8) scan would finally cause the box to enter a livelock/deadlock where locate would stuck in 'zfs:lo' state, and new process accessing the file system would stuck in 'zfs:&buf_hash_table.ht_locks[i].ht_lock]'. Is this a known issue (any workaround)? This is a production box without serial port, and is several miles away from me, but I will try to provide some debugging aid if necessary. Cheers, - -- Xin LI <delphij@delphij.net> http://www.delphij.net/ FreeBSD - The Power to Serve! -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFHZrswhcUczkLqiksRAlbQAJ96llqty5JzasitfaEvFXeRzKYuGACg2L7H xJcPPYmu1AwgMAT56JdwSPg=SEG+ -----END PGP SIGNATURE-----