Hello! After suffering through http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/72451 and experiencing occasional ufs2 panics (rather than the usual system hangs), we recreated the troublesome partition as ufs1 without softupdates, with other newfs parameters being default. We also set the mpsafenet and mpsafevm to 0, as it seemed to have helped increase the system's uptime in the past. Under the load, the system eventually paniced with: 'ffs_blkfree: freeing free block'. The stack was: ffs_blkfree+0x433 ffs_indirtrunc+0x37c ffs_indirtrunc+0x33c ffs_truncate+0xd24 ufs_inactive+0x1d7 vput+0x113 kern_unlink+0x129 unlink+0x3e The machine runs fresh amd64-current on a single Opteron in a dual capable Tyan K8W motherboard. The troublesome harddrive is connected directly to the on-board SATA connector at SATA150 speed. We went through two drives, two SATA cables, already, and tried other on-board SATA connectors. Is this panic just as likely to hit us if we go back to RELENG_5? Could the bug be stemming from some unsafeness in ata's reset/reinit code -- testing these must be painfully difficult, I understand. Thanks! -miReceived on Thu Oct 21 2004 - 16:53:50 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:19 UTC