Bug 29169
| Summary: | FC loop that 'goes away' never times out I/O cmds | ||
|---|---|---|---|
| Product: | Base System | Reporter: | wilko <wilko> |
| Component: | kern | Assignee: | Matt Jacob <mjacob> |
| Status: | Closed FIXED | ||
| Severity: | Affects Only Me | ||
| Priority: | Normal | ||
| Version: | Unspecified | ||
| Hardware: | Any | ||
| OS: | Any | ||
Responsible Changed From-To: freebsd-bugs->mjacob Per previous discussion.. State Changed From-To: open->feedback With bugmeister hat on, reassign from inactive committer. Responsible Changed From-To: mjacob->freebsd-bugs Is this still a problem with modern versions of FreeBSD? Responsible Changed From-To: freebsd-bugs->mjacob mjacob has reactived his commit bit. mea culpa for the bogus reassignment. State Changed From-To: feedback->open Put this back in open state. State Changed From-To: open->closed Fixed as of about 3 weeks ago. |
Powering down FC disk cab results in LIP, noted by the isp driver. This is OK. Running for example camcontrol rescan on the FC loop/bus results in camcontrol hanging in 'cbwait' forever. The I/O request obviously never times out. Fix: Some timeout mechanism needs to be implemented so that a dead FCAL can be detected and attached FC/SCSI devices can be declared 'gone' properly. How-To-Repeat: See description