https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=288345
--- Comment #34 from Craig Leres <[email protected]> --- (In reply to Craig Leres from comment #32) I saw another case of D wait processes for ~10 minutes, twice today. The timing was similar to last time: the build started at midnight, the first D wait happened 48 minutes in, resolved after 10 minutes. The second D wait was at 1:15 and lasted for 9 minutes. I don't see anything interesting in the logs. I strongly suspect if I had been running the non DEBUG kernel these would have been system hangs. Looking at DEADLKRES I don't see any printf's that would indicate that a deadlock was resolved (only panics). Is it possible DEADLKRES is silently preventing the problematic D wait from happening? -- You are receiving this mail because: You are the assignee for the bug.
