Comment # 8 on bug 1158241 from
OK, makes sense, but I didn't have such problems for a long time on the
machine. As a matter of fact the system disk (Seagate Barracuda) may be failing
soon (it has some defective sectors), so maybe an I/O delay was the reason for
delaying the journal write. Unfortunately killing processes (and retrying the
I/O) is not the best recovery strategy in this case...


You are receiving this mail because: