I have noticed a weird instability in my system when I use the RH6.1 tulip driver: Symptom: Disk accesses hang up, with the "Disk Busy" light continously on (but no disk seek noise). System response to already running processes (e.g. X) is sluggish, suggestive of a lot of interrupt activity. Eventually, I'm forced to hit the reset button. Configuration: ASUS P2L97 motherboard, 2x300MHz PII, BusLogic BT-958 SCSI Controller, SMC EtherPower NIC. Also note BT-958 on interrupt 12, while NIC is on interrupt 11. Observations: When I booted with the 2.2.12-20smp kernel, the system would freeze up within an hour or so of light activity. The system would stay up for days with the the 2.2.12-20 kernel, but it could be made to fall over at will with intensive disk activity, n particular running mkisofs. Falling back to 2.2.5-22 didn't help. The system didn't fall over when running fsck in single user mode, however. Work Around: I switched NIC drivers from tulip to de4x5. Now I'm back to using the 2.2.12-20smp kernel, with no difficulites observed.
Assigned to dledford
is this ok in 6.2
I cannot say whether this is still visible in RH6.2. The affected system is indeed running RH6.2, kernel 2.2.16, but no longer has any SCSI harddrives in it. I consider the matter closed, and it's fine with me if you do, too.
Ok I will close this. If anyone else is seeing something with the same hardware and has a problem please reopen this Now if bugzilla only had a resolved->weird button