Messages in this thread | | | Date | Tue, 21 Sep 1999 23:24:46 +0200 | From | Andreas Tobler <> | Subject | Re: strange mounting problem |
| |
Guest section DW wrote: > > From toa@pop.agri.ch Tue Sep 21 20:29:03 1999 > > > You may check, just to be sure, that the kernel indeed sees > > the disk as removable. (It will say so in the boot messages.) > > Ha, you're right, the kernel doesn't see the HP as removable, only the > sony is detected as a removable. (/var/log/dmesg) > > Good. This explains why no check_media_change() was done. > > > I seem to recall that you mentioned that this disk was SCSI-1 > > and that things are fine for SCSI-2 compliant drives. > > Yes. > > > The ioctl BLKRRPART will do revalidate_scsidisk() which does > > the required invalidate_*(). > > This is somewhat Alan suggested to me. So I inserted the > ioctl(fd,BLKFLSBUF) directly into the eject code from Jeff Tranter. > This works so far. > > Yes. I mentioned BLKRRPART and not BLKFLSBUF because the latter does > invalidate_buffers() but not invalidate_inodes(), and you want to do both, > I think. Alan suggested both, first BLKFLSBUF and then BLKRRPART I did this in the eject code and if I do it soft like it ends in an infinite loop. So I only did the BLKFLSBUF. The infinite loop was/is caused due to the fact that the drive pulls in the already ejected dev. If I do it manually it works, means push the 'eject button', but sending the command eject again it spits out the media and then it pulls it in to BLKRRPART. Well I have to play what's best... But anyway I feel much better than before. (A cheaper solution would have been to throw away the old drive, but then I wouldn't have the experiences I have now!!)
(eject code from Jeff!!)
Yours I have to try out tomorrow... Thanks
Andreas
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |