Can it be that those modules are not loaded when mount in fstab occurs?

that's what I was thinking...could I just do a mkinitrd -m (which module) -k 2.6.23.8 and then put the appropriate initrd line into /etc/lilo.conf?

If so, which module (lsmod shows a couple usb modules....I didn't build any into the kernel)?

Those I remember are ehci, ohci, uhci. There are more, you can check at kernel config.

You would also need the scsi disk module and usb mass storage module, however they are called.

lsmod shows this:

Module Size Used by
ohci_hcd 16644 0
xt_tcpudp 2944 5
nf_conntrack_ipv4 12424 1
xt_state 2048 1
nf_conntrack 46364 2 nf_conntrack_ipv4,xt_state
nfnetlink 4248 2 nf_conntrack_ipv4,nf_conntrack
iptable_filter 2304 1
ip_tables 9928 1 iptable_filter
x_tables 9860 3 xt_tcpudp,xt_state,ip_tables
ipv6 196324 12
sg 23196 0
psmouse 32528 0
usb_storage 46596 1
8139too 19328 0
mii 4224 1 8139too
ehci_hcd 25996 0
uhci_hcd 18444 0
shpchp 26132 0
serio_raw 4868 0
evdev 6912 0
i2c_i810 4100 0
i2c_algo_bit 4996 1 i2c_i810
i2c_i801 7440 0
i2c_core 17680 3 i2c_i810,i2c_algo_bit,i2c_i801
intel_agp 20116 1
agpgart 24112 1 intel_agp



I'm thinking I should just build the modules into the kernel that I need for this USB 2.0 drive. Where in the kernel config is the options for building these modules into kernel 2.6.23.8 (I use menuconfig)?

How did you get it to mount? I am having the same problem, not using automount though. Whereas the USB thumbdrive usually gets assigned a /dev, such as /dev/sda1, that is no longer happening for me and I can't figure out why!

@LinuxHippy: Press space


@Vincent Vega: Run dmesg. Don't know how those mount points at dev are assigned (somebody tell me) and never found a better way to check what's happening.

Plug it in after the machine has booted and see if the syslog says anything intelligent about it.

Either way, the last two fields should be "1 2" and not "0 0".

The way I watch all of this happen is with:There is the initial messages that the system recognized something was plugged into the USB and usually a second after that it shows something like:
"sda : sda1"
But that's the part that never happens!

Why use 1 2 instead of 0 0? I always set up the usb thumbdrives with 0 0 and I have never had any problems.

The "1" can be useful if you make backups with a tool that works like the old "dump" tool did, and understands that field from the "fstab" file. It's usually a good idea to set it to "1", for tradition, with mount points that hold data that would typically be copied as part of a backup. In this case it could be right.

The second number, a "2" in this case, is used by "fsck" when invoked with the "-A" option to check every filesystem, in order. "fsck" is invoked this way from "rc.S" but could also be invoked by hand. It's, like the previous field, useful when the mount point refers to a "system" mount point like /, /home, etc. The user would have to decide if it's appropriate in this case. If you end up mounting the filesystem from rc.local because you're not able to appropriately load all the required modules from the initrd (or other problems) and you never run fsck by hand this way, it wouldn't be used. Read "man fstab" and "man fsck" for more details, as well as /etc/rc.d/rc.S.