Hi all,
I have a problem where 50% of the time when a node is rebooted, it
gets stuck with one of the following messages being the last on the
console. I haven't had this with non-testbed nodes of the same spec
also on RH 7.3. Anyone any ideas ?
Jan 29 16:15:55 epcf36 kernel: VFS: Disk quotas vdquot_6.5.1
Jan 29 16:15:55 epcf36 kernel: pty: 2048 Unix98 ptys configured
Jan 29 16:15:55 epcf36 kernel: Serial driver version 5.05c (2001-07-08)
with MANY_PORTS MULTIPORT SHARE_IRQ SERIAL_PCI ISAPNP enabled
Jan 29 16:15:55 epcf36 kernel: ttyS0 at 0x03f8 (irq = 4) is a 16550A
Jan 29 16:15:55 epcf36 kernel: ttyS1 at 0x02f8 (irq = 3) is a 16550A
Jan 29 16:15:55 epcf36 kernel: Real Time Clock Driver v1.10e
Jan 29 16:15:55 epcf36 kernel: FDC 0 is a post-1991 82077
Going back to some previous problems, now fixed:
> You can of course just chown the /tftpboot/pxelinux.cfg to apache.
> Of course this should have happened anyway, obviously something broken.
Yep, what had happened is that my minimal RH 7.3 system on the LCFG
had omitted apache, which I added later, so at the time the chown
happened the 'apache' userid didn't exist. When I re-did things from
scratch in the right order ;-) that problem didn't happen.
I have an issue with my BIOS doing a hard-disk boot and not a network
boot even if network boot is listed first, unless I suppress the
hard-disk boot with a !, but that's another story.
>
> So the fstab error was fixed just by making the WP8/Atlas, or whatever
> it was directory?
Yes: but in 2_1_9 (or for whatever reason) that directory WP8/Alice does
exist, so I didn't need to make it by hand.
In 2_1_9 there's a mismatch between the edg-pxelinux-supp version listed
in /opt/local/linux/7.3/rpmcfg/ngserver73-rpm, and the actual version at
/opt/local/linux/7.3/RPMS/LCFGng/edg-pxelinux-supp-1.0.16-edg1.i386.rpm
but that was easily fixed.
Lawrie.
|