[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: still -- building kernel rpm problem



Chuck Campbell wrote:
I didn't hear any response, so I re-downloaded the kernel-2.4.20-24.9.src.rpm
and tried again. I still see the same thing during build. (See prior note below).

If the kernel built properly (the make didn't error out), you're probably fine. I don't use "rpmbuild", I do the normal configuration, build cycle:

make xconfig;make dep clean bzImage modules install modules_install

I don't recall the checksum errors popping up, but the kernel does build
based on the config files included in /usr/src/linux-2.4/configs.

I'm loathe to reboot w/ the new kernel to try it, because the machine is
50 miles away from me, and I can't get there easily to access the console
to force a boot of the prior kernel.
> Is there a way around this issue for remote machines?

You need one of three things: a remote KVM switch, someone there to
watch it, or set up serial port console operations with either a modem
on the machine's serial port or its serial port connected to another
machine you CAN get at via a null-modem cable.

You can redirect the console output to a serial port and talk to the
machine via a modem or another machine connected to its serial port.
The grub commands are:

	serial --unit=0 --speed-9600
	terminal serial

(see the grub manual for details). For LILO, the commands are:

serial=0,9600n8

(see the lilo manual).

Chuck previously wrote:


I downloaded the recently released kernel-2.4.20-24.9.src.rpm and did an
install as root. rpm -Uvh kernel-2.4.20-24.9.src.rpm

Next I went to the /usr/src/redhat/SPECS directory and did this:

rpmbuild -ba --clean --target i686 kernel-2.4.spec

during this build I have seen amny occurrences of the following message:

1 of 13 checksums did not match

What implications does this have for this build, and what should I do about it?


I wanted to build this specifically for the i686 platform, instead of installing i386 binaries. This is (ostensibly) for performance reasons. Is my
reasoning all wet?


thanks,
-chuck



--
----------------------------------------------------------------------
- Rick Stevens, Senior Systems Engineer     rstevens vitalstream com -
- VitalStream, Inc.                       http://www.vitalstream.com -
-                                                                    -
-    Overweight:  When you step on your dog's tail...and it dies.    -
----------------------------------------------------------------------




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]