[CentOS] Simple Install of x86_64 CentOS 5 is failing

Wed May 23 19:09:23 UTC 2007
James Olin Oden <james.oden at gmail.com>

On 5/23/07, Stephen John Smoogen <smooge at gmail.com> wrote:
> On 5/23/07, James Olin Oden <james.oden at gmail.com> wrote:
> > It gets all the way through the install, and then when it reboots,
> > grub can't find the kernel.  Grub can't find the kernel because the
> > kernel was not instaled for some reason?  I looked at the box through
> > rescue mode and in /boot there is no kernel.
> >
> > Any ideas what might have happened?
> >
>
> No idea without some idea about any errors generated during the
> install, what kind of x86_64 hardware (cpu, mb, memory, disk type),
> and what a simple install is, and if there were any errors listed in
> the /root/*log files.
The failure mode is so dramatic that I was wondering if in general if
anyone has seen this before.  As it turn out rpm for all intensive
purposes died, and anaconda itself never caught the errors.    First
you start to see  over and ove in the install.log:

   error: db4 error(22) from dbcursor->c_close: Invalid argument

And then it switches to:

   error: db4 error(-30977) from dbcursor->c_put: DB_RUNRECOVERY:
Fatal error, run database recovery
   error: error(-30977) storing record p^L^^oêJ·^Tm^P¥Å~B¬~Sß0û into Filemd5s

this repeats for a while (with different records of course), but then
degrades/stabalizes too repeated:

  error: db4 error(-30977) from db->sync: DB_RUNRECOVERY: Fatal error,
run database recovery

Essentially, I'm getting errors concerning the integrity of the rpm db
mid rpm transaction during the install.

I don't know if this is repeatable, at this point.  I suspect the type
of hardware is irrelevant (though I know its not completely ruled
out).  I suspect it could be a memory problem, but also recall an
issue with mmap that effect rpm in this way (more to the point Jeff
Johnson muttered something about that as he was looking at it)

Again has anyone seen this?


Thanks...james