|Summary:||Installer seems to fail to set any partitions bootable|
|Product:||[Retired] Red Hat Linux||Reporter:||Preston Brown <pbrown>|
|Component:||installer||Assignee:||Matt Wilson <msw>|
|Status:||CLOSED CURRENTRELEASE||QA Contact:|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||1999-03-13 21:58:50 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
|Cloudforms Team:||---||Target Upstream Version:|
Description Preston Brown 1998-11-12 11:42:33 UTC
After a server install, the installer seems to fail to set any partitions bootable. Some computers need to have a partition marked bootable to boot. :) This failure necessitates booting from floppy, entering fdisk, and setting the first partition bootable before the system can be fully usable.
Comment 1 David Lawrence 1998-11-19 13:51:59 UTC
I was able to install to a machine that currently had NT 4.0 installed on the entire hard drive. I ran the server installation and it correctly created the partitions and was able to boot up normally. Please let use know what the prior configuration was on your system was before you attemmpted the server installation.
Comment 2 Edward Schlunder 1998-12-05 00:44:59 UTC
No, the problem is with machihnes that have absolutely nothing on them and you want to install Linux all by itself on the machine. In this case, if you use Disk Druid to create your Linux partitions, the first partition is never marked "active" and thus LILO you can't boot after the install. If you are installing Red Hat 5.2 on a machine with some other OS already on it, chances are you will already have a partition marked active and thus you'll never see this problem.
Comment 3 Preston Brown 1998-12-05 01:46:59 UTC
matt/dave: I believe this is still a very valid bug. Please take the bootable flag off all partitions on a drive, and then wipe the partitions, and see if an install works correctly.
Comment 4 David Lawrence 1998-12-07 21:30:59 UTC
I have not been able to replicate this problem in the test lab. I low-leveled the first few tracks of the hard disk as to emulate a brand new hard disk with no information on it and the server install still worked properly.
Comment 5 dank 1999-03-10 15:44:59 UTC
This has become a very public bug; see Jerry Pournelle's column in this month's resurrected Byte, http://www.byte.com/columns/chaosmanor/1999/030199a.html At the first installfest I attended, one user was stymied by this bug for about an hour, until we realized the installer had failed to mark anything bootable. In neither of the above cases was it a server install. I think it can happen regardless of workstation, custom, or server install. If you can't reproduce it, email me, and I'll try to come up with a recipe.
Comment 6 Preston Brown 1999-03-10 16:46:59 UTC
this bug is getting fixed in our next distribution.
Comment 7 Matt Wilson 1999-03-13 21:58:59 UTC
Fixed in the next release.