Red Hat Bugzilla – Bug 48773
Not enough memory to install Red Hat on this machine during install
Last modified: 2007-04-18 12:34:40 EDT
From Bugzilla Helper:
User-Agent: Mozilla/4.7 [en] (Win95; U)
Description of problem:
Redhat 7.1 Anaconda does not load the install program on a Compaq Proliant 1500 with 128MB and an integrated
wide fast scsi-2/p controller. I can install Red Hat 5.2 on this computer from the CD, but the 7.1 install does not work.
Steps to Reproduce:
2.Loads image files
3.Then Anacondo tried to start the install and an errorr occurs - not enough memory to install on Red Hat on this machine.
Actual Results: The Install process aborts and the machine reboots.
Expected Results: The install process to load and continue.
I opened a support call with Red Hat and they said there are known problems with 7.1 and scsi controllers.
Boot the machine with:
"linux mem=exactmap mem=128M@0"
"linux mem=exactmap mem=640k@0 mem=128M@1M"
The install process went further with the "linux mem=exactmap mem=128M0" however I got the error message
python: header.c:511 headerLoad: Assertion 'rdlen == dl' failed install exited abnormally -- recieved signal 6 sending termination
signals...done sending kill signals...done disabling swap... /tmp/swap/sda7 unmounting filesystems...<F1> for help|/mnt/sysimage/bootments
| <<<Space> select | <F12> next screen /mnt/sysiamge/home /mnt/sysimage/rpoc /mnt/sysimage/var /mnt/sysimage/use /mnt/sysimage
/mnt/runtime /mnt/source umount failed () /dev/pts /proc you may safely reboot your system.
CTL-DEL-F4 screen last line displays
<6>Adding Swap: 265032k swap-space (priority -)
CTL-DEL-F3 screen last linees display
* Detected 128M of memory
* Swap attempt of 128M to 256M
* Maximum cylinder is 7
It appears the swap space was not created properly on the drives.
Removed two of the three Compaq Hot pluggable 4.2 GB drives to see if maybe the drive array was causing a problem.
When I booted with one hard drive it got as far as copying the image to the hard drive and then the kernal paniced.
pmd entry c7fc84a8: 00000000000000000000000
...pmd not present!
eax: 4a9b3820 ebx: c00f2020 ecx: c0073000 ed c004a010
esi: c7ffdcdc edi: 00000202 ebp: 00000968 esp: c7ff1e54
ds: 0010 es: 0018 ss:0018
Process linuxrc (pid:8, stackpage=c7ff1000)
Stack: c0073a20 c0073a20 c0073a20 c0073a20 00000000 c01327a8 c7ffdcdc c0073a20
00000003 0000000001 00000000 00000000 00000017 c0073a20 00001000 c0073a20
00000000 c01310ba c10017f8 000000000 0000008f2 c10017f8 c100183c 00000000
Call Trace: [<c-1327a8>] ..... more numbers
Code: 89 08 eb 28 90 8b 46 08 8b 58 04 8b 51 04 8b 01 89 50 04 89
EXT2-fs: unable to read superblock
isofs_read_super: bread failed, dev=09:00, iso_blknum=16, block 32
FAT bread failed
FAT bread failed
Kernal Panic: VFS: Unable to mount root fs on 09:00
Looks like it is unable to mount the root. SCSI controller problem? The scsi controlle works fine with Red Hat 5.2.
that is with
"linux mem=exactmap mem=640k@0 mem=127M@1M"
I booted with linux mem=exactmap mem=640k@0 mem=128M@1M
Thanks that worked!
Ehm I made a typo, it needs to be "127" not "128", eg 1 less than you have,
so it is 640Kb in "low memory" and an additional 127Mb in "high" memory.
Also, if you install the updated 2.4.3-12 kernel, you can use the old style
This workaround is confirmed to work with at least one other machine, this time
a ProLiant 1500 with a SmartArray 221 (4 drives, RAID5).
*** This bug has been marked as a duplicate of 36940 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.