[Solved] Kernel downgrade? Newbie question

MTUser2007 · Last edited by MTUser2007 on Tue Sep 15, 2009 2:01 am; edited 2 times in total

I upgraded from linux-2.6.27-gentoo-r8 to linux-2.6.30-r6. I though everything was working fine, but then I discovered a third party tool I use, Acronis True Image Echo Enterprise Server was broken. I tried uninstalling the tool and re-installng it. It will not compile against this new kernel. I would like to go back to 2.6.27-gentoo-r8, but when I try to eselect this source and compile it with genkernel, I get errors related to drivers. Here is a small fragment of the error log in case it helps:

CC [M] drivers/scsi/lpfc/lpfc_mem.o
CC [M] drivers/usb/storage/shuttle_usbat.o
CC [M] drivers/scsi/lpfc/lpfc_sli.o
CC [M] drivers/usb/storage/sddr09.o
drivers/scsi/lpfc/lpfc_sli.c: In function 'lpfc_sli_poll_fcp_ring':
drivers/scsi/lpfc/lpfc_sli.c:1365: warning: array subscript is above array bounds
drivers/scsi/lpfc/lpfc_sli.c:1365: warning: array subscript is above array bounds
drivers/scsi/lpfc/lpfc_sli.c: In function 'lpfc_sli_handle_fast_ring_event':
drivers/scsi/lpfc/lpfc_sli.c:1541: warning: array subscript is above array bounds
drivers/scsi/lpfc/lpfc_sli.c:1541: warning: array subscript is above array bounds
--
CC [M] drivers/video/display/display-sysfs.o
CC [M] drivers/scsi/lpfc/lpfc_scsi.o
LD [M] drivers/video/display/display.o
CC [M] drivers/watchdog/pcwd.o
drivers/scsi/lpfc/lpfc_scsi.c: In function 'lpfc_bus_reset_handler':
drivers/scsi/lpfc/lpfc_scsi.c:1245: warning: 'status' may be used uninitialized in this function
--
LD [M] drivers/scsi/sym53c8xx_2/sym53c8xx.o
LD [M] drivers/scsi/scsi_tgt.o
LD [M] drivers/scsi/qla2xxx/qla2xxx.o
Building modules, stage 2.
MODPOST 1588 modules
WARNING: modpost: Found 5 section mismatch(es).
--
* >> Appending busybox cpio data...
* >> Appending modules cpio data...
* Gentoo Linux Genkernel; Version 3.4.10.904
* Running with options: --menuconfig all

* ERROR: Could not copy the initramfs image to /boot!

* -- End log... --

* Please consult /var/log/genkernel.log for more information and any
* errors that were reported above.

* Report any genkernel bugs to bugs.gentoo.org and
* assign your bug to genkernel@gentoo.org. Please include
* as much information as you can in your bug report; attaching
* /var/log/genkernel.log so that your issue can be dealt with effectively.
*
* Please do *not* report compilation failures as genkernel bugs!
*
gentoo-mirror1 ~ #

What do I need to do to go back? If it involves manually compiling the kernel and not using genkernel, please post explicit instructions as I have never manually compiled a kernel before, all I have used is genkernel. Thank you.

Hu · Moderator Joined: 06 Mar 2007 Posts: 21619

We will need to wait another ~2 years to see the 2.6.40 kernel to be sure, but assuming nothing major changed with it, it did not do anything difficult to revert. I see no compilation errors in the output you posted. The messages you cite are warnings, not errors. However, assuming you did not clean out /boot, your old 2.6.27 kernel is already there. Just switch grub back to it and go.

For reference, what is wrong with Acronis?

MTUser2007 · Posted: Mon Sep 14, 2009 3:42 am Post subject:

I am Sorry, typo on my part, 2.6.30-r6. (I've edited the original post to fix it now) Thank you for the reply. I'll give your suggestion a try. Acronis did provide a log of the failure. I'll dig it out and post it here.

MTUser2007 · Posted: Mon Sep 14, 2009 4:25 pm Post subject:

I tried to go back after receiving the warnings above. I must have damaged the image in boot as the system complains about part of it missing and goes into kernel panic after failing to mount the partitions. Can anyone advise me on how to manually get the boot image of 2.6.28 gentoo-r8 back into boot?

I have an Acronis Image of the version with 2.6.30-gentoo-r6 so I can try again to recompile the lower version kernel. This machine is an IBM xServe 226 running raid 5 so perhaps that is why boot directory image was damaged?

John R. Graham · Posted: Mon Sep 14, 2009 8:33 pm Post subject:

MTUser2007 · Posted: Mon Sep 14, 2009 9:37 pm Post subject:

Thank you for the help!

At the risk of opening a can of worms, and wearing out my welcome here at the forum by asking too many newbie questions, how do I go about checking the partition for space and contents?

As an alternative to deleting something, can I easily make the partition larger? This machine has 4 partitons totaling 72GB in RAID5. The biggest data partition is 68GB. If it is technically easy to accomplish, I could make the data partition smaller and the boot partition bigger. ( I know I could do this with Acronis Disk Director for Server on a Windows based system, I am betting it is possible on Linux based machine too)

John R. Graham · Posted: Mon Sep 14, 2009 11:19 pm Post subject:

No worries. If /boot is not mounted, mount it. Then, to show used and available space, use df:

MTUser2007 · Posted: Tue Sep 15, 2009 1:41 am Post subject:

I am so happy, thank you!

gentoo ~ # df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda4 65G 14G 48G 23% /
udev 10M 160K 9.9M 2% /dev
/dev/sda1 39M 38M 0 100% /boot
/dev/sda3 1.9G 431M 1.4G 25% /webroot
none 1013M 0 1013M 0% /dev/shm
//buffalo/designer 924G 923G 867M 100% /mnt/jdrive
//buffalo/backup 924G 923G 867M 100% /mnt/avast
gentoo ~ #

The minuscule boot partition was full. I had 3 boot images in there. I removed the two oldest and recompiled with genkernel, and viola, it worked. I am back in business with Acronis True Enterprise Echo Image Server installed. Another small step in my continuing journed to Linux/Gentoo competency, hoorah!

If you or anyone else wanted too, I'd sure like to learn why boot partitions are set so small.

John R. Graham · Posted: Tue Sep 15, 2009 2:32 am Post subject:

I make mine a little bigger for exactly that reason: