Kickstart fails on boot

General
1

Hello Team,
I am automating rocky linux installation.
Rocky9.1 image worked fine.
but rocky8.7 kickstart boot fails with the error
Started cancel waiting for multipath siblings of sda
ive tried this ks script for disk mangment and didnt work this is the exact script that i used for rocky9 that worked fine

ignoredisk --only-use=sda

Partition clearing information

clearpart --all --initlabel

Disk partitioning information

part /boot/efi --fstype=“efi” --ondisk=sda --size=600 --fsoptions=“umask=0077,shortname=winnt”
part / --fstype=“xfs” --ondisk=sda --size=5000 --grow
part /boot --fstype=“xfs” --ondisk=sda --size=1024

and tried this script that use zerombr insted and also same issue
zerombr

Partition clearing information

clearpart --all --initlabel

Disk partitioning information

#autopart --type=lvm
part /boot --fstype=“xfs” --ondisk=vda --size=1024 --label=boot --asprimary
part pv.01 --fstype=“lvmpv” --ondisk=vda --size=15359
volgroup vg_os pv.01
logvol /tmp --fstype=“xfs” --size=1024 --label=“lv_tmp” --name=lv_tmp --vgname=vg_os
logvol / --fstype=“xfs” --size=14331 --label=“lv_root” --name=lv_root --vgname=vg_os

2

I’ve seen that message mentioned on forum, so you are not the first and issue might not be in kickstart.

What happens if you start interactive 8.7 installer on that machine?

3

normal instillation works,
i saw that the problem was present in all rhel 8.5 but it was fixed in rhel 8.7 but rocky 8.7 still not working
Thanks

4

Also these are the changes that i made to isolinux.cfg and to the grup.cfg

isolinux.cfg
label kickstart
menu label ^Kickstart
kernel vmlinuz
append initrd=initrd.img inst.stage2=hd:LABEL=Rocky-8-7-x86_64-dvd inst.ks=hd:LABEL=Rocky-8-7-x86_64-dvd:/anaconda-ks.cfg

grup.cfg
menuentry ‘Kickstart’ --class fedora --class gnu-linux --class gnu --class os {
linuxefi /images/pxeboot/vmlinuz inst.stage2=hd:LABEL=Rocky-8-7-x86_64-dvd inst.ks=hd:LABEL=Rocky-8-7-x86_64-dvd:/anaconda-ks.cfg quiet
initrdefi /images/pxeboot/initrd.img
}

5

after two days, the issue was the name of the iso image :smiley:

6

I have the same problem with an 8.6: A normal installation works but one with a kickstart file gives me the error: “started cancel waiting for multipath siblings of sda”.
Please, how did you solve it?

7

Hi dominTxomin,

I suspect this will be down the label or as mhk states the name of the ISO. I would take a look at the LABEL in isolinux.cfg and grup.cfg.

After that if you are still having issues, how are you generating the ISO?

Thanks Tom.

8

Hello, thanks for the answers.
I have tried with version 8.7 and the result is the same on a physical server(On a virtual server it works fine):
“started cancel waiting for multipath siblings of sda”.

I generate the iso like this:

mkisofs -o test.iso -b isolinux/isolinux.bin -J -R -l -v -T -c isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table -eltorito-alt-boot -e images/efiboot.img -no-emul-boot -graft-points -V 'Rocky-8-7-x86_64' .

The grub.cgf file:

menuentry 'ISO install' --class fedora --class gnu-linux --class gnu --class os {
        linuxefi /images/pxeboot/vmlinuz inst.stage2=hd:LABEL=Rocky-8-7-x86_64 inst.ks=cdrom:/isolinux/my-ks.cfg
        initrdefi /images/pxeboot/initrd.img
}

I check the iso file:

isoinfo -d -i ./test.iso
CD-ROM is in ISO 9660 format
System id: LINUX
Volume id: Rocky-8-7-x86_64
Volume set id:
Publisher id:
Data preparer id:
Application id: GENISOIMAGE ISO 9660/HFS FILESYSTEM CREATOR (C) 1993 E.YOUNGDALE (C) 1997-2006 J.PEARSON/J.SCHILLING (C) 2006-2007 CDRKIT TEAM
Copyright File id:
Abstract File id:
Bibliographic File id:
Volume set size is: 1
Volume set sequence number is: 1
Logical block size is: 2048
Volume size is: 1127041
El Torito VD version 1 found, boot catalog is in sector 210
Joliet with UCS level 3 found
Rock Ridge signatures version 1 found
Eltorito validation header:
    Hid 1
    Arch 0 (x86)
    ID ''
    Key 55 AA
    Eltorito defaultboot header:
        Bootid 88 (bootable)
        Boot media 0 (No Emulation Boot)
        Load segment 0
        Sys type 0
        Nsect 4
        Bootoff D3 211
9

Hi,

In grub.cfg try replacing:

inst.ks=cdrom:/isolinux/my-ks.cfg

With:

inst.ks=hd:LABEL=Rocky-8-7-x86_64 :/isolinux/my-ks.cfg

Thanks Tom.

10

Do u still have the problem ?

11

In the future, if you are trying to add a kickstart to an ISO, I woudl recommend using mkksiso. It will make things easier over trying to arbitrarily run mkisofs.

12

Thank you very much for everything. With the changes you mentioned it worked also on the physical server. I copied the ISO to a CDROM and it booted without any problem. Thank you.
Now I will make the changes to be able to use a usb stick instead of the cdrom but that’s another matter.

13

Good afternoon. I am trying to install from a usb stick:
In the kickstart file I have two sections %pre and two sections %post.
In the two %pre sections and in the first %post section, it correctly recognizes the usb stick:

/dev/sdb1 on /run/install/repo type vfat (ro,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,errors=remount-ro)

The problem, and I don’t know why it happens, is that in the second section %post, it doesn’t recognise it and what’s more, it does something with it that makes it inaccessible, it doesn’t even recognise it on my laptop.
I have no idea what can happen
Any idea?
Thanks

14

i understand that u have issue in the post section is unrecognized mean the instillation don’t do that the commands that u listed in the post section ??
can u share ur kick start file so we can help u
i think u need to specify the weather should ur code run in non chroot environment and what compiler u should use sh, shell, even u can use python3

15

Hello, I can’t test again until Monday.
I have made the following changes:
In the %post I have removed “interpreter” and in the next line I have put #!/bin/bash.
In the first %post, which works correctly, at the end what I do is to unmount the usb device.
In the second %post, I mount the usb device again.

On monday when I try it, I’ll let you know how it works and if so, I’ll put the kickstart file.

Thanks

16

Good afternoon. I have finally been able to continue with the tests.
The break was a good thing because when I checked again the kickstart file I saw the possible failure of the usb device deletion. It was me who caused it. I thought I was only erasing the sda and I erased all the sd*. Tomorrow I hope to try again and I’ll let you know but it’s almost ready.
Thanks

17

Good morning. Well, there is no way, I no longer delete the content of the usb device but I still cannot access it in the second %post.

First %post

%post --nochroot
#!/bin/bash
echo “POST 1”
if [ ! -d /mnt/sysimage/root ]; then
mkdir /mnt/sysimage/root
fi

mount
→ output: /dev/sdb1 on /run/install/repo type vfat (ro,…)

lsblk -d -n -p -o NAME,TYPE | grep -E “/dev/sr|usb|/dev/sd|cdrom”
–>output:
–>/dev/sda disk
–>/dev/sdb disk

blkid -s UUID
–>output: /dev/sdb1: UUID=“4E6C-CD85”

cp /tmp/type /mnt/sysimage/root/
echo “End POST 1”
%end

Second %post:
%post --logfile=/root/test.log
#!/bin/bash
echo date
unalias mv &>/dev/null
unalias cp &>/dev/null

ls -l /etc/sysconfig/network-scripts/ifcfg-*

echo “… mount (0) …”
mount
–>output: NOT mounted (At another time I put it to mount and it told me it was busy)

echo “… df -h (0) …”
df -h
–>output: NOT shown

lsblk -d -n -p -o NAME,TYPE | grep -E “/dev/sr|usb|/dev/sd|cdrom”
–>output:
–>/dev/sda disk
–>/dev/sdb disk <----

blkid -s UUID
→ output: /dev/sdb1: UUID=“4E6C-CD85” <----

What do I still need to do, do I need to do/activate something other than chroot like the first %post to make it work?
Thanks again