Unable to use autogenerated kickstart using Intel Matrix Storage Manager RAID1 on 2 disks
Issue
-
Sometimes, it is not possible to kickstart a device when using the RAID1 volume. When installing interactively, it works.
-
It seems that
ignorediskandclearpartcommand do not deal correctly with theVolume0_0. These are important configuration commands to avoid installing the RHEL system on an unwanted device. -
How can we safely kickstart a node with an IMSM RAID volume ?
-
Doing an interactive installation generates a kickstart
- Using this kickstart to perform a new installation doesn't work, it displays the following message:
the following problem occurred on line 11 of the kickstart file:
Specified nonexistent disk vol0_0 on ignoredisk command.
- The
ignorediskerror message is raised inpyanaconda.kickstart.IgnoreDisk().
# python
Python 2.7.5 (default, Feb 11 2014, 07:46:25)
[GCC 4.8.2 20140120 (Red Hat 4.8.2-13)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import pyanaconda.kickstart
>>> ks = pyanaconda.kickstart
>>> ks.deviceMatches('sda')
['sda']
>>> ks.deviceMatches('Volume0_0')
[]
>>> ks.deviceMatches('md/Volume0_0')
['md126']
>>>
Obviously, Volume0_0 is not a valid device name, whereas it is the name found in /root/anaconda-ks.cfg after a successful installation.
-
It has been possible to continue the installation by specifying
md/Volume0_0instead ofVolume0_0in the whole kickstart file. -
However, the
clearpartdoes not work neither, and anaconda is claiming that there is not enough space in filesystems. One must go to interactive mode to be able to select the BIOS RAID set and use all space for installation.
Environment
- Red Hat Entreprise Linux (RHEL) 7.1
- Anaconda
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.