We use Warewulf and create chroots that are ultimately PXE booted to compute nodes. Recently, we've been attempting to pin the YUM config in the images to specific CentOS point releases. Our systems are isolated from the internet, and there are driver compatibility questions so we at least need to manage the upgrades to different point releases on our own timetable.
That said, we've suddenly began to have trouble reliably being working from the 7.5 area in the vault.centos.org. We have our CentOS-Base.repo configured thusly:
Code: Select all
[base]
name=CentOS-7.5 - Base
baseurl=http://vault.centos.org/centos/7.5.1804/os/$basearch/
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
[updates]
name=CentOS-7.5 - Updates
baseurl=http://vault.centos.org/centos/7.5.1804/updates/$basearch/
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
[extras]
name=CentOS-7.5 - Extras
baseurl=http://vault.centos.org/centos/7.5.1804/extras/$basearch/
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
[centosplus]
name=CentOS-7.5 - Plus
baseurl=http://vault.centos.org/centos/7.5.1804/centosplus/$basearch/
gpgcheck=1
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
Can someone help me troubleshoot? I was already pretty sure that --installroot used the installroot PATH for /etc/yum.repos.d and confirmed this with strace since there doesn't appear to be any significant documentation on how this works.
Thanks in advance.
PS: no lectures please on holding to point releases, please. It's occasionally necessary due to factors outside of our control, and we move as soon as is practical.