CentOS 8 AMI availability in AWS

[akrueger]

Is there a date on when an official CentOS 8 AMI will be available in AWS please? We're keen to upgrade, but I'm hesitant to use a third-party image due to security reasons.
I've looked at https://wiki.centos.org/Cloud/AWS and the AWS marketplace and cannot find anything.

[TimL]

I have been wondering the same thing.
Maybe they have bugs they want to fix first...
Where is the best place to prod the developers with this question?

[TrevorH]

I am fairly sure there is a bugs.centos.org bug ticket for this already.

[speculatrix]

I am fairly sure there is a bugs.centos.org bug ticket for this already.

https://bugs.centos.org/view.php?id=16614

[aks]

You can upload your own vm image to AWS - last did this a few years ago & I think I used a qemu image, possibly converted via virt-image (for something like that). See https://aws.amazon.com/ec2/vm-import/
Also I think the Cloud SIG provides qcow images already.

[aks]

EDIT:
Just checked. Uploaded "raw" image, converted via qemu-img command.

[RobertHue]

Hey guys,
Did someone succeed to generate a working AMI based on CentOS-8-ec2-8.1.1911-20200113.3.x86_64.qcow2 ?
I've converted it into the 'raw' format, uploaded into S3, imported as a VM, converted the snapshot into an AMI and then started an EC2 instance.
But it doesn't boot (status check show "1/2: Instance reachability check failed"),

Any hints ?

edit: same issue with GenericCloud image

[aks]

So what, the boot partition/disk/whatever ain't working? How do you know?
When I last did it (C7.x) I used this json as the --disk-containers-file (it's meant to be an OVA after all):
[
{
"Description": "CentOS 7.3 RAW",
"Format": "raw",
"UserBucket": {
"S3Bucket": "my-amis",
"S3Key": "aws-import.raw"
}
}]
Perhaps you did something else?

[TrevorH]

CentOS 8 uses blscfg for grub2 and kernel entries are now in /boot/loader/entries not in /boot/grub2/grub2.cfg. If you use e.g an Amazon xen based host then it will not know about that and will be unable to boot. You can try turning it off in /etc/default/grub but when I tried that here, the next kernel update reverted it for me and started using blscfg again. At that point I gave up.

[RobertHue]

Actually grub and kernel seem (mostly) fine, but the system is not coming up entirely:

Code: Select all

[    1.825011] Freeing unused kernel memory: 316K
[    1.837656] systemd[1]: systemd 239 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD +IDN2 -IDN +PCRE2 default-hierarchy=legacy)
[    1.857331] systemd[1]: Detected virtualization xen.
[    1.861986] systemd[1]: Detected architecture x86-64.
[    1.866750] systemd[1]: Running in initial RAM disk.
Welcome to CentOS Linux 8 (Core) dracut-049-27.git20190906.el8_1.1 (Initramfs)
...
[   OK   ] Reached target Basic System.
         Starting dracut initqueue hook...
[    4.106025] scsi host0: ata_piix
[    4.128088] scsi host1: ata_piix
[    4.130871] ata1: PATA max MWDMA2 cmd 0x1f0 ctl 0x3f6 bmdma 0xc100 irq 14
[    4.136602] ata2: PATA max MWDMA2 cmd 0x170 ctl 0x376 bmdma 0xc108 irq 15
[    4.902011] random: crng init done
[    4.905208] random: 7 urandom warning(s) missed due to ratelimiting
[  133.033085] dracut-initqueue[325]: Warning: dracut-initqueue timeout - starting timeout scripts
[  133.581706] dracut-initqueue[325]: Warning: dracut-initqueue timeout - starting timeout scripts

This timeout error is displayed every 0.5sec, no services are started.