Updated: Dec 7, 2017, 8AM PT
Our previous patch provided the fix of ext4 bug with KernelCare included a bug in unpatch. The bug was triggered by latest patch, and caused some number of servers to crash. The effect of the bug is that servers that use ext4 will crash. If they didn't crash - they will not crash now.
Note that the new patch that triggered the uninstall was delivered in real-time and 12h feeds only. 24h & 48h feeds are not affected. As this patch will not be delivered to 24h & 48h feeds, people on those feeds should not see crashes.
Affected:
- CloudLinux 7 & CloudLinux 6 Hybrid kernels (3.10 lve version before 1.4.50)
- Servers that use ext4
Not Affected:
- Any other kernels & distributions, including CloudLinux 6 kernels;
- CloudLinux 7 & CloudLinux 6 Hybrid lve1.4.50+ version kernels.
We will make every effort by the asap to have a solution that will automatically update the patches with a minimum risk of repeated crashes.
Due to bug being triggered by unpatching, we strongly advise you - don't run the following commands:
kcarectl --update
kcarectl --unload
uninstall kernelcare
As this can trigger crash (CloudLinux 7 & CloudLinux 6 Hybrid only, all feeds).
Note. More info about different feeds.
Delayed feed ensures that your servers will receive patches 24 hours after the release.
To implement the delayed feed, add PREFIX=’N’h to /etc/sysconfig/kcare/kcare.conf.
We have options for:
- PREFIX=12h
- PREFIX=24h
- PREFIX=48h
