fuel-dev team mailing list archive
-
fuel-dev team
-
Mailing list archive
-
Message #00433
Re: Ubuntu HA failures on powerful servers
>
>
> Sergey - please make an effort to write a good bug description for it.
> Right now it's hard to understand what the bug is about, and how it was
> fixed.
>
>
This bug was not fixed. Only workarrounded (by insert sleeps) for customer.
The essence of what is happening is this.
All processes in Pacemaker - asynchronous.
Puppet works synchronously.
We use modifyed manifests from puppetforge.
For fixing we should add missing verifications to providers code.
Now we analize return code of crm utility call. It's not enouпh.
This werifications should check and wait for end previous operation in
cluster resource manager.
References