fuel-dev team mailing list archive

[Sergey Vasilenko <svasilenko@xxxxxxxxxxxx>]

>
>
>
>    1. Do we have bug reported for that "not fatal" error in puppet in the
>    middle of installation? If it is expected that there is no connectivity for
>    a certain amount of time - it is fine, but we should not log that as an
>    error.
>    2. Can you please upload this ISO to Google Drive?
>    3. Did you try to run HealtCheck against this deployment (video is
>    long, so I was jumping over it and might miss some stuff)?
>
>
1. I catched this bug only on this deploy. It happens due this deploy was
very slow. In some puppet resources we use retries instead timeouts. I
think we need inspect code and replace retries to timeouts (in seconds). I
say "non fatal" because this error happened in call prefetch method while
1st network create. After deploy I checked, that all networks created
successfuly. This bug was reported to Jira.

3. No, I don't run health checking. I check possibility of start two VM per
tenant (on different computes) and network connectivity betwen them and
gateway.

4. This ISO was build from "havana" branch of fuel-lib repo.