-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/31177/#review73133
-----------------------------------------------------------
ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/package/alerts/alert_webhcat_server.py
<https://reviews.apache.org/r/31177/#comment119362>
Service isn't called Templeton anymore (property names aside). OK_MESSAGE should say
WebHCat.
ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/package/alerts/alert_webhcat_server.py
<https://reviews.apache.org/r/31177/#comment119364>
We've repeated this pattern so much. Time for an r_m function? Curl(host, [kinit], [keytab],
[principal] [timeout] , etc) ?
- Nate Cole
On Feb. 18, 2015, 9:43 p.m., Jonathan Hurley wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/31177/
> -----------------------------------------------------------
>
> (Updated Feb. 18, 2015, 9:43 p.m.)
>
>
> Review request for Ambari, Nate Cole and Tom Beerbower.
>
>
> Bugs: AMBARI-9697
> https://issues.apache.org/jira/browse/AMBARI-9697
>
>
> Repository: ambari
>
>
> Description
> -------
>
> After enabling Kerberos, the WebHCat Server Status alert is always CRITICAL. This is
caused by two problems:
>
> 1) The lack of any kinit and curl combination when the cluster is kerberized
> 2) Assuming that cluster-env/security_enabled means SSL (which is does not!)
>
> Consulted with the Kerberos folks that cluster-env/security_enabled is used to indicate
kerberization.
>
> It appears that WebHCat doesn't have an SSL mode; So going back to the 1.7.0 alert with
a hard-coded http:// scheme for now.
>
>
> Diffs
> -----
>
> ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/package/alerts/alert_webhcat_server.py
44840de
> ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/WEBHCAT/package/files/alert_webhcat_server.py
44840de
>
> Diff: https://reviews.apache.org/r/31177/diff/
>
>
> Testing
> -------
>
> Tested in a kerberized cluster to ensure that the changed code fixed the CRITICAL alert.
Took the service down to verify that the alert triggered correctly.
>
> Tested the alert in a regular cluster to ensure the non-kerberized path worked as it
has in the past.
>
>
> Thanks,
>
> Jonathan Hurley
>
>
|