You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@knox.apache.org by Kevin Minder <ke...@hortonworks.com> on 2013/03/21 14:10:01 UTC

[DISCUSS] Bulk import of Jira issues

Hi Everyone,
We don't have our Jira setup yet but in preparation I would like to 
discuss if we should bulk import the Jira issues from the Hortonworks 
Jira.  Below I have a list of the currently filed issues.  Personally I 
would like to import them in an effort to move everything into the public.
Kevin.

BUG-4336    Gateway starts even if incorrect master password entered
BUG-4335    Support finer grain control over what is included in the URL 
rewrite
BUG-4334    Prepare Maven project structure for Apache submission
BUG-4333    Support horizontal scalability of gateway via clustering
BUG-4332    Provide option to have gateway behave transparently
BUG-4331    Prepare for Apache submission
BUG-4330    Support translation of cluster internal hostnames to 
external hostnames
BUG-4329    Smart load balancing among gateway cluster intances
BUG-4328    Ensure that all cluster topology details are rewritten for 
Oozie REST APIs
BUG-4327    Prevent Shiro rememberMe cookie from being returned
BUG-4326    Cluster topology must not be exposed in datanode redirect 
query parameters
BUG-4325    Clean up i18n logging and any System.out or printStackTrace 
usages
BUG-4324    Parse and convert Hadoop config files to Gateway cluster 
topology
BUG-4323    Clean up usage and help output from server command line
BUG-4322    Support HUE interacting with a Hadoop cluster via the Gateway
BUG-4321    Add support for Oozie REST APIs
BUG-4320    Change gateway URLs to match service URLs as closely as possible
BUG-4319    Integrate execution of open source governance scanning into 
release build
BUG-4318    Verify LDAP over SSL
BUG-4317    Convert from service lookup pattern to dependency injection 
pattern where possible
BUG-4316    Authentication failure via Shiro/LDAP/BasicAuth doesn't 
produce diagnostic output when misconfigured
BUG-4315    Investigate OpenID/OAuth as identity federation/SSO mechanism
BUG-4314    Cluster topology discovery and synchronization via Ambari
BUG-4313    Integrate hadoop-auth maven module as authentication provider
BUG-4312    Verify authentication via Active Directory using LDAP
BUG-4311    Provide support for hosting Jersey services for the purposes 
of protocol mediation of non-REST services
BUG-4310    Allow plug-ins to be loaded by adding JARs to bin directory 
(or similar)
BUG-4309    Create lifecycle scripts for gateway server
BUG-4308    Limit HTTP basic auth chalenges to HTTPS
BUG-4307    Provide option to have gateway automatically follow WebHDFS 
redirects
BUG-4306    Access Kerberos secured Hadoop cluster via gateway using 
basic auth credentials
BUG-4305    Support federation/SSO using external tokens (MS Azure)
BUG-4304    Access Kerberos secured Hadoop cluster via gateway using SPNEGO
BUG-4303    Support HTTPS gateway endpoints
BUG-4302    Generate audit log of all gateway activity
BUG-4301    Invoke HDFS via gateway using hadoop CLI and FileSystem API
BUG-4300    Utilize hadoop.auth cookie to prevent re-authentication for 
every request
BUG-4299    Administrative REST API
BUG-4298    Misleading logging output when files in 
{GATEWAY_HOME}/deployments changed
BUG-4297    IdentityAssertionHttpServletRequestWrapper should only be 
used when required by the service
BUG-4296    Error message resulting from invalid deployment file are not 
diagnostic
BUG-4295    Several very inefficient implementation of InputStream and 
ServletInputStream need to be rewritten
BUG-4294    Improved test coverage for invalid request/response bodies
BUG-4293    Clean up APIs and SPIs to follow either Fluent or Bean style 
consistently.

Re: [DISCUSS] Bulk import of Jira issues

Posted by Alan Gates <ga...@hortonworks.com>.
I have a couple of concerns about this:

1) I'm not sure how those outside of Hortonworks are supposed to evaluate this.  If they can't see the contents of those bugs how can they vote?  We could provide a raw dump for them to look at I supposed, but that seems painful.

2) Was the git history maintained during the import to Apache?  If so, then I see some value in the JIRA import since the two can still be correlated.  If not, then the value seems dubious since people won't be able to look back and see what source changes relate to what JIRAs.

Alan.

On Mar 21, 2013, at 6:10 AM, Kevin Minder wrote:

> Hi Everyone,
> We don't have our Jira setup yet but in preparation I would like to discuss if we should bulk import the Jira issues from the Hortonworks Jira.  Below I have a list of the currently filed issues.  Personally I would like to import them in an effort to move everything into the public.
> Kevin.
> 
> BUG-4336    Gateway starts even if incorrect master password entered
> BUG-4335    Support finer grain control over what is included in the URL rewrite
> BUG-4334    Prepare Maven project structure for Apache submission
> BUG-4333    Support horizontal scalability of gateway via clustering
> BUG-4332    Provide option to have gateway behave transparently
> BUG-4331    Prepare for Apache submission
> BUG-4330    Support translation of cluster internal hostnames to external hostnames
> BUG-4329    Smart load balancing among gateway cluster intances
> BUG-4328    Ensure that all cluster topology details are rewritten for Oozie REST APIs
> BUG-4327    Prevent Shiro rememberMe cookie from being returned
> BUG-4326    Cluster topology must not be exposed in datanode redirect query parameters
> BUG-4325    Clean up i18n logging and any System.out or printStackTrace usages
> BUG-4324    Parse and convert Hadoop config files to Gateway cluster topology
> BUG-4323    Clean up usage and help output from server command line
> BUG-4322    Support HUE interacting with a Hadoop cluster via the Gateway
> BUG-4321    Add support for Oozie REST APIs
> BUG-4320    Change gateway URLs to match service URLs as closely as possible
> BUG-4319    Integrate execution of open source governance scanning into release build
> BUG-4318    Verify LDAP over SSL
> BUG-4317    Convert from service lookup pattern to dependency injection pattern where possible
> BUG-4316    Authentication failure via Shiro/LDAP/BasicAuth doesn't produce diagnostic output when misconfigured
> BUG-4315    Investigate OpenID/OAuth as identity federation/SSO mechanism
> BUG-4314    Cluster topology discovery and synchronization via Ambari
> BUG-4313    Integrate hadoop-auth maven module as authentication provider
> BUG-4312    Verify authentication via Active Directory using LDAP
> BUG-4311    Provide support for hosting Jersey services for the purposes of protocol mediation of non-REST services
> BUG-4310    Allow plug-ins to be loaded by adding JARs to bin directory (or similar)
> BUG-4309    Create lifecycle scripts for gateway server
> BUG-4308    Limit HTTP basic auth chalenges to HTTPS
> BUG-4307    Provide option to have gateway automatically follow WebHDFS redirects
> BUG-4306    Access Kerberos secured Hadoop cluster via gateway using basic auth credentials
> BUG-4305    Support federation/SSO using external tokens (MS Azure)
> BUG-4304    Access Kerberos secured Hadoop cluster via gateway using SPNEGO
> BUG-4303    Support HTTPS gateway endpoints
> BUG-4302    Generate audit log of all gateway activity
> BUG-4301    Invoke HDFS via gateway using hadoop CLI and FileSystem API
> BUG-4300    Utilize hadoop.auth cookie to prevent re-authentication for every request
> BUG-4299    Administrative REST API
> BUG-4298    Misleading logging output when files in {GATEWAY_HOME}/deployments changed
> BUG-4297    IdentityAssertionHttpServletRequestWrapper should only be used when required by the service
> BUG-4296    Error message resulting from invalid deployment file are not diagnostic
> BUG-4295    Several very inefficient implementation of InputStream and ServletInputStream need to be rewritten
> BUG-4294    Improved test coverage for invalid request/response bodies
> BUG-4293    Clean up APIs and SPIs to follow either Fluent or Bean style consistently.


Re: [DISCUSS] Bulk import of Jira issues

Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
Hey Kevin,

Awesome, great thoughts.

We had to deal with this when we moved OODT from JPL to Apache, even after.
1+ years after I still had an internal JIRA with over 200+ unresolved
issues.
I moved them 1 by 1 (not kidding!) to Apache. It sucked. I closed many
along
the way that I knew we wouldn't deal with at the ASF, and that were dating
back to like 2007.

So, it's really up to you here. If all of the below make sense to work on
Apache Knox (incubating), I say, bring em' over! Helps to decommission
existing
work internally (e.g., at Hortonworks) that's separate from the Apache
stuff.

+1.

Cheers,
Chris


On 3/21/13 6:10 AM, "Kevin Minder" <ke...@hortonworks.com> wrote:

>Hi Everyone,
>We don't have our Jira setup yet but in preparation I would like to
>discuss if we should bulk import the Jira issues from the Hortonworks
>Jira.  Below I have a list of the currently filed issues.  Personally I
>would like to import them in an effort to move everything into the public.
>Kevin.
>
>BUG-4336    Gateway starts even if incorrect master password entered
>BUG-4335    Support finer grain control over what is included in the URL
>rewrite
>BUG-4334    Prepare Maven project structure for Apache submission
>BUG-4333    Support horizontal scalability of gateway via clustering
>BUG-4332    Provide option to have gateway behave transparently
>BUG-4331    Prepare for Apache submission
>BUG-4330    Support translation of cluster internal hostnames to
>external hostnames
>BUG-4329    Smart load balancing among gateway cluster intances
>BUG-4328    Ensure that all cluster topology details are rewritten for
>Oozie REST APIs
>BUG-4327    Prevent Shiro rememberMe cookie from being returned
>BUG-4326    Cluster topology must not be exposed in datanode redirect
>query parameters
>BUG-4325    Clean up i18n logging and any System.out or printStackTrace
>usages
>BUG-4324    Parse and convert Hadoop config files to Gateway cluster
>topology
>BUG-4323    Clean up usage and help output from server command line
>BUG-4322    Support HUE interacting with a Hadoop cluster via the Gateway
>BUG-4321    Add support for Oozie REST APIs
>BUG-4320    Change gateway URLs to match service URLs as closely as
>possible
>BUG-4319    Integrate execution of open source governance scanning into
>release build
>BUG-4318    Verify LDAP over SSL
>BUG-4317    Convert from service lookup pattern to dependency injection
>pattern where possible
>BUG-4316    Authentication failure via Shiro/LDAP/BasicAuth doesn't
>produce diagnostic output when misconfigured
>BUG-4315    Investigate OpenID/OAuth as identity federation/SSO mechanism
>BUG-4314    Cluster topology discovery and synchronization via Ambari
>BUG-4313    Integrate hadoop-auth maven module as authentication provider
>BUG-4312    Verify authentication via Active Directory using LDAP
>BUG-4311    Provide support for hosting Jersey services for the purposes
>of protocol mediation of non-REST services
>BUG-4310    Allow plug-ins to be loaded by adding JARs to bin directory
>(or similar)
>BUG-4309    Create lifecycle scripts for gateway server
>BUG-4308    Limit HTTP basic auth chalenges to HTTPS
>BUG-4307    Provide option to have gateway automatically follow WebHDFS
>redirects
>BUG-4306    Access Kerberos secured Hadoop cluster via gateway using
>basic auth credentials
>BUG-4305    Support federation/SSO using external tokens (MS Azure)
>BUG-4304    Access Kerberos secured Hadoop cluster via gateway using
>SPNEGO
>BUG-4303    Support HTTPS gateway endpoints
>BUG-4302    Generate audit log of all gateway activity
>BUG-4301    Invoke HDFS via gateway using hadoop CLI and FileSystem API
>BUG-4300    Utilize hadoop.auth cookie to prevent re-authentication for
>every request
>BUG-4299    Administrative REST API
>BUG-4298    Misleading logging output when files in
>{GATEWAY_HOME}/deployments changed
>BUG-4297    IdentityAssertionHttpServletRequestWrapper should only be
>used when required by the service
>BUG-4296    Error message resulting from invalid deployment file are not
>diagnostic
>BUG-4295    Several very inefficient implementation of InputStream and
>ServletInputStream need to be rewritten
>BUG-4294    Improved test coverage for invalid request/response bodies
>BUG-4293    Clean up APIs and SPIs to follow either Fluent or Bean style
>consistently.