We should switch to not storing the scheme with base_url or how about
completely relying on nodename (and eliminating base_url eventually)


On Mon, Feb 24, 2014 at 2:25 AM, Mark Miller <[email protected]> wrote:

> No, it’s a different issue than " urlScheme property should be
> whitelisted”. The issue is that you can't setup a cluster with http and
> then later switch it to https without some manual workaround steps. You can
> create a fresh cluster with SSL with no manual workaround steps though.
>
> - Mark
>
> http://about.me/markrmiller
>
> On Feb 23, 2014, at 2:04 PM, Simon Willnauer <[email protected]>
> wrote:
>
> > mark, I am a bit confused - the issue you are mentioning here is not
> > fixed yet, right? It's not part of
> >
> > r1570798 | noble | 2014-02-22 07:50:02 +0100 (Sat, 22 Feb 2014) | 1 line
> > SOLR-3854 urlScheme property should be whitelisted
> >
> > ?
> >
> > On Sat, Feb 22, 2014 at 10:44 PM, Mark Miller <[email protected]>
> wrote:
> >> I have simliar feelings to the white list issue (which looks like it
> did make it in). You can still use the feature, it’s a new feature and so
> no regression, and so I’d vote to document the limitation around migrating
> from http to https (you have to start with https without manual work) and
> address this in a 4.7.1 or 4.8.
> >>
> >> I do think its something we should address if a more serious issue
> causes a respin - it’s a straightforward fix - we should always be using
> the coreNodeName to match state - never the url or address.
> >>
> >> - Mark
> >>
> >> http://about.me/markrmiller
> >>
> >> On Feb 22, 2014, at 8:19 AM, Steve Davids <[email protected]> wrote:
> >>
> >>> Hate to bring this up, though it must have gotten lost in the shuffle.
> When migrating from http -> https in SOLR-3854 shards aren’t obtaining
> their old core node name and resuming their prior assignments. This is
> because the base_url is being compared instead (which changed) instead of
> something more constant like the node_name. A patch was attached yesterday:
> https://issues.apache.org/jira/browse/SOLR-3854?focusedCommentId=13908014&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13908014.
> It is a quick patch that hasn’t really been tested yet, will do so later
> this evening.
> >>>
> >>> The current work-around would be that if clients want to migrate to
> https, they will need to shutdown their servers, migrate the cluster
> state’s base_url from http to https and bring the server back up.
> >>>
> >>> -Steve
> >>>
> >>> On Feb 22, 2014, at 5:46 AM, Simon Willnauer <
> [email protected]> wrote:
> >>>
> >>>> Please vote for the forth Release Candidate for Lucene/Solr 4.7.0
> >>>>
> >>>> you can download it here:
> >>>>
> http://people.apache.org/~simonw/staging_area/lucene-solr-4.7.0-RC4-rev1570806/
> >>>> or run the smoke tester directly with this commandline (don't forget
> >>>> to set JAVA6_HOME etc.):
> >>>>
> >>>> $ python3.2 -u dev-tools/scripts/smokeTestRelease.py
> >>>>
> http://people.apache.org/~simonw/staging_area/lucene-solr-4.7.0-RC4-rev1570806/
> >>>> 1570806 4.7.0 /tmp/smoke_test_4_7
> >>>>
> >>>> Smoketester said: SUCCESS!
> >>>>
> >>>> here is my +1
> >>>>
> >>>> This RC includes the following fixes compared to RC3
> >>>>
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570798 | noble | 2014-02-22 07:50:02 +0100 (Sat, 22 Feb 2014) | 1
> line
> >>>> SOLR-3854 urlScheme property should be whitelisted
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570795 | noble | 2014-02-22 07:42:22 +0100 (Sat, 22 Feb 2014) | 1
> line
> >>>> SOLR-5762 broke backward compatibility of Javabin format
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570772 | sarowe | 2014-02-22 01:49:11 +0100 (Sat, 22 Feb 2014) | 1
> line
> >>>> Fix CHANGES.txt to reflect the twisted evolution and current state of
> >>>> the Admin UI "Files" conf directory File Browser.  (merged branch_4x
> >>>> r1570771)
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570741 | sarowe | 2014-02-21 23:51:47 +0100 (Fri, 21 Feb 2014) | 1
> line
> >>>> LUCENE-5465: Solr Contrib map-reduce breaks Manifest of all other JAR
> >>>> files by adding a broken Main-Class attribute (merged trunk r1570738)
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570628 | sarowe | 2014-02-21 17:38:59 +0100 (Fri, 21 Feb 2014) | 1
> line
> >>>> SOLR-5729: intellij config (merge trunk r1570626)
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570576 | mvg | 2014-02-21 15:06:41 +0100 (Fri, 21 Feb 2014) | 2
> lines
> >>>> Fixed typo in CHANGES.txt for issue LUCENE-5399 and moved that issue
> >>>> under optimizations.
> >>>>
> ------------------------------------------------------------------------
> >>>> r1570562 | mikemccand | 2014-02-21 13:49:47 +0100 (Fri, 21 Feb 2014)
> | 1 line
> >>>> LUCENE-5461: fix thread hazard in ControlledRealTimeReopenThread
> >>>> causing a possible too-long wait time when a thread was waiting for a
> >>>> specific generation
> >>>>
> >>>> ---------------------------------------------------------------------
> >>>> To unsubscribe, e-mail: [email protected]
> >>>> For additional commands, e-mail: [email protected]
> >>>>
> >>>
> >>>
> >>> ---------------------------------------------------------------------
> >>> To unsubscribe, e-mail: [email protected]
> >>> For additional commands, e-mail: [email protected]
> >>>
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [email protected]
> >> For additional commands, e-mail: [email protected]
> >>
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [email protected]
> > For additional commands, e-mail: [email protected]
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>


-- 
-----------------------------------------------------
Noble Paul

Reply via email to