SOLR-18091: Separate out core specific info into CoreInfoHandler#4084
SOLR-18091: Separate out core specific info into CoreInfoHandler#4084epugh wants to merge 18 commits intoapache:mainfrom
Conversation
|
|
||
| @Override | ||
| public void handleRequestBody(SolrQueryRequest req, SolrQueryResponse rsp) throws Exception { | ||
| rsp.setHttpCaching(false); |
There was a problem hiding this comment.
Does this request handler need if (AdminHandlersProxy.maybeProxyToNodes(req, rsp, getCoreContainer(req))) { like SystemInfoHandler has? Is that soemthing every class needs?
epugh
changed the title
|
@gerlowskija @igiguere you may be interested in this PR with your work on the V2 apis... |
github-actions
bot
added
documentation
|
I think this could go to 10x as it turns out the Admin UI, which is the primary consumer, was easy to update. |
|
I think that the fact that you got system info at the core level was a bug... not a feature. And that by seperating them and making the urls endpoints clearer, I fixed the bug, and therefore this can be backported to branch_10x.... Thoughts? |
I was working on #4078 instead. Draft... a couple of questions before I make it official. |
|
Okay, this looks ready for final review and hopefully a merge! |
|
I'd like to merge this on Monday.... |
dsmiley
left a comment
dsmiley
left a comment
There was a problem hiding this comment.
Worth mentioning in the upgrade guide. Technically a back-compat break but I'm +1 for 10.1.
changelog/unreleased/SOLR-18091.yml
Outdated
| @@ -0,0 +1,8 @@ | |||
| # See https://github.com/apache/solr/blob/main/dev-docs/changelog.adoc | |||
| title: Seperate the node/container level info from the core level info for the admin end points. | |||
There was a problem hiding this comment.
It'd be more helpful to list the actual affected endpoints
| dirs.add("data", core.getDirectoryFactory().normalize(core.getDataDir())); | ||
| } catch (IOException e) { | ||
| log.warn("Problem getting the normalized data directory path", e); | ||
| dirs.add("data", "N/A"); |
There was a problem hiding this comment.
yuck; lets not add dummy/invalid values
| dirs.add("index", core.getDirectoryFactory().normalize(core.getIndexDir())); | ||
| } catch (IOException e) { | ||
| log.warn("Problem getting the normalized index directory path", e); | ||
| dirs.add("index", "N/A"); |
There was a problem hiding this comment.
i'm happy to make this change, just FYI that is a change compared to 9x and 10.0... I was just moving things, not redoing things...
|
Okay, I think i've responded to feedback and I am looking to merge this on Monday. Hopefully all of this goes away as our V2 API's come online! |
3 participants
https://issues.apache.org/jira/browse/SOLR-18091
Description
Instead of having SystemInfoHandler accessible at two different URLS, and serving different data, just split them and have proper names.
V1: http://localhost:8983/solr/admin/info/system
V2: http://localhost:8983/api/node/system
and
http://localhost:8983/solr/techproducts/admin/info
There appears to be no V2 for the core level
Solution
Make loading pattern clearer. CoreInfoHandler is in ImplicitPlugins.json, while SystemInfoHandler is part of InfoHandler loaded in CoreContainer.
Tests
manual and unit
I still need to check the solr Admin UI to make sure the new /solr/techproducts/admin/info url makes.