janhoy commented on issue #1327: SOLR-13942: /api/cluster/zk/* to fetch raw ZK data URL: https://github.com/apache/lucene-solr/pull/1327#issuecomment-601604990 Not sure I like the magic switch between returning actual data and returning a list of metdatata simply based on whether a znode happens to have children or not. This is not deterministic. I'd rather have an explicit action, e.g. `GET /api/cluster/zookeeper/collections/_children` to list children for a node, and let pure `GET <path>` calls always return UTF-8 data. Futher, it must be possible to retrieve the metadata of any znode. The existing `/admin/zookeeper` API returns both metadata (ctime etc), utf-8 data and children in the same response. So consider either a `GET /api/cluster/zookeeper/collections/_metadata` call, or putting the metadata in HTTP headers `X-solr-zkmeta-ctime: foo`? An example to show the problem is the `/collections` znode which happens to have no children on a clean node after install. Try this: docker run --rm -ti solr:8.4.1 bash solr start -c curl "http://localhost:8983/solr/admin/zookeeper?path=/collections&detail=true"; It will return `children_count:0`. Your current impl would just return the data of that node, i.e. an empty string, but after creating the first collection it would return a list, and you'd need to add `leaf=true` to get to the data, which is very inconsistent.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
