--
You received this message because you are subscribed to the Google Groups "Magnolia User Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to user-list+...@magnolia-cms.com.
To view this discussion on the web, visit https://groups.google.com/a/magnolia-cms.com/d/msgid/user-list/04467de1-0351-48a5-8e02-b341ac9af8e9n%40magnolia-cms.com.
6.2.44 already have total in delivery
you can check specs here: https://docs.magnolia-cms.com/product-docs/6.2/developing/api/delivery-api/#:~:text=Result%20count%20and%20pagination%20metadata
On Wed, May 15, 2024 at 6:09 PM Patrik Pahulák <patrik.pahulak@servermechanics.cz> wrote:Hello, we have this one specific workspace that is kinda the most important workspace in our project. The workspace contains 7k ish nodes atm. The issue we are faced with are slow(ish) queries. The workspace is structured into folders that contain the nodes (all of these nodes are of the same type). This workspace is queried quite often. We have a rest api endpoint that handles querying this workspace with filtering and pagination (around 10 properties the filter works with). To make this work, we limit the queries to a couple of nodes per query (10-50). However we also need to return a total number of nodes that fit the query for our pagination (first, page, next page, last page, ...). AFAIK there isnt really a COUNT or anything like that to get this result so we are basically forced to run the query with no limit/offset to get the total number of nodes. This can get pretty slow, atm it's at around 1.5 seconds, which is not that bad, but the number of nodes is expected to grow, which will slow this down even more. We know for a fact that the slow part is the process of running the query with no limit.
There is no index setup on our side. I have tried setting up an index for the workspace, but I have most likely failed, since I do not see any speed improvements. I was wondering if setting up an index would even help us out in the first place and if so I was wondering if anyone could share some example index setup with Magnolia. Our repository setup is basically identical to this one, atleast the SearchIndex part.We are running magnolia 6.2.44.
Thanks a lot
--
You received this message because you are subscribed to the Google Groups "Magnolia User Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to user-list+unsubscribe@magnolia-cms.com.
To view this discussion on the web, visit https://groups.google.com/a/magnolia-cms.com/d/msgid/user-list/04467de1-0351-48a5-8e02-b341ac9af8e9n%40magnolia-cms.com.
Maybe you might consider adjusting the index configuration file to index only the properties you need for the queries. This page goes into some of the details https://docs.magnolia-cms.com/product-docs/6.2/features/search/jackrabbit-search/ and there is also this https://jackrabbit.apache.org/archive/wiki/JCR/IndexingConfiguration_115513411.htmlHTHRichOn Wednesday, May 15, 2024 at 2:33:56 PM UTC-4 undi...@gmail.com wrote:6.2.44 already have total in delivery
you can check specs here: https://docs.magnolia-cms.com/product-docs/6.2/developing/api/delivery-api/#:~:text=Result%20count%20and%20pagination%20metadataOn Wed, May 15, 2024 at 6:09 PM Patrik Pahulák <patrik....@servermechanics.cz> wrote:Hello, we have this one specific workspace that is kinda the most important workspace in our project. The workspace contains 7k ish nodes atm. The issue we are faced with are slow(ish) queries. The workspace is structured into folders that contain the nodes (all of these nodes are of the same type). This workspace is queried quite often. We have a rest api endpoint that handles querying this workspace with filtering and pagination (around 10 properties the filter works with). To make this work, we limit the queries to a couple of nodes per query (10-50). However we also need to return a total number of nodes that fit the query for our pagination (first, page, next page, last page, ...). AFAIK there isnt really a COUNT or anything like that to get this result so we are basically forced to run the query with no limit/offset to get the total number of nodes. This can get pretty slow, atm it's at around 1.5 seconds, which is not that bad, but the number of nodes is expected to grow, which will slow this down even more. We know for a fact that the slow part is the process of running the query with no limit.
There is no index setup on our side. I have tried setting up an index for the workspace, but I have most likely failed, since I do not see any speed improvements. I was wondering if setting up an index would even help us out in the first place and if so I was wondering if anyone could share some example index setup with Magnolia. Our repository setup is basically identical to this one, atleast the SearchIndex part.We are running magnolia 6.2.44.
Thanks a lot--
You received this message because you are subscribed to the Google Groups "Magnolia User Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to user-list+...@magnolia-cms.com.