You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When a page generates a LDAP query that returns a huge number of records, then all of them are processed on the backend to generate a HTML page containing all the results. Pagination is completely handled on the frontend side, hiding/showing elements corresponding to the page number.
There are drawbacks to this approach:
when the HTML page is too long to generate, the request may reach reverse-proxies timeout
the response may be very huge (measured >50MB)
The pagination system architecture should be reconsidered to only return current page of results and request the backend at each page change.
The text was updated successfully, but these errors were encountered:
naïve approach: each time a search is requested, read every entry in the LDAP directory. This request can be slow (~5s for 5000 entries returned)
store the results in the user session (php session). advantage: when another user add or modify an entry while a first one is doing a paginated search, the one doing the search keeps having a coherent view of the results
store the result in a global cache (shared between users). Drawbacks: deal with the synchronization issues between cache and directory
When a page generates a LDAP query that returns a huge number of records, then all of them are processed on the backend to generate a HTML page containing all the results. Pagination is completely handled on the frontend side, hiding/showing elements corresponding to the page number.
There are drawbacks to this approach:
The pagination system architecture should be reconsidered to only return current page of results and request the backend at each page change.
The text was updated successfully, but these errors were encountered: