Repository navigation
Replies: 3 comments
|
Thanks for this report, I'll check it and prepare solution for it. |
0 replies
|
Done in 3.2.0 release. |
0 replies
|
@MateosTDD - let me know if this index helped in 3.2.0 thanks |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
We are testing a deployment based on Skoruba Duende IdentityServer Admin 3.1. The Admin UI loads, and client and user lists work, but some dashboard counters, including the total number of clients, remain as loading placeholders.
The API request to GetDashboardIdentityServer fails after the 30-second SQL command timeout. The stack trace points to AuditLogRepository.GetDashboardAuditLogsAsync and GetDashboardAuditLogsAverageAsync. One of the logged queries groups [AuditLog] records by CONVERT(date, [Created]) after filtering on Created.
Our AuditLog table has approximately 794,000 rows and only a clustered primary-key index on Id; there is no index on Created. During the failures, Azure SQL’s Data IO percentage briefly reached 93–97%. We have not captured an execution plan, so we cannot yet confirm the exact access path or attribute all of that I/O to this query.
Expected: Dashboard counters load, or remain available if audit-log statistics fail.
Actual: A timeout in the audit-log portion appears to fail the dashboard response, leaving unrelated counters in their loading state. Client and user list endpoints continue to work.
Is an index on AuditLog.Created expected to be created by the 3.1 migrations? Would you recommend changing these dashboard queries, or separating the audit statistics from the counters so a timeout does not block the entire dashboard?
All reactions