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
The UI and HTTP API end point (/v1/query) is not respecting the history setting.
At least not when using the jdbc driver but also seems to return too many queries in general.
We ran a query too fast(100s per second) via a jdbc driver . It was a fast query (sub 20 ms)
SELECT TABLE_CAT, TABLE_SCHEM, TABLE_NAME, TABLE_TYPE, REMARKS,
TYPE_CAT, TYPE_SCHEM, TYPE_NAME, SELF_REFERENCING_COL_NAME, REF_GENERATION
FROM system.jdbc.tables
WHERE TABLE_CAT = 'iceberg' AND TABLE_SCHEM LIKE 'sa' ESCAPE '' AND TABLE_NAME LIKE '%' ESCAPE '' AND TABLE_TYPE IN ('TABLE')
ORDER BY TABLE_TYPE, TABLE_CAT, TABLE_SCHEM, TABLE_NAME
It's a bug on our side but it brought out this issue.
When hitting /v1/query or looking at the trino UI I am getting back 28 K (thousand!) finished queries.
This brings the trino UI to a halt and taxes my own code when looking at /v1/query
1.) This needs to be fix and respect the max_history setting of 100(default). We do not set this, assuming the default is in place. I misunderstood the config option but this still causes and issue - see extra comment.
2.) The query endpoint could use a limit=x or a sort=asc, desc
Side note: is there a way to get the same data through a normal trino query?
The text was updated successfully, but these errors were encountered:
I reread the docs, and it seems I mis understood. It reads more that the max-history is only cleared after the min-expired.
This still causes the trino UI issues(and the /v1/query endpoint issues) if there are a lot of queries in a short period of time. This includes crashing browser tabs.
A limiter to the endpoint request still seems like a relevant idea.
Linux centos 7.
The UI and HTTP API end point (/v1/query) is not respecting the history setting.
At least not when using the jdbc driver but also seems to return too many queries in general.
We ran a query too fast(100s per second) via a jdbc driver . It was a fast query (sub 20 ms)
SELECT TABLE_CAT, TABLE_SCHEM, TABLE_NAME, TABLE_TYPE, REMARKS,
TYPE_CAT, TYPE_SCHEM, TYPE_NAME, SELF_REFERENCING_COL_NAME, REF_GENERATION
FROM system.jdbc.tables
WHERE TABLE_CAT = 'iceberg' AND TABLE_SCHEM LIKE 'sa' ESCAPE '' AND TABLE_NAME LIKE '%' ESCAPE '' AND TABLE_TYPE IN ('TABLE')
ORDER BY TABLE_TYPE, TABLE_CAT, TABLE_SCHEM, TABLE_NAME
It's a bug on our side but it brought out this issue.
When hitting /v1/query or looking at the trino UI I am getting back 28 K (thousand!) finished queries.
This brings the trino UI to a halt and taxes my own code when looking at /v1/query
1.)
This needs to be fix and respect the max_history setting of 100(default). We do not set this, assuming the default is in place.I misunderstood the config option but this still causes and issue - see extra comment.2.) The query endpoint could use a limit=x or a sort=asc, desc
Side note: is there a way to get the same data through a normal trino query?
The text was updated successfully, but these errors were encountered: