Appian Community
Site
Search
Sign In/Register
Site
Search
User
DISCUSS
LEARN
SUCCESS
SUPPORT
Documentation
AppMarket
More
Cancel
I'm looking for ...
State
Not Answered
Replies
7 replies
Subscribers
4 subscribers
Views
6327 views
Users
0 members are here
Share
More
Cancel
Related Discussions
Home
»
Discussions
»
Reports
Query Process Analytics getting time out
shailjas
over 8 years ago
Hi,
We have a task report, even if the filters are applied to the report it is fetching all the data.
Since the number of active tasks for the application is more than 1000, it increases the CPU utilization.
We are receiving some warning logs because of this.
snippets from the logs : "Total: Total (Report named 'CIC Tasks Report' (report document id 14972), maximum report time 2000ms) affected 694 rows"
To reduce this we are thinking of reducing the unnecessary columns from the report and to increase the timelimit in custom.properties for this.
Can someone help regarding this.
OriginalPostID-231189
Discussion posts and replies are publicly visible
0
Alok Mohare
Appian Employee
over 8 years ago
Review the following link from our Documentation to tackle the issue of Portal Reports that are timing out.
forum.appian.com/.../Report_Performance_Details.html
Do not increase the Report timeout limit in custom.properties. This may cause further degradation in performance.
My suggestion would be to work on optimizing the existing report that you have.
forum.appian.com/.../Report_Performance_Details.html
You can check the "Report Performance Details" to figure out what columns take the most amount of time to populate and work on optimizing the expressions within those columns.
Hope this helps!
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel
0
karthiksaik
over 8 years ago
You can select and hit with query only few columns that your are displaying in the report so that the report wont hit on all the columns this will reduce the time taken to fetch.
selection: a!querySelection(
columns: {
a!queryColumn(field: "Column1"),
a!queryColumn(field: "Column2")
}
),
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel
0
sikhivahans
over 8 years ago
@shailjas Did you consider making use of the pagingInfo and retrieve the records in batches? Or did it not support your use case?
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel
0
shailjas
over 8 years ago
@sikhivahans We are applying paging info but it seems like it is fetching the data from the memory
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel
0
sikhivahans
over 8 years ago
@shailajas In this context, what I meant to say is, each query should be made afresh with the changed pagingInfo. That is, while loading the report, you will fetch only 25 rows (let's assume you are asked by Business to show only 25 rows per page) in the load(). Get the next 25 rows only when the user tries to paginate, and this should be written in saveInto of the Report Grid. This way, you will hold only 25 rows in the memory or query 25 rows each time. In this case, definitely, there won't be an issue at all even though the columns are many in the process report and further hold complex and time consuming expressions.
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel
0
shailjas
over 8 years ago
Thanks Sikhi,
We are applying paging info but it does not seem to be working.
Even after turning batchSize to 1,its giving us same error.
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel
0
sikhivahans
over 8 years ago
@shailjas In this case, I think you may need to optimise the report. As pointed by @alok.mohare earlier, it might be worth going through the documentation at https://forum.appian.com/suite/help/16.2/Report_Performance_Details.html and try to improve the performance as much as you can. There might be various ways of doing this, for instance, you may split the data across multiple reports or rework on the complex expressions that are being used.
Another way is to increase the value of server.conf.processcommon.MAXIMUM_REPORT_MS in the server but you should keenly observe the @csteward's experience of increasing the value.
Cancel
Vote Up
0
Vote Down
Sign in to reply
Verify Answer
Cancel