This isn't good. The SSS_TIME_LIMIT_EXCEEDED erro...
# suitescript
s
This isn't good. The SSS_TIME_LIMIT_EXCEEDED error I had yesterday came back again today in our M/R script. Has anyone seen anything similar in their environment in the last few days? Within the map phase of a Map/Reduce script, what can cause SSS_TIME_LIMIT_EXCEEDED ? Is it total execution time, or just the time the map instance has been running?
s
I have not seen any new ones popping up, that error usually comes from the individual execution time for that particular invocation of map, not the cumulative total of every time it is called
We process large blocks of information and the only way we've found to fix is to send smaller blocks to each invocation (which increases total MR runtime but each invocation is shorter)
s
The odd thing with this, is that at the point the script fails, it has not even made a single SuiteScript api call, just a few lines of plain old JavaScript (creating a JS object, assigning some properties, etc.). The only thing different about the line where it times out is that it calls
new Date()
That seems like it shouldn't be a problem, and it never has been until yesterday.
s
How many places you got this running? Are there 2021.2 accounts vs 2021.1 and its only failing in 2021.2 ones? The timing of it starting to fail seems like an unlikely coincidence
s
It's running in production and three sandboxes + release preview. Only failing in production (still on 2021.1)
The sandboxes are also 2021.1
b
you can try taking a look at the details of the Map/Reduce Script status page
id expect a timeout error to occur if any of the stages took longer than an hour
s
the timeout occurs within the first 7 or 8 minutes of the M/R script starting up. that's part of what is so strange
Huh, I think I see a pattern. Yesterday, the script ran, and processed about 4,000 results, from 2:01 to 2:03 pm, then nothing happened for 5 minutes, and all of the map phases died at 2:08 pm. Then again today, the M/R script ran and processed about 3,00 results from 2:02 to 2:06 pm, then nothing happened for five minutes and the map phases all died again, at 2:11 pm.
So it looks like it is running normally for a while, then something happens and all of the map phases just hang for 5 minutes doing nothing, then fail with a timeout.
s
Doesnt sound fun to diagnose or explain
s
According to SuiteAnswers, the timeout for Map/Reduce scripts should be 3600 seconds (60 minutes) so this makes no sense. And yeah, I feel like I am going to have to bubble this up to support, since the code hasn't changed in years.
s
I thought each invocation of map or reduce had like 15 minutes for some reason
s
I have Yield After Minutes set to 20 (I think it can be anything from 3-60)
d
@scottvonduhn I've been seeing SSS_TIME_LIMIT_EXCEEDED as well all of a sudden. I am curious if it has anything to do with 2021.2 update for my account on the 17th. I have at least 2 scripts throwing that error. Both of them interact with the File Cabinet, but so far that is the only correlation I have seen between the two scripts. I'm eager to know any other details that you might have. I'll fish out the error messages I have to see if it helps find a cause.
Environment: Production Date & Time: 09/08/2021 01:04 pm Execution Time: 0.00s Script Usage: 0 Script: GPB Get Delivery Slip Type: Workflow Action Function:
onAction
Error:
SSS_TIME_LIMIT_EXCEEDED
Script Execution Time Exceeded. Stack Trace:
onAction(/SuiteScripts/WizardGPB/Wizard 2.0/gpbGetDelSli.js:62)
s
My script doesn't even interact with the filing cabinet. In fact, all it does is either sends an email, or creates a Task, for each search result. No other SuiteScript calls are made besides those. The rest is just a bunch of numerical and Date comparisons (it's a collections/dunning script) to see how overdue and by how much each customer is, and whether to just email them or create a task for a collections agent. What is your script doing at/on the line referenced in the stack trace? Mine is creating a new Date (current date/time), and that is where it is consistently timing out. I did open a case with NS support yesterday, as this error makes no sense, and did just start suddenly sometime between Friday and yesterday (our script doesn't run on the weekend). If it matters, this is a SS 2.0 Map/Reduce script. I haven't tried switching it to SS 2.1 to see if that would make a difference.
j
@David Hayden @scottvonduhn Our instance too started getting this yesterday seemingly out of the blue and intermittently at that. We are not receiving the release until mid-October but it certainly has that smell of when NetSuite is making changes behind the scenes.