Closed
Bug 1478437
Opened 7 years ago
Closed 7 years ago
fx-sig-verify memory tuning needed
Categories
(Cloud Services :: Operations: Miscellaneous, task)
Cloud Services
Operations: Miscellaneous
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: hwine, Unassigned)
Details
Now that we're processing larger files [#46], the memory allocated to the lambda function may also need to be updated.
From today's report (~24 hours after deployment [#47]), we ran out of memory a number of times (which is a hard fail):
```sh
$ analyze_cloudwatch --report --summarize `lf`
24,089 runs
5,275 seconds execution time
7,577 seconds billed
68,447,689 GBi seconds (AWS Billing Unit)
219 average milliseconds per run
4,122 times we used all memory
17% of runs maxing out memory
384 MBi max memory used
0 times run aborted for excessive time
0% of runs exceeding time limit
0 times retry did not succeed
```
The failed runs should only be on the new nightly builds.
[#46]: https://github.com/mozilla-services/fx-sig-verify/issues/46
[#47]: https://github.com/mozilla-services/fx-sig-verify/issues/47
Comment 1•7 years ago
|
||
We have a PR here [0] that addresses this by increasing the memory available to the lambda function to 1024mb.
[0]: https://github.com/mozilla-services/cloudops-deployment/pull/2309
Flags: needinfo?(miles)
Comment 2•7 years ago
|
||
This has landed in prod.
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
You need to log in
before you can comment on or make changes to this bug.
Description
•