My MemoryStore analytics show that I am not even close to hitting my limits.
Usage:
Requests:
Given that I’m barely at 1% utilization on these charts, you’d expect smooth sailing. Let’s check our response chart:
Failure Status Requests:
These analytics show only 3 non-successful requests over the last 7 days, which is definitely the smooth sailing we anticipated. 1 TotalRequestsOverLimit, which is strange when the above graph shows my request count nowhere near the limit. 2 InternalError, which is annoying but no big deal. Cool.
Then we go to my error logs and search for “MemoryStoreService”… and it tells a different story.
MemoryStores are failing far more than the above graphs are claiming.
By my error log counts, I’m seeing a very different story. I’m not going through 20 pages of error logs and adding up the counts, but I can assure you it’s a lot higher than 3.
Expected behavior
I have several expectations that are not being met:
- MemoryStoreService should be a lot more reliable- this is an unacceptable amount of
InternalErrorresponses. - My analytics should reflect my true situation. I am unsure what my real utilization is. Are the throttling errors mistaken, or am I overutilizing?
- MemoryStoreService errors are quite unhelpful.
MemoryStoreService: InternalError: Internal Error.does not provide any insight into what is going wrong and how I can work around it.
RequestThrottled: Recent MemoryStores requests hit one or more limits.Are my total requests too high, or is it failing because I hit the limit for a single queue/map?
