Contents Cache failed for <Object name>

Brief summary of issue / Description of requested feature:

This happened before in issue #1064 and I thought it was resolved, but when checking my logs for the past week, I occasionally will see dozens of lines of error messages of Contents cache failed for <object name here>, a character or a room. Reloading fixes it, though I’m unsure why this would suddenly pop back after seeming to be resolved for months.

Steps to reproduce the issue / Reasons for adding feature:

1. Long enough period between reloads seems to occasionally create this error.

Error output / Expected result of feature

Dozens of lines of Contents cache failed for <object name>. Usually more than one object is affected - somewhere between 1-6 characters or rooms seems to be typical.

Extra information, such as Evennia revision/repo/branch, operating system and ideas for how to solve / implement:

Probably related to issue #1064

[Griatch]

@TehomCD, @aceamro @strikaco I have investigated the error and found this to be benign. What happens is that the central idmapper cache is flushed (this happens on a timer to cut down on memory usage), meaning the lookup fails. Reloading from database rebuilds that part of the cache and it will after that not longer give the error. So this is the normal operation of rebuilding the cache after a flush. It’s just the ‘error’ in the log that’s confusing. I have removed the logging of this (it shouldn’t even be a warning).

[strikaco]

In #1064, the reason for this error was the firing of the cache flush mechanism (see my comment that reproduced it). This will start firing if the memory footprint of your game is increasing past the memory limiter breakoff point as set in settings. Is the memory limiter firing for you in the same way now?

@Griatch I’m starting to experience this error myself. The memory limiter is definitely kicking in; the rest of the conditions are as reported-- 12h+ uptime, active (10k+ objects created).

But everything still seems to be working inasmuch as I can observe it. Is this log entry misclassified (should it be a warn event?) or is its presence indicative of an actual error that needs addressing?