There are a lot of options in persistConfig that aren't mentioned in the README. We should add a section to the README that documents each option you can declare in a persistConfig.

Prometheus v2.26.0 has the flag --enable-feature=promql-negative-offset to enable negative offset.
It works fine.

But I can't enable it with Thanos 0.19.0.
It looks like the flag is not available for Thanos Query.
So the next query my_metric_status offset -24h fails with the error 1:31: parse error: unexpected <op:-> in offset, expected duration

Is this feature implemented?

Before opening, please confirm:

• I have installed the latest version of the Amplify CLI (see above), and confirmed that the issue still persists.
• I have searched for duplicate or closed issues.
• I have read the guide for [submitting bug reports](https://github.com/aws-amplify/amplify-cli/blob/master/CONTR

See https://jsfiddle.net/ygaprkbe/2/

In the console you will see {} but it should be the object in the update call.

So far I've only seen this when updating items inside of an array.

Also, most likely a related bug: https://jsfiddle.net/ygaprkbe/3/

The modifications object here is { foo.bar: 444 } which seems similar to this old bug #1195

Where is the list of abbreviations used in the standard? The term "UA" is used extensively throughout the text but not explained anywhere...

https://html.spec.whatwg.org/commit-snapshots/009a7316ad7f1b69de978101b043b9b29878d975/#table-of-contents

Here is an issues with detailed description: NekR/offline-plugin#77

We want to have consistent and structured logs to improve the ability to filter and analyse longhorn-manager logs.
Currently the controller uses the global logrus package logger, the task here would be to switch the log calls to the instance logger that was implemented in longhorn/longhorn-manager#661

While doing this also improve the log messages, where possible.
For an example of a successf