Scott Parkerson

This is definitely `mktime`'s fault, as `fromdate` is just an alias for `strptime("%Y-%m-%dT%H:%M:%SZ")|mktime`. ``` echo 1665511200 | jq -c 'todate|strptime("%Y-%m-%dT%H:%M:%SZ")' [2022,9,11,18,0,0,2,283] echo "[2022,9,11,18,0,0,2,283]" | jq mktime 1665514800

@mkj-pendo that checks out, as the code for `mktime` ... is [messy](https://github.com/stedolan/jq/blob/77417c1335a12c4ceef469caf38c0cbfb6315b45/src/builtin.c#L1266-L1308).

Tried this on `jq 1.7.1`, and it looks good ``` ❯ echo "[2022,9,11,18,0,0,2,283]" | jq mktime 1665511200 ❯ echo 1665511200 | jq -c 'todate|strptime("%Y-%m-%dT%H:%M:%SZ")' [2022,9,11,18,0,0,2,283] ❯ echo 1665511200 | jq...

I think I can explain the test code's performance as due to the interaction between goroutine scheduling and connection pooling. `MaxPoolSize` is, by default, the number of CPUs * 10,...

I'm curious if option (3) is a version of what [StackExchange.Redis](https://stackexchange.github.io/StackExchange.Redis/) has implemented with [its approach to multiplexing](https://stackexchange.github.io/StackExchange.Redis/PipelinesMultiplexers#multiplexing). It seems to be a good idea for _many_ applications except for...

> Please create a bug describing the issue. I did, didn't I? It's #10684.

I've addressed your comments, @bhshkh -- let me know what you think.

@bhshkh -- > To minimize friction, consider setting 'Allow edits from maintainers' on the PR, which will enable project committers and automation to update your PR. Sadly, I created this...

Ok, apparently this is a losing battle, as I cannot continually fight the progress of the monorepo changing every day. I'm going to close this PR and reopen it using...

I am willing to bet this you are being bit by #2871. We had issues happen to us where a connection pool timeout occurred inside the pipeline for one or...