simple logger panics in async/multi-threaded environment possibly due to "local_offset_at" in `time-rs` dependency

[cromulentbanana]

First, thanks for your contribution to rust logging!

As of version 1.16, I've oserved that simple logger will panic when executed in the tokio async runtime with the following output:

thread 'main' panicked at 'Could not determine the UTC offset on this system. Possible causes are that the time crate does not implement "local_offset_at" on your system, or that you are running in a multi-threaded environment and the time crate is returning "None" from "local_offset_at" to avoid unsafe behaviour. See the time crate's documentation for more information. (https://time-rs.github.io/internal-api/time/index.html#feature-flags): IndeterminateOffset', /var/home/dlevin/.cargo/registry/src/github.zerozr99.workers.dev-1ecc6299db9ec823/simple_logger-1.16.0/src/lib.rs:409:85
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Panic in Arbiter thread.

I've provided an example repo that you can clone, in which this bug is triggered:

(analytics) ❯❯ git clone git@github.com:cryptobanana/simpleloggerpanic.git
Cloning into 'simpleloggerpanic'...
remote: Enumerating objects: 15, done.
remote: Counting objects: 100% (15/15), done.
remote: Compressing objects: 100% (12/12), done.
remote: Total 15 (delta 0), reused 15 (delta 0), pack-reused 0
Receiving objects: 100% (15/15), 15.29 KiB | 5.10 MiB/s, done.
(analytics) ❯❯ cd simpleloggerpanic
(analytics) ❯❯ cargo run
   Compiling simpleloggerpanic v0.1.0 (/tmp/simpleloggerpanic)
    Finished dev [unoptimized + debuginfo] target(s) in 5.88s
     Running `/home/dlevin/nobackup/cargo/debug/simpleloggerpanic`
2022-01-09 12:23:04,327 INFO [simpleloggerpanic] Logging with level INFO
2022-01-09 12:23:04,327 INFO [actix_server::builder] Starting 8 workers
thread 'main' panicked at 'Could not determine the UTC offset on this system. Possible causes are that the time crate does not implement "local_offset_at" on your system, or that you are running in a multi-threaded environment and the time crate is returning "None" from "local_offset_at" to avoid unsafe behaviour. See the time crate's documentation for more information. (https://time-rs.github.io/internal-api/time/index.html#feature-flags): IndeterminateOffset', /var/home/dlevin/.cargo/registry/src/github.zerozr99.workers.dev-1ecc6299db9ec823/simple_logger-1.16.0/src/lib.rs:409:85
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Panic in Arbiter thread.

I realize this panic is probably not due directly to the simple logger codebase itself, but perhaps if the problem indeed lies in time-rs, then the dependency time-rs can be updated... or something.

Currently simple_logger v1.15 is not affected by this problem.

Many thanks in advance!

[borntyping]

This is a known issue (#43, #44, #47).

The time crate's documentation provides more information about this issue: https://time-rs.github.io/internal-api/time/index.html#feature-flags, as does this issue: time-rs/time#293.

There's a list of workarounds for simple_logger, depending on how you'd prefer to deal with the issue (mostly taken from #43).

1. Use UTC time: SimpleLogger::new().with_utc_timestamps().init().unwrap();
2. Disable timestamps: SimpleLogger::new().without_timestamps().init().unwrap();
3. Use the time crate's unsafe features: RUSTFLAGS="--cfg unsound_local_offset" (docs)

Older versions of simple_logger use a different dependency (chrono) which is vulnerable to the same problem the time crate is trying to avoid. The unsound_local_offset flag from the time crate makes it work the same way as the chrono crate. This problem still affects the latest version of the time crate.

[borntyping]

UTC timestamps are now the default in simple_logger 2.0.0.