Problem description:
Infinite loop when booting repl (loading nix environment)
Expected behavior:
Repl starts as normal, allowing code to be edited and run.
Actual behavior:
Repl alternates between “booting repl”, “connected”, and “reconnecting”, in an infinite loop. The “connected” phase lasts less than a second, while the other phases typically take 10-15 seconds.
Steps to reproduce:
Create a new nix repl (or any repl, really), and edit the replit.nix file to add/remove packages. At least on my account, this causes the infinite restart loop, even in a new, empty *nix repl.
This bug does not occur if the replit.nix file isn’t edited, suggesting it’s some issue with how packages are being loaded.
cargo fmt --all && CARGO_HOME=cargocache cargo test --all -- --nocapture --test-threads=1
Compiling libc v0.2.139
Compiling memchr v2.5.0
Compiling version_check v0.9.4
Compiling cfg-if v1.0.0
Compiling once_cell v1.17.1
Compiling regex-syntax v0.6.28
Compiling unicode-width v0.1.10
Compiling smawk v0.3.1
Compiling ahash v0.7.6
Compiling getrandom v0.2.8
Compiling aho-corasick v0.7.20
Compiling regex v1.7.1
Compiling hashbrown v0.12.3
Compiling unicode-linebreak v0.1.4
Compiling textwrap v0.16.0
Compiling fsmregex v0.1.0 (/home/runner/KYbcLzKqd1L)
warning: unused variable: `num_registers`
--> src/tests.rs:179:10
|
179 | let (num_registers, num_tags, dfa, nfa) = compiler.compile(regex);
| ^^^^^^^^^^^^^ help: if this is intentional, prefix it with an underscore: `_num_registers`
|
= note: `#[warn(unused_variables)]` on by default
warning: unused variable: `num_tags`
--> src/tests.rs:179:25
|
179 | let (num_registers, num_tags, dfa, nfa) = compiler.compile(regex);
| ^^^^^^^^ help: if this is intentional, prefix it with an underscore: `_num_tags`
warning: unused variable: `dfa`
--> src/tests.rs:179:35
|
179 | let (num_registers, num_tags, dfa, nfa) = compiler.compile(regex);
| ^^^ help: if this is intentional, prefix it with an underscore: `_dfa`
warning: unused variable: `nfa`
--> src/tests.rs:179:40
|
179 | let (num_registers, num_tags, dfa, nfa) = compiler.compile(regex);
| ^^^ help: if this is intentional, prefix it with an underscore: `_nfa`
warning: `fsmregex` (lib test) generated 4 warnings
Finished test [unoptimized + debuginfo] target(s) in 2m 02s
Running unittests (target/debug/deps/fsmregex-62ee0ade41e0bacc)
running 15 tests
test charset::tests::test_disjoint ... ok
test charset::tests::test_intersection ... ok
test charset::tests::test_minus ... ok
test parser_tests::test_balanced ... ok
test parser_tests::test_escapes ... ok
test parser_tests::test_parse ... Some(
Tag(
Optional(
Repeat(
Chain(
[
Char(
Allow(
{
'a',
},
),
),
Char(
Allow(
{
'b',
},
),
),
Char(
Allow(
{
'c',
},
),
),
Char(
Allow(
{
'd',
},
),
),
],
),
),
),
"hi",
),
)
ok
test tests::test_alternation ... thread 'main' panicked at 'assertion failed: `(left == right)`
left: `None`,
right: `Some((2, []))`', src/tests.rs:47:5
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
FAILED
test tests::test_asdf ... ok
test tests::test_chain ... thread 'main' panicked at 'assertion failed: `(left == right)`
left: `None`,
right: `Some((17, []))`', src/tests.rs:74:5
FAILED
test tests::test_quoted_string ... ok
test tests::test_string_match_casesensitive ... ok
test tests::test_string_match_ignorecase ... ok
test tests::test_tag_repeating_complex ... ok
test tests::test_tag_repeating_simple ... ok
test tests::test_tag_simple ... ok
failures:
failures:
tests::test_alternation
tests::test_chain
test result: FAILED. 13 passed; 2 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.14s
error: test failed, to rerun pass '--lib'
exit status 101
(It’s very long.) I don’t see any Nix error, it seems this error is to do with your code… That said, I don’t know if you can see any Nix errors without actually being in the editor, but the fact is I could run the Repl.
I suppose I could try re-making the repl with the errors fixed, and that would probably eliminate the issue. However, I believe that it’s still a replit bug, as a code error shouldn’t make it impossible to modify files within the repl.
I fixed all the errors and uploaded it to a new repl, but the moment I paste in my replit.nix file, the restart loop occurs again.
Reverting to the old, default replit.nix file does nothing, presumably cuz it’s not saving due to the restart loop.
Well, the issue appears to have resolved itself. IDK was up with it yesterday, but even so much as creating a new rust repl and editing the replit.nix file would reproduce the error, but today, I can re-open all repls perfectly fine.
But to answer your question, at least yesterday, forking did not do anything. The solution that did work for me (even yesterday) was to create a brand new *nix repl (NOT a rust repl, because editing the replit.nix file would cause issues there), and then uploading my code files one-by-one onto the *nix repl.
But just to confirm, it’s all good now - even the original rust repl is OK. IDK what exactly caused it but
