Phritz

Results 18 issues of Phritz

It defaults to 10s: https://github.com/rocicorp/repc/blob/ff452fd4a4a733b2bed623dad483d6d7716fc45c/src/fetch/mod.rs#L1

https://rocicorp.slack.com/archives/C01JJGGS6CU/p1621426062298400 ``` UnhandledRejection Non-Error promise rejection captured with value: DBReadError(MapLoadError(CorruptChunk(Corrupt(“missing key”)))) Pull returned: PullFailed(FetchFailed(RequestTimeout(TimeoutError { _private: () }))) logger: console arguments: [“Pull returned: PullFailed(FetchFailed(RequestTimeout(TimeoutError { _private: () })))“] ```

In https://github.com/rocicorp/repc/issues/351 we can see a DbReadError and a PullFailed error. These are variants of OpenTransactionError and BeginTryPullError respectively. However when printed the name of the enum (OTE and BTPE)...

It's easy to exceed eg a 1MB request body in push if there are lots of mutations (eg, mouse movements) and the server is slow or unresponsive for a while....

Proposal: https://www.notion.so/Differential-Client-View-12be3a636c9f404b88d49ecbd100a694 TODOs in repc: - [x] remove checksum from pull and checksum itself. @phritz - it is a question if we do this in a breaking vs nonbreaking way...

aaron ran into this and we discussed it in yesterday's sync. even if we are moving away from string patch to JSON which would eliminate the possibility of non-json in...

1. wasm bench - doesn't compile eg `./tool/benchmark |tee > wasm-bench.txt` - when it fails in CI it still is shown as green. go to https://github.com/rocicorp/repc/runs/1974200630?check_suite_focus=true, click wasm size, click...

this has come up several times and seems totally natural to have a keyspace of some sort that is not synced in other contexts eg https://github.com/rocicorp/replicache/issues/79 we have discussed having...

this is an indexing followup https://github.com/rocicorp/repc/issues/218#issuecomment-726253248

in https://github.com/rocicorp/replicache/issues/30#issue-599410531 i proposed some invariants that we should check during sync to reduce the chances that we have a bug / data corruption. i had planned to do that...