Conversation
fabriziodemaria
left a comment
fabriziodemaria
left a comment
There was a problem hiding this comment.
It's a bit outside my realm of expertise, but it looks good to me. I was mostly focused on cases of irrecoverable panics, ensuring there is minimal impact on the hosting application in that case
| private reloadInstance(error:unknown) { | ||
| logger.error('Failure calling into wasm:', error); | ||
| try { | ||
| this.bufferedLogs.push(this.delegate.flushLogs()); |
There was a problem hiding this comment.
What happens with unrecoverable errors? I am thinking if the wasm doesn't restart and we keep pushing data to the bufferedLogs we might have a memory leak, perhaps cap the bufferedLogs
There was a problem hiding this comment.
The only thing that can produce logs is a successful resolve. And what's happening here is that we try once to extract logs from a broken instance before we throw it away. I don't think it's a problem from a memory leak perspective, i.e. this can't cause logs to grow faster than what successful operation does.
andreas-karlsson
force-pushed
the
handle-panics
branch
from
6bc951b to
c82a2bd
Compare
3 participants
No description provided.