Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I’ve done this with retry logic in each concurrent writer (retry if you can’t lock the file with a short timeout) and it works fine enough.


If threads have to contend for db with retry loops, does concurrency actually buy you anything?


It definitely doesn't buy you performance. But I thought you were asking if it's possible, or how difficult it is. A use case might be if you have an existing architecture of N processes and want to collect stuff from them in a central file.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: