Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Do not untrap exits if they were already trapped before callingfix/do-not-untrap-trapped-processes | rinpatch | 2020-09-05 | 1 | -7/+10 |
| | |||||
* | Merge branch 'fix/trap-exits' into 'master' | href | 2020-09-05 | 1 | -4/+29 |
|\ | | | | | | | | | Fix counters not decrementing when the process is exited See merge request pleroma/elixir-libraries/concurrent_limiter!1 | ||||
| * | Fix counters not decrementing when the process is exited | rinpatch | 2020-09-05 | 1 | -4/+29 |
|/ | | | | | | | | | | | | | This fixes the issue by trapping exits for the duration of the fun, then turning exit trapping off and killing the process if there is an exit message in the mailbox. The real-world case where this fixes things is Pleroma MediaProxy. Without exit trapping, the process would get killed if the connection was closed client-side and a counter would be left incremented. I am not sure if trapping exits is the optimal solution, but I don't see any other option. | ||||
* | Fix decrement after max_retries been reached + telemetry events | Jordan Bracco | 2020-07-07 | 4 | -0/+18 |
| | |||||
* | 0.1.0 (for realz) | Jordan Bracco | 2020-05-16 | 1 | -0/+1 |
| | |||||
* | 0.1.0.0.1.0 | Jordan Bracco | 2020-05-16 | 1 | -2/+2 |
| | |||||
* | Fix tests & add ConcurrentLimiter.delete/1 | Jordan Bracco | 2020-05-16 | 2 | -11/+22 |
| | |||||
* | Format and prepare for release | Jordan Bracco | 2020-05-16 | 11 | -103/+527 |
| | |||||
* | Remove ETS backend, add max_retries, add options to limit/3. | Jordan Bracco | 2020-05-14 | 1 | -70/+34 |
| | |||||
* | Rename to ConcurrentLimiter | Jordan Bracco | 2020-05-14 | 8 | -111/+132 |
| | |||||
* | minor fixes | Jordan Bracco | 2020-05-08 | 1 | -27/+19 |
| | |||||
* | Switch to atomics, add shared ets, .. | Jordan Bracco | 2020-05-08 | 9 | -61/+337 |
| | |||||
* | Initial commit | Jordan Bracco | 2020-05-08 | 8 | -0/+180 |