View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0008954 | Taler | release | public | 2024-06-16 16:50 | 2026-09-19 14:43 |
| Reporter | Assigned To | Florian Dold | Resolution deadline | ||
| Implementer | Reviewer | ||||
| Release signed by | Deployed by | ||||
| Verifier | Communicator | ||||
| Data protection authority notification required | No | ||||
| CSIRT notification required | No | ||||
| Estimated effort (hours) | Actual effort (hours) | 0 | |||
| Fast-track | No | ||||
| Work package | none | ||||
| Priority | low | Severity | feature | Reproducibility | have not tried |
| Status | closed | Resolution | won't fix | ||
| Target Version | 1.6 | Fixed in Version | 1.6 | ||
| Summary | 0008954: Non-systemd reliance | ||||
| Description | For platforms like Alpine/Gentoo (using OpenRC), or Docker/Podman (where running systemd is generally considered bad practice & should be avoided d), we need to [be able to] handle: - Automatically Restarting of Process - Shared Socket Access without Systemd. An example architecture for this would be to have 2 layers; a shell script that simply handles forking & restarting(1) the processes, and a Rust/C/Go wrapper that creates said socket before spawning the bash process (2, as bash cannot start a socket it owns on it's own). | ||||
| Additional Information | (1) This part would be composed of 2 parts: a function that checks if the process dies (and how fast it does so), and restarts it either immediately or after some delay, and a loop that kills the process on an interval (& lets the restart handler start it again) (2) This part would simply open a socket if the config tells it that it's a Taler component that needs it, then spawn part 1. This part is needed as to ensure that the bash script - and thereby also it's children - have access to the socket. Signals should be passed 1:1 to the child bash script, and if the child dies (inclusive as a result of a signal we send it), we should exit with the exact same code. The user should not need to know that the process they spawned internally runs a bash script, they should be able to send signals to the rust process and get the outcome they expect. ^ Based on a discussion with Christian | ||||
| Tags | No tags attached. | ||||
| We currently have no good use case for this, and we don't want to maintain component / feature we don't have a good use case for. | |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2024-06-16 16:50 | New Issue | ||
| 2024-06-16 16:50 | Status | new => assigned | |
| 2024-06-16 16:50 | Assigned To | => nullptrderef | |
| 2024-06-16 16:52 | Description Updated | ||
| 2024-06-16 16:52 | Additional Information Updated | ||
| 2024-06-16 16:57 | Description Updated | ||
| 2024-06-17 08:33 | Christian Grothoff | Target Version | git (master) => post-1.0 |
| 2024-06-23 16:54 | Description Updated | ||
| 2024-07-28 19:49 | Christian Grothoff | Severity | minor => feature |
| 2025-04-17 21:31 | Christian Grothoff | Assigned To | nullptrderef => |
| 2025-04-17 21:31 | Christian Grothoff | Status | assigned => implementing |
| 2026-01-25 16:33 | Christian Grothoff | Target Version | post-1.0 => post-2.0 |
| 2026-08-22 17:42 | Florian Dold | Assigned To | => Florian Dold |
| 2026-08-22 17:42 | Florian Dold | Status | implementing => closed |
| 2026-08-22 17:42 | Florian Dold | Resolution | open => won't fix |
| 2026-08-22 17:42 | Florian Dold | Note Added: 0029553 | |
| 2026-09-19 14:43 | Christian Grothoff | Fixed in Version | => 1.6 |
| 2026-09-19 14:43 | Christian Grothoff | Target Version | post-2.0 => 1.6 |