Re: [pbs-devel] [PATCH proxmox 2/2] rest-server: close race window when updating worker task count

[Thomas Lamprecht <t.lamprecht@proxmox.com>]

Am 29.11.24 um 14:13 schrieb Fabian Grünbichler:
> this mimics how the count is updated when spawning a new task - the lock scope
> needs to cover the count update itself, else there's a race when multiple
> worker's log their result at the same time..
> 
> Co-developed-by: Dominik Csapak <d.csapak@proxmox.com>
> Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
> ---
>  proxmox-rest-server/src/worker_task.rs | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/proxmox-rest-server/src/worker_task.rs b/proxmox-rest-server/src/worker_task.rs
> index 3ca93965..018d18c0 100644
> --- a/proxmox-rest-server/src/worker_task.rs
> +++ b/proxmox-rest-server/src/worker_task.rs
> @@ -1023,7 +1023,8 @@ impl WorkerTask {
>  
>          WORKER_TASK_LIST.lock().unwrap().remove(&self.upid.task_id);
>          let _ = self.setup.update_active_workers(None);
> -        set_worker_count(WORKER_TASK_LIST.lock().unwrap().len());
> +        let lock = WORKER_TASK_LIST.lock().unwrap();

why not use this also for the remove operation above? I.e. something like:

let locked_worker_tasks = WORKER_TASK_LIST.lock().unwrap();

locked_worker_tasks.remove(&self.upid.task_id);

set_worker_count(locked_worker_tasks.len())

If there are technical reason speaking against this, which I hope not, then a
comment would be definitively warranted, otherwise using a single lock would
IMO make this a bit clearer and locking twice isn't exactly cheaper.

Looks OK besides that, but would still want to take a closer look.

> +        set_worker_count(lock.len());
>      }
>  
>      /// Log a message.



_______________________________________________
pbs-devel mailing list
pbs-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel