Vladimir Uogov

1) I am using a sync client to publish/subscribe messages on Zenoh 2) Handling messages and establish connections/push/close happens in pre-spawned threads from me code. But there is only one...

2. I've tried to spawn 1,2,3,4,10 http threads and result is the same. There is a single thread on sender that shall consume a single http thread. 3. Agree 4....

Here is common situation with CPU/Memory and other utilization 

I am running my app with extended backtrace. I will have some results in an hour or two.

``` [2024-02-03T00:18:06Z INFO zbus::cmd::zbus_export_stream] Request is served by thread #0 [2024-02-03T00:18:06Z INFO zenoh::net::runtime] Using PID: 706b64c5d3a618ca4ea57962e9558259 [2024-02-03T00:18:06Z INFO zenoh::net::runtime::orchestrator] Zenoh can be reached at: tcp/[fe80::1]:64017 [2024-02-03T00:18:06Z INFO zenoh::net::runtime::orchestrator] Zenoh can...

Each pre-spaned thread performing this: 1. Establish connection with Zenoh 2. Parse received JSON's 3. Publish succesfully parsed JSON 4. Close connection with Zenoh If I am to exhaust any...

Sure, I exhaust the RAM, but I do have swapins and swapouts going. Even if that, shall app crash in library code if something like this occurs ?

One of my points, while I do exhaust the RAM, I do not exhaust VM, so as far as I concern, I do have resources. Although I will eventually get...