Emmanuel Bosquet
Emmanuel Bosquet
This crate fits all my needs up to a feature I'd like to see: ```rust #[derive(Envconfig)] struct Config { #[envconfig(from = "POSTGRESQL_URL", or_from = "POSTGRES_URL")] pub postgresql_url: String, } ```...
For now, querying a cluster's metrics by doing: ``` sozu --config config.toml query metrics -k MyCluster ``` does not return anything (this is addressed in #789 ), but once this...
The pull request #783 restores the querying of proxy metrics (`accept_queue`, `http_responses`, `event_loop_time`…), but leaves this problems: - Querying cluster metrics returns nothing. - Querying a list of metric names...
For now, when we query metrics… nothing appears. Also metrics enabling and disabling doesn't work. Let's fix it. - [x] fix "metrics enable" and disable and clear - [x] fix...
Here is an example of metrics formatting: ``` Worker 0 ========= +--------------------+-------+-------+ | | gauge | count | +--------------------+-------+-------+ | accept_queue.count | 0 | | +--------------------+-------+-------+ | client.connections | 0...
A number of issues for 0.14 are linked to inaproppriate behaviour of sozuctl: - #718 - #708 - #631 Before fully addressing those issues, we need to address the timeout...
While refactoring the CommandServer, I noticed that, at least since #724, the CLI reads only ONE message and then closes. This is problematic because : 1. the timeout doesn't really...
Fixes #353 We want to create a web component that sends a "ping all VMs" message to the CC Admin (in fact, ping the action dispatcher instances, there is one...
For a better worker management in the main process, it would be handy to have a `worker_id` in the `ProxyResponseData` sent by a worker (may be renamed as `ProxyResponseContent` by...
We should add those metrics within a proxy instance. Within the `frontend_from_request` function, we should check for possible optimization on the lookup time. Possibly too: does the regex matching time...