Polish docs & messages

CLOUD.md

@@ -16,22 +16,22 @@ az account subscription list
 
 Create a dedicated Resource Group. See [documentation][az_manage_rg].
 For example, consider `yascheduler-rg` in `westeurope` location.
-Write down the resource group and location to the config.
+Save the resource group and location to the cloud config.
 
 ```bash
 az group create -l westeurope -g yascheduler-rg
 ```
 
 Create a dedicated _Enterprise Application_ for service.
 See [documentation][az_app_create].
-Write down `appId` as `client_id` to config.
+Save `appId` as `az_client_id` to the cloud config.
 
 ```bash
 az ad app create --display-name yascheduler
 ```
 
 Assign roles _Network Contributor_ and _Virtual Machine Contributor_
-in the _Resource Group_. Use the correct `appId`.
+in the _Resource Group_. Use the correct `appId`:
 
 ```bash
 az role assignment create \
@@ -45,14 +45,15 @@ az role assignment create \
 ```
 
 Create an _Application Registration_.
-Add _Client Secret_ to the Application Registration. Use correct `appId`.
-Write down `tenant` as `tenant_id` and `password` as `client_secret`.
+Add the _Client Secret_ to this Application Registration. Use the correct `appId`:
 
 ```bash
 az ad app credential reset --id 00000000-0000-0000-0000-000000000000 --append
 ```
 
-Create virtual networks.
+Use `tenant` as the `az_tenant_id` and `password` as the `az_client_secret` cloud settings.
+
+Create virtual networks:
 
 ```bash
 az network nsg create \
@@ -71,14 +72,16 @@ az network vnet create \
     --subnet-prefix 10.0.0.0/22
 ```
 
-It is possible to setup a _jump host_. It allows connections from the outside.
-If `yascheduler` is installed in the internal network, it is optional.
+According to our experience, while creating the nodes, the Azure allocates the new public
+IP-addresses slowly and unwillingly, so we support **the internal IP-addresses** only.
+This is no problem, if `yascheduler` is installed in the internal network.
+If this is not the case, one has to setup a _jump host_, allowing connections from the outside:
 
 ```bash
 az vm create \
     -g yascheduler-rg -l westeurope \
     --name yascheduler-jump-host \
-    --image Debian \
+    --image Debian11 \
     --size Standard_B1s \
     --nsg yascheduler-nsg \
     --public-ip-address yascheduler-jump-host-ip \
@@ -90,7 +93,7 @@ az vm create \
     --ssh-key-values "$(ssh-keygen -y -f path/to/private/key)"
 ```
 
-Write down `publicIpAddress` as `jump_host`. `jump_user` will be `yascheduler`.
+Save the `publicIpAddress` as `az_jump_host`, and `az_jump_user` will be `yascheduler`.
 
 [az_cli_install]: https://docs.microsoft.com/en-us/cli/azure/install-azure-cli
 [az_manage_rg]: https://docs.microsoft.com/en-us/cli/azure/manage-azure-groups-azure-cli

README.md

@@ -47,6 +47,20 @@ result = yac.queue_submit_task(
 print(result)
 ```
 
+Or run directly in console with `yascheduler` (use a key `-l DEBUG` to change the log level).
+
+_Supervisor_ config reads e.g.:
+
+```
+[program:scheduler]
+command=/usr/local/bin/yascheduler
+user=root
+autostart=true
+autorestart=true
+stderr_logfile=/data/yascheduler.log
+stdout_logfile=/data/yascheduler.log
+```
+
 File paths can be set using the environment variables:
 
 - `YASCHEDULER_CONF_PATH`
@@ -104,7 +118,7 @@ Connection to a PostgreSQL database.
   Path to root directory of local data files.
   Can be relative to the current working directory.
 
-  _Default_: `./data`
+  _Default_: `./data` (but it's always a good idea to set up explicitly!)
 
   _Example_: `/srv/yadata`
 
@@ -362,7 +376,7 @@ Settings prefix is `upcloud`.
 
   Password.
 
-#### Engines `[engine.*]`
+### Engines `[engine.*]`
 
 Supported engines should be defined in the section(s) `[engine.name]`.
 The name is alphanumeric string to represent the real engine name.

yascheduler/clouds/cloud_api_manager.py

@@ -129,8 +129,9 @@ async def select_best_provider(
 
         self.log.debug("Used providers: %s", used_providers)
         if not suitable_providers:
-            self.log.debug("No suitable cloud provides")
+            self.log.debug("No suitable cloud providers")
             return
+
         ok_apis = filter(lambda x: x.name in suitable_providers, self.apis.values())
         ok_apis_sorted = sorted(ok_apis, key=lambda x: x.config.priority, reverse=True)
         api = ok_apis_sorted[0]

yascheduler/scheduler.py

@@ -497,7 +497,7 @@ async def allocator_producer(
         tasks = await self.db.get_tasks_by_status((TaskStatus.TO_DO,), tlim)
         if tasks:
             ids = [str(t.task_id) for t in tasks]
-            self.log.debug("Want allocate tasks: %s" % ", ".join(ids))
+            self.log.debug("Want to allocate tasks: %s" % ", ".join(ids))
         for task in tasks:
             yield UMessage(task.task_id, task)