mirror of https://github.com/coder/coder.git
5.5 KiB
5.5 KiB
Usage: coder scaletest create-workspaces [flags]
Creates many users, then creates a workspace for each user and waits for them
finish building and fully come online. Optionally runs a command inside each
workspace, and connects to the workspace over WireGuard.
It is recommended that all rate limits are disabled on the server before running this scaletest. This test generates many login events which will be rate limited against the (most likely single) IP.
Options
--cleanup-concurrency int, $CODER_SCALETEST_CLEANUP_CONCURRENCY (default: 1)
Number of concurrent cleanup jobs to run. 0 means unlimited.
--cleanup-job-timeout duration, $CODER_SCALETEST_CLEANUP_JOB_TIMEOUT (default: 5m)
Timeout per job. Jobs may take longer to complete under higher
concurrency limits.
--cleanup-timeout duration, $CODER_SCALETEST_CLEANUP_TIMEOUT (default: 30m)
Timeout for the entire cleanup run. 0 means unlimited.
--concurrency int, $CODER_SCALETEST_CONCURRENCY (default: 1)
Number of concurrent jobs to run. 0 means unlimited.
--connect-hold duration, $CODER_SCALETEST_CONNECT_HOLD (default: 30s)
How long to hold the WireGuard connection open for.
--connect-interval duration, $CODER_SCALETEST_CONNECT_INTERVAL (default: 1s)
How long to wait between making requests to the --connect-url once the
connection is established.
--connect-mode derp|direct, $CODER_SCALETEST_CONNECT_MODE (default: derp)
Mode to use for connecting to the workspace.
--connect-timeout duration, $CODER_SCALETEST_CONNECT_TIMEOUT (default: 5s)
Timeout for each request to the --connect-url.
--connect-url string, $CODER_SCALETEST_CONNECT_URL
URL to connect to inside the the workspace over WireGuard. If not
specified, no connections will be made over WireGuard.
-c, --count int, $CODER_SCALETEST_COUNT (default: 1)
Required: Number of workspaces to create.
--job-timeout duration, $CODER_SCALETEST_JOB_TIMEOUT (default: 5m)
Timeout per job. Jobs may take longer to complete under higher
concurrency limits.
--no-cleanup bool, $CODER_SCALETEST_NO_CLEANUP
Do not clean up resources after the test completes. You can cleanup
manually using coder scaletest cleanup.
--no-plan bool, $CODER_SCALETEST_NO_PLAN
Skip the dry-run step to plan the workspace creation. This step
ensures that the given parameters are valid for the given template.
--no-wait-for-agents bool, $CODER_SCALETEST_NO_WAIT_FOR_AGENTS
Do not wait for agents to start before marking the test as succeeded.
This can be useful if you are running the test against a template that
does not start the agent quickly.
--output string-array, $CODER_SCALETEST_OUTPUTS (default: text)
Output format specs in the format "<format>[:<path>]". Not specifying
a path will default to stdout. Available formats: text, json.
--run-command string, $CODER_SCALETEST_RUN_COMMAND
Command to run inside each workspace using reconnecting-pty (i.e. web
terminal protocol). If not specified, no command will be run.
--run-expect-output string, $CODER_SCALETEST_RUN_EXPECT_OUTPUT
Expect the command to output the given string (on a single line). If
the command does not output the given string, it will be marked as
failed.
--run-expect-timeout bool, $CODER_SCALETEST_RUN_EXPECT_TIMEOUT
Expect the command to timeout. If the command does not finish within
the given --run-timeout, it will be marked as succeeded. If the
command finishes before the timeout, it will be marked as failed.
--run-log-output bool, $CODER_SCALETEST_RUN_LOG_OUTPUT
Log the output of the command to the test logs. This should be left
off unless you expect small amounts of output. Large amounts of output
will cause high memory usage.
--run-timeout duration, $CODER_SCALETEST_RUN_TIMEOUT (default: 5s)
Timeout for the command to complete.
-t, --template string, $CODER_SCALETEST_TEMPLATE
Required: Name or ID of the template to use for workspaces.
--timeout duration, $CODER_SCALETEST_TIMEOUT (default: 30m)
Timeout for the entire test run. 0 means unlimited.
--trace bool, $CODER_SCALETEST_TRACE
Whether application tracing data is collected. It exports to a backend
configured by environment variables. See:
https://github.com/open-telemetry/opentelemetry-specification/blob/main/specification/protocol/exporter.md.
--trace-coder bool, $CODER_SCALETEST_TRACE_CODER
Whether opentelemetry traces are sent to Coder. We recommend keeping
this disabled unless we advise you to enable it.
--trace-honeycomb-api-key string, $CODER_SCALETEST_TRACE_HONEYCOMB_API_KEY
Enables trace exporting to Honeycomb.io using the provided API key.
--trace-propagate bool, $CODER_SCALETEST_TRACE_PROPAGATE
Enables trace propagation to the Coder backend, which will be used to
correlate server-side spans with client-side spans. Only enable this
if the server is configured with the exact same tracing configuration
as the client.
--use-host-login bool, $CODER_SCALETEST_USE_HOST_LOGIN (default: false)
Use the use logged in on the host machine, instead of creating users.
---
Run `coder --help` for a list of global options.
Creates many users, then creates a workspace for each user and waits for them
finish building and fully come online. Optionally runs a command inside each
workspace, and connects to the workspace over WireGuard.
It is recommended that all rate limits are disabled on the server before running this scaletest. This test generates many login events which will be rate limited against the (most likely single) IP.
Options
--cleanup-concurrency int, $CODER_SCALETEST_CLEANUP_CONCURRENCY (default: 1)
Number of concurrent cleanup jobs to run. 0 means unlimited.
--cleanup-job-timeout duration, $CODER_SCALETEST_CLEANUP_JOB_TIMEOUT (default: 5m)
Timeout per job. Jobs may take longer to complete under higher
concurrency limits.
--cleanup-timeout duration, $CODER_SCALETEST_CLEANUP_TIMEOUT (default: 30m)
Timeout for the entire cleanup run. 0 means unlimited.
--concurrency int, $CODER_SCALETEST_CONCURRENCY (default: 1)
Number of concurrent jobs to run. 0 means unlimited.
--connect-hold duration, $CODER_SCALETEST_CONNECT_HOLD (default: 30s)
How long to hold the WireGuard connection open for.
--connect-interval duration, $CODER_SCALETEST_CONNECT_INTERVAL (default: 1s)
How long to wait between making requests to the --connect-url once the
connection is established.
--connect-mode derp|direct, $CODER_SCALETEST_CONNECT_MODE (default: derp)
Mode to use for connecting to the workspace.
--connect-timeout duration, $CODER_SCALETEST_CONNECT_TIMEOUT (default: 5s)
Timeout for each request to the --connect-url.
--connect-url string, $CODER_SCALETEST_CONNECT_URL
URL to connect to inside the the workspace over WireGuard. If not
specified, no connections will be made over WireGuard.
-c, --count int, $CODER_SCALETEST_COUNT (default: 1)
Required: Number of workspaces to create.
--job-timeout duration, $CODER_SCALETEST_JOB_TIMEOUT (default: 5m)
Timeout per job. Jobs may take longer to complete under higher
concurrency limits.
--no-cleanup bool, $CODER_SCALETEST_NO_CLEANUP
Do not clean up resources after the test completes. You can cleanup
manually using coder scaletest cleanup.
--no-plan bool, $CODER_SCALETEST_NO_PLAN
Skip the dry-run step to plan the workspace creation. This step
ensures that the given parameters are valid for the given template.
--no-wait-for-agents bool, $CODER_SCALETEST_NO_WAIT_FOR_AGENTS
Do not wait for agents to start before marking the test as succeeded.
This can be useful if you are running the test against a template that
does not start the agent quickly.
--output string-array, $CODER_SCALETEST_OUTPUTS (default: text)
Output format specs in the format "<format>[:<path>]". Not specifying
a path will default to stdout. Available formats: text, json.
--run-command string, $CODER_SCALETEST_RUN_COMMAND
Command to run inside each workspace using reconnecting-pty (i.e. web
terminal protocol). If not specified, no command will be run.
--run-expect-output string, $CODER_SCALETEST_RUN_EXPECT_OUTPUT
Expect the command to output the given string (on a single line). If
the command does not output the given string, it will be marked as
failed.
--run-expect-timeout bool, $CODER_SCALETEST_RUN_EXPECT_TIMEOUT
Expect the command to timeout. If the command does not finish within
the given --run-timeout, it will be marked as succeeded. If the
command finishes before the timeout, it will be marked as failed.
--run-log-output bool, $CODER_SCALETEST_RUN_LOG_OUTPUT
Log the output of the command to the test logs. This should be left
off unless you expect small amounts of output. Large amounts of output
will cause high memory usage.
--run-timeout duration, $CODER_SCALETEST_RUN_TIMEOUT (default: 5s)
Timeout for the command to complete.
-t, --template string, $CODER_SCALETEST_TEMPLATE
Required: Name or ID of the template to use for workspaces.
--timeout duration, $CODER_SCALETEST_TIMEOUT (default: 30m)
Timeout for the entire test run. 0 means unlimited.
--trace bool, $CODER_SCALETEST_TRACE
Whether application tracing data is collected. It exports to a backend
configured by environment variables. See:
https://github.com/open-telemetry/opentelemetry-specification/blob/main/specification/protocol/exporter.md.
--trace-coder bool, $CODER_SCALETEST_TRACE_CODER
Whether opentelemetry traces are sent to Coder. We recommend keeping
this disabled unless we advise you to enable it.
--trace-honeycomb-api-key string, $CODER_SCALETEST_TRACE_HONEYCOMB_API_KEY
Enables trace exporting to Honeycomb.io using the provided API key.
--trace-propagate bool, $CODER_SCALETEST_TRACE_PROPAGATE
Enables trace propagation to the Coder backend, which will be used to
correlate server-side spans with client-side spans. Only enable this
if the server is configured with the exact same tracing configuration
as the client.
--use-host-login bool, $CODER_SCALETEST_USE_HOST_LOGIN (default: false)
Use the use logged in on the host machine, instead of creating users.
---
Run `coder --help` for a list of global options.