App Update
update volume
If your app is deployed in EverAI platform, when you want to update your files in volume, you can run everai app update. EverAI platform support app hot upgrade, and your application service will always be running online during the entire upgrade and update process.
The following example is based on the app in the quickstart.
Run everai worker list, you can see a worker in running. Note that CREATED_AT and DELETED_AT use UTC time display.
ID STATUS DETAIL_STATUS CREATED_AT DELETED_AT
---------------------- -------- --------------- ------------------------ ------------
kEEBkaoEaZrxPgzab2ChjQ RUNNING FREE 2024-07-05T08:41:49+0000
Use curl to run the test case, the output of the example is on the terminal.
curl -H'Authorization: Bearer <your_token>' https://everai.expvent.com/api/routes/v1/<your namespace>/<your app name>/show-volume
hello world
hello world
Run everai volume get get-start-volume, you can find the path of the volume that is used in your app.
<Volume: id: R9cFpk2JgmLcuiEfgAxPRd, name: get-start-volume, revision: 000001-08a, files: 1, size: 11 B>
path: /Users/<username>/.cache/everai/volumes/R9cFpk2JgmLcuiEfgAxPRd
Enter the directory of the volume, and find the file named my-model, then update the file content. In this example, the new file content like this:
hello world
hello world
hello world
hello world
Return to your app's directory, run everai app prepare to push the new file into the cloud.
Run the following command to finish app update.
everai app update
Run everai worker list -a, you can see the original worker is still running. And the new worker is deploying now.
ID STATUS DETAIL_STATUS CREATED_AT DELETED_AT
---------------------- ---------- --------------- ------------------------ ------------------------
kEEBkaoEaZrxPgzab2ChjQ RUNNING FREE 2024-07-05T08:41:49+0000
mNTAJoqyMRHDyDoLTCSjPn CREATED IN_FLIGHT 2024-07-05T13:11:59+0000
The original worker is terminated still the new worker is deployed. Run everai worker list -a, you can see is new worker is running.
ID STATUS DETAIL_STATUS CREATED_AT DELETED_AT
---------------------- ---------- --------------- ------------------------ ------------------------
kEEBkaoEaZrxPgzab2ChjQ TERMINATED 2024-07-05T08:41:49+0000 2024-07-05T13:12:23+0000
mNTAJoqyMRHDyDoLTCSjPn RUNNING FREE 2024-07-05T13:11:59+0000
Use curl to run the test case agian, the new output of the example is on the terminal now.
curl -H'Authorization: Bearer <your_token>' https://everai.expvent.com/api/routes/v1/<your namespace>/<your app name>/show-volume
hello world
hello world
hello world
hello world
update image
If your app is deployed in EverAI platform, when you want to update your code, you can run everai app update. EverAI platform supports app hot upgrade, and your application service will always be running online during the entire upgrade and update process.
The following example is based on the app in the quickstart.
Run everai worker list, you can see a worker in running. Note that CREATED_AT and DELETED_AT use UTC time display.
ID STATUS DETAIL_STATUS CREATED_AT DELETED_AT
---------------------- -------- --------------- ------------------------ ------------
PNLENbRP7AabNy6ZQWTogb RUNNING FREE 2024-07-05T06:45:19+0000
Use curl to run the test case, the output of the example is on the terminal. This example sends active messages from the server to the client.
curl -H'Authorization: Bearer <your_token>' https://everai.expvent.com/api/routes/v1/<your namespace>/<your app name>/sse
hello again 0
hello again 1
hello again 2
In this step, Open app.py, and change code from hello again to hello world again. Then you run everai app run to check your code locally.
import flask
@app.service.route('/sse', methods=['GET'])
def sse():
def generator():
for i in range(10):
yield f"hello world again {i}\n\n"
time.sleep(1)
return flask.Response(generator(), mimetype='text/event-stream')
And then open image_builder.py, update your image registry's version. Run everai image build to buld image and push the image to quay.io.
from everai.image import Builder
IMAGE = 'quay.io/<username>/<repo>:<tag>'
Aftering build image, you can run the following command to update your app.
everai app update
Run everai worker list -a, you can see the new worker's detail status is INITIALIZED.
ID STATUS DETAIL_STATUS CREATED_AT DELETED_AT
---------------------- ----------- --------------- ------------------------ ------------
PNLENbRP7AabNy6ZQWTogb RUNNING FREE 2024-07-05T06:45:19+0000
kEEBkaoEaZrxPgzab2ChjQ INITIALIZED IN_FLIGHT 2024-07-05T08:41:49+0000
Until the new worker is in the RUNNING state, the original worker is already in the TERMINATED state.
ID STATUS DETAIL_STATUS CREATED_AT DELETED_AT
---------------------- ---------- --------------- ------------------------ ------------------------
PNLENbRP7AabNy6ZQWTogb TERMINATED 2024-07-05T06:45:19+0000 2024-07-05T08:42:28+0000
kEEBkaoEaZrxPgzab2ChjQ RUNNING FREE 2024-07-05T08:41:49+0000
Use curl to run the test case again, the new output of the example is on the terminal. This example sends new active messages from the server to the client.
curl -H'Authorization: Bearer <your_token>' https://everai.expvent.com/api/routes/v1/<your namespace>/<your app name>/sse
hello world again 0
hello world again 1
hello world again 2
hello world again 3
update resource
If your application has been deployed on the EverAI platform cloud, you can use everai app update when you need to adjust your computing resources (CPU, memory, number of GPU cards or GPU model). The EverAI platform supports hot application upgrades, and your application services are always running online during the entire upgrade and update process.
Open the app.py file, and in the code that creates and defines an app instance through the App class, modify the resource_requests parameter:
resource_requests=ResourceRequests(
cpu_num=2,
memory_mb=20480,
gpu_num=1,
gpu_constraints=[
"A100 40G",
"RTX 4090"
],
)
After saving and exiting the app.py file, execute the following command to complete the adjustment of application computing resources.
everai app update
Run everai worker list -a, you can see the original worker's detail status is TERMINATED. And the new worker is running now.