API rate limits
For uninterrupted access to BILL API services, follow the API rate limit rules and use exponential backoff to space out your API requests.
What is an API rate limit?
An API rate limit is the number of API requests your solution or users can make within a given time period.
The BILL API rate limit is calculated as the number of API requests per developer key per hour. In addition, the API concurrent rate limit is calculated as the number of simultaneous API requests per developer key per organization.
BILL API rate limits
The rate limit for BILL API requests per developer key per hour is 20000
. In addition, the rate limit for /Login.json
and /ListOrgs.json
API requests per developer key per hour is 200
.
When you reach this limit, you receive the BDC_1144
error. All subsequent API requests must wait until the beginning of the next hour.
{
"response_status" : 1,
"response_message" : "Error",
"response_data" : {
"error_code" : "BDC_1144",
"error_message" : "Max number of allowed requests per hour reached: 20000."
}
}
BILL API concurrent rate limits
The concurrent rate limit for BILL API requests per developer key per organization is 3
.
When you reach this limit, you receive the BDC_1322
error. All subsequent API requests fail until one concurrent request is completed.
{
"response_status" : 1,
"response_message" : "Error",
"response_data" : {
"error_code" : "BDC_1322",
"error_message" : "Max number of concurrent requests per organization reached."
}
}
Implementing exponential backoff
Implement exponential backoff in your code and appropriately space out your API requests to avoid hitting the BILL rate limits and concurrent rate limits.
In this Python example, when you hit the BILL rate limits and receive the BDC_1144
error, the time.sleep()
function is used to set a backoff in seconds before retrying the API request.
The delay in seconds is calculated as 2 ^ (number of unsuccessful retries)
.
NOTE
In the example,
max_retry
is set as12
for the maximum possible backoff time in seconds to go past one hour. For BILL rate limits, this logic ensures that all subsequent API requests wait until the beginning of the next hour.
import time, requests, json
headers = {
"Accept": "application/json",
"Content-Type": "application/x-www-form-urlencoded"
}
sessionId = "set_your_session_id"
devKey = "set_your_developer_key"
vendorId = "set_your_vendor_id"
max_retry = 12
retry = 0
while retry < max_retry:
# Make a BILL API call (Read a vendor)
url = 'https://api-stage.bill.com/v2/Crud/Read/Vendor.json'
data = {
"sessionId": sessionId,
"devKey": devKey,
"data": "{\"id\":\"" + vendorId + "\"}"
}
result = requests.post(url, headers=headers, data=data)
parse_result = json.loads(result)
# If not rate limited, break & continue with your code logic
if parse_result['response_data']['error_code'] != "BDC_1144":
break
else:
# If rate limited, add exponential backoff
time.sleep(2 ** retry)
retry = retry + 1
Updated 4 months ago