PRO API (Endpoint #3, Asia) DOWN - Work around options?

Engine 2 Asia is down and I am looking for work around options.

Engine 1 is not a lot of use for me as I am reading numbers and mostly engine 1 gets them wrong or can not read them at all.

I get a 30 second timeout for OCR requests. Is there any point in extending that time since the server is reported as down (OCR.Space API Status) ? If so how?

Am I able to access an Engine 2 server at a different location that is not down until the problem is resolved?

Any other suggested solutions?

Hello, Engine2 in Asia is up again.

Yes! In the future, if e. g. the Asia endpoint is down again, we recommend that you (automatically) switch over to the EU or US endpoint. Of course, our goal is that all endpoints are up all the time.

Sounds ideal. It did not appear to fail to an alternate server this time. It appeared to try the same server twice? Is that something I can resolve?

2022-08-16T06:36:17.120Z - [info] Executing: | store | true | !ErrorIgnore |
2022-08-16T06:36:17.450Z - [info] Executing: | store | true | !OCRScale |
2022-08-16T06:36:17.777Z - [info] Executing: | OCRExtractRelative | AlphaSOC_dpi_144.png | SOC |
2022-08-16T06:36:21.279Z - [info] OCR (eng) started (1.7 KB)
2022-08-16T06:36:51.297Z - [warning] Error in OCR endpoint 3 after 30.0s: timeout - trying next.
2022-08-16T06:37:21.311Z - [error][ignored] OCR timeout
2022-08-16T06:37:21.366Z - [warning] Error in OCR endpoint 3 after 30.0s: timeout - trying next.
2022-08-16T06:37:21.651Z - [info] Executing: | echo | ${SOC} | |
2022-08-16T06:37:21.663Z - [echo] 0
2022-08-16T06:37:22.007Z - [info] Executing: | store | true | !StatusOK |
2022-08-16T06:37:22.356Z - [info] Executing: | store | false | !ErrorIgnore |


Oh, that is interesting. It should have picked a different endpoint. See here my test (I tested by disconnecting my PC from the Internet):

But if you are on Windows and Mac: Have you tried the new offline RPA OCR engine? It works fast and requires no Internet connection. Or is its OCR quality not good enough for you?

Thanks. Interestingly I am getting a similar result to you now. Let’s see what happens if the endpoint 3 server goes down again one day.