OneSpan Sign Callback failed
Thursday, April 29, 2021 at 04:23pmHello,
We have an azure function setup to listen to onespan callbacks which works great nearly all the time, but on the rare occasion we get a failed callback. I supposed there is a number of reasons that could happen. In this specific case the error indicates a timeout exception.
I have a couple of questions
1 - When this exception occurs it appears that the onespan package sender/creator is notified via an email. here is a sample of the email
"
From: OneSpan Sign <[email protected]>
Sent: April 01, 2021 1:23 PM
To: Sender Name <senderEmailhere>
Subject: OneSpan Sign Callback failed
Hi Sender Name,
Failed to communicate with the callback server.
URL: https://myazurefxname.azurewebsites.net/api/webhookendpoint<https://urldefense.com/v3/__https:/myazureFxName.azurewebsites.net/api/OSSWebhook__;!!---something
Payload: {"@class":"com.silanis.esl.packages.event.ESLProcessEvent","name":"PACKAGE_ACTIVATE","sessionUser":"sessionUserId","packageId":"FailedPackageId","message":null,"documentId":null,"createdDate":"2021-00-00T20:45:22.316Z"}
Reason(s): java.util.concurrent.TimeoutException
"
This is not the most helpful way for us to catch these exceptions so I was wondering if there is a way to set that exception email address explicitly? maybe to someone in a support group? so that all the onespan callback exceptions go to [email protected]?
2 - Since there's always some small chance that the callback will fail does Onespan keep track of these failed callbacks in a way we can query for them using the onespan api. I didn't see anything but just thought I would ask.
Thank you
Chris
Reply to: OneSpan Sign Callback failed
Thursday, April 29, 2021 at 04:36pmHi Chris,
Thanks for your post and great to hear that the callback listener works nearly all the time for you! For your two questions
(1)This email template is called "email.callbackFailure", by default its recipient points to the package owner. You can definitely customize it by contacting our support team ([email protected]).
(2)This is temporary not supported, but there are already similar enhancement ideas being raised, like to "retry and send email after connector callback fails" or "store callback failure alerts". If you preferred, I can definitely vote on behalf of your organization for these requests.
Duo
Reply to: Hi Chris, Thanks for…
Thursday, April 29, 2021 at 04:59pmThanks for the super quick reply :)
Good to know it's relatively easy to change that recipient email on the failed callback.
Both features seem like good ideas, I have a feeling that the failure on our end for the callback receiver is very intermittent. I looked into our logs and we have the webhooks working all around that time of the reported failure with maybe less than a minute gap exactly at the time that callback failed so who knows exactly what happened. Some kind of exponential back off probably would have worked in that case, but also in general, having a repository to query these failed callbacks would also be potentially really useful.
Cheers,
Chris