We have apex classes through which we send emails based on user action in force.com sites. It was working great until Friday, But yesterday when one of our customer try to submit application, they were getting error, In debug it says “SendEmail failed. First exception on row 0; first error: UNKNOWN_EXCEPTION, INVALID_TYPE: sObject type 'SalesforceIqDataSource' is not supported.: []”. I am pretty sure we did not change anything related to apex class, email template, vf page etc.. Does anyone know the reason or best way to fix this?

We have a number of proceses that work by rescheduling themselves, ie chained scheduled jobs. This has worked perfectly for 3-4 years now.

Since the update this weekend, all of a sudden jobs that are scheduled to start just don't get started and it seems to be completely random. The job is definitely scheduled because I can see it in the list of scheduled jobs but the next scheduled run time comes and goes and the job never gets run.

There is nothing in the logs that I can see. This seems like a bug to me - has anyone else got any experience of this or any idea what might be the issue?

I don't even believe it is neccessarily related to the chaining as I have seen this happen when scheduling a job to start in 5 minutes time the job does not even get started the first time.

We received an email from Salesforce this morning regarding additional protection against forced logins that is being introduced in the next update.

The details in the email are not particularly detailed and we are unclear at this stage as to whether it will in fact affect us.

We have a custom scheduled job in Salesforce that generates reports on a quarterly basis. Due to various restrictions around scheduled jobs and the inability to call PageReference.getContent() in certain contexts that existed at the time this process was implemented, it works in a very round about way.

The scheduled job calls a future method that then gets an access token using a HTTP Post to login to an OAuth-Enabled connected app. The POST request includes the grant_type, username, password, client id and client secret in the body of the request in order to login to the connected app and the request is sent to https://login.salesforce.com/services/oauth2/token.

Once we have an access token we include this in authorization header and then call a custom Salesforce Webservice method that actually generates the reports.

We believe this is the only place where any functionality that might be considered a "forced login" is implemented or used in our environment. It is not clear, however, from the information available at the moment whether or not what we are doing counts as a forced login and will not work after the update.

It is also not clear if it does what changes we would need to make in order to ensure this still works after the update.

Does anyone have any more information about this change and whether it is likely to break the way this process works? If so can anyone recommend a better way of doing this? I think I remember reading that many of the restrictions around the usage of PageReference.getContent() have been removed in the past year or two so perhaps it is the case that all this additional complexity is no longer required.

We have apex classes through which we send emails based on user action in force.com sites. It was working great until Friday, But yesterday when one of our customer try to submit application, they were getting error, In debug it says “SendEmail failed. First exception on row 0; first error: UNKNOWN_EXCEPTION, INVALID_TYPE: sObject type 'SalesforceIqDataSource' is not supported.: []”. I am pretty sure we did not change anything related to apex class, email template, vf page etc.. Does anyone know the reason or best way to fix this?

We have a number of proceses that work by rescheduling themselves, ie chained scheduled jobs. This has worked perfectly for 3-4 years now.

Since the update this weekend, all of a sudden jobs that are scheduled to start just don't get started and it seems to be completely random. The job is definitely scheduled because I can see it in the list of scheduled jobs but the next scheduled run time comes and goes and the job never gets run.

There is nothing in the logs that I can see. This seems like a bug to me - has anyone else got any experience of this or any idea what might be the issue?

I don't even believe it is neccessarily related to the chaining as I have seen this happen when scheduling a job to start in 5 minutes time the job does not even get started the first time.

We received an email from Salesforce this morning regarding additional protection against forced logins that is being introduced in the next update.

The details in the email are not particularly detailed and we are unclear at this stage as to whether it will in fact affect us.

We have a custom scheduled job in Salesforce that generates reports on a quarterly basis. Due to various restrictions around scheduled jobs and the inability to call PageReference.getContent() in certain contexts that existed at the time this process was implemented, it works in a very round about way.

The scheduled job calls a future method that then gets an access token using a HTTP Post to login to an OAuth-Enabled connected app. The POST request includes the grant_type, username, password, client id and client secret in the body of the request in order to login to the connected app and the request is sent to https://login.salesforce.com/services/oauth2/token.

Once we have an access token we include this in authorization header and then call a custom Salesforce Webservice method that actually generates the reports.

We believe this is the only place where any functionality that might be considered a "forced login" is implemented or used in our environment. It is not clear, however, from the information available at the moment whether or not what we are doing counts as a forced login and will not work after the update.

It is also not clear if it does what changes we would need to make in order to ensure this still works after the update.

Does anyone have any more information about this change and whether it is likely to break the way this process works? If so can anyone recommend a better way of doing this? I think I remember reading that many of the restrictions around the usage of PageReference.getContent() have been removed in the past year or two so perhaps it is the case that all this additional complexity is no longer required.

We have apex classes through which we send emails based on user action in force.com sites. It was working great until Friday, But yesterday when one of our customer try to submit application, they were getting error, In debug it says “SendEmail failed. First exception on row 0; first error: UNKNOWN_EXCEPTION, INVALID_TYPE: sObject type 'SalesforceIqDataSource' is not supported.: []”. I am pretty sure we did not change anything related to apex class, email template, vf page etc.. Does anyone know the reason or best way to fix this?