Notifications

101 views

Description

This alert is reported when the instance get a lot of emails to process and it takes some time to process, and hence appear stuck. Sometimes, the inactivity of the reader can also cause this.

Cause

The Email Reader jobs can sometimes retrieve a lot of emails at the same time, which can cause the event queue to be overloaded, so it takes some time for the incoming processes (e.g. inbound actions) to clear out the pending actions. So, at this time, they appear to be stuck, but they could be functioning. Sometimes it can also happen that the Email Readers are not active.

Resolution

There are different causes for the alert. Here is a list of possible causes:


Case   

Possible problem

Recommended solution

Case 1

If the alert is caused by an large amount of unwanted emails being received in the instance, if they are sent by unwanted senders (e.g. postmaster@domain  ) or have a common subject (e.g. Undeliverable) or identifiable header

You can check with the following link: <instance>/sys_email_list.do?sysparm_filter_only=true&sysparm_query=typeINreceived%2Creceived-ignored%5Esys_created_onONToday%40javascript%3Ags.beginningOfToday()%40javascript%3Ags.endOfToday()%5EORDERBYDESCsys_created_on


In some cases, outbound emails sent from the instance are being rejected by the target mail server or undelivered and are bouncing back to the instance. This could causing email reader job to run for a long time and hence overall email processing is being delayed.
e.g. incoming emails will show in the body a message like "Your mail server is rejecting with below message: <reason>"

Please create an email filter to ignore them. Also review if you can disable the notification for the users with bounced emails


Case 2

If at the same time, the event processor looks overloaded (e.g. there are 500+ events waiting to be processed)

You can check with the following link: <instance>/sysevent_list.do?&sysparm_filter_only=true&sysparm_query=sys_created_onONToday@javascript:gs.beginningOfToday()@javascript:gs.endOfToday()%5Estate%3Dready%5EqueueISEMPTY%5EORDERBYDESCsys_updated_on

Validate if the instance is sending a large volume of notifications at the time and validate if there is a problem with the notifications

Case 3

Reviewing the active transactions on the instances, and validate if you find one or two inbound actions taking a long time to execute. It will show as "event process" running a long time.

You can check with the following link: <instance>/loading_transactions.do


You might also find the transaction on the nodes stats.do if you are connected to the one processing the inbound action

You can check with the following link: <instance>/stats.do

Validate why the inbound action is taking a large amount of time. They are usually related to inbound actions with large queries, or non-optimized queries.

Case 4

If when checking the Email Diagnostic Page the email receiving does not look operational.

You can check with the following link: <instance>/email_diagnostics.do

Please validate if the Email reader job is correctly running. Check KB0520595 



Please ensure to validate any change on your non-production instance before making changes to production.

Article Information

Last Updated:2020-10-01 10:25:09
Published:2020-10-01