SXI_CACHE is hanging or Not working
Case 1: Problem with Cache Refresh
When we perform any changes to our design objects and configuration objects which were created in our IR and ID and if the changes were not reflected to that objects, at that time every one mind would point to perform the cache refresh. We have different options available to perform the cache refresh.
1. In IR from MenuàEnvironmentàClear SLD Data cache.
2. In ID from Menu ClearàEnvironmentàSLD Data Cache.
When we log in to ABAP stack and hit the transaction SXI_CACHE there you can observe the status will be in red color and the cache will not be up-to-date and unable to refresh cache. Even though if we perform the cache refresh the changes are not reflecting since the cache is not working properly.
From IR from menu using cache notifications we can see whether cache is refreshing or not. From that option also it shows the status as RED color and cache is not up-to-date and cache is not refreshing.
So how to resolve this issue and what might be the problem for displaying the status as red and cache is not up-to-date.
To resolve this issue logon ABAP stack and check the RFC Destination of
Type H: INTEGRATION_DIRECTORY_HMI in this check whether the path prefix is maintained or not.
If we observe the above screenshot we can identify that path prefix as not maintained anything. So maintain the path prefix as in the below screen shot and save it.
After performing the above step go to SXI_CACHE and perform cache refresh. Now cache will perform the refresh and the status will be as below.
Case 2: Problem with Cache Refresh
When Path Prefix is fine and you unable to perform the Cache Refresh
GO to SM21, search for the system logs. You could find multiple entries as below for user “SAP_PI_RFC”.
During processing of asynchronous messages a lock was triggered that prevents a message being processed more than once in parallel. Either the lock prevented this, or the lock was not cleared in an earlier call and so now prevents further processing. It should not be possible to execute a message more than once. Notify support of the problem. If the lock was not cleared, you can remove it manually SM12 – You can confirm with PI team). You should then be able to continue processing without problems.
Solution: Please make sure these are NOT recent entries. Delete these entries from SM12 after confirmation from functional/Technical team.
Case 3: Problem with Cache Refresh
There is a chance that system has too many suspended DB commands at DB02. There might be any SQL command blocking other inbound queue.
Solution: It just happens due to huge blocking. Kill the culprit process which is causing blockage.
Case 4: Problem with Cache Refresh
Once I have resolved the issue by restarting the Java stack system from SMICM T-code.