800xA HMI Update Problem
We have 800xA system version 6.0.3 with merged aspect connectivity server. Also Connectivity server connect to seperate control network (non-safety and safety with network area 20 and 21) via seperate NIC. Time synchronization is ok in all network node (workstation and controller). Now we have problem in all of client when HMI used. Many time in day operator workplace can't update for 5-10 seconds. In this time show red box and red icon in upper right corner and 800xA workplace hanged. We synchronize aspect directory database manually and for 10 days we have no problem. Now we have HMI problem again. We check service structure system status viewer and Aspect directory and FFDataStorageAndDistribution serivce provider in ASCS2 automatically stopped. and we have to uncheck and check again manually for service running. We check microsoft windows system event in ASCS1 and see AfwCSLib system error with EventID 3. I attach these pictures. Please see attached picture and guide me for solve this problem.
Asked 7 months ago
Voted best answer
1
This looks like a severe issue - please file a support case with your regional ABB support center.
"CSLib Message Queue Full" indicates that a service or process is subjected to more network application traffic/messages than it can handle.
Given that the service involved is "AfwADServer" (the Aspect Directory) the effects can quickly become system wide (e.g. a transaction need to be acknowledged by all aspect servers).
The workplace watchdog will detect a hung up workplace window after a 30 seconds delay. The small button in the upper right corner of the screen can be used to create memory dumps of the workplace process. For troubleshooting this problem, I recommend that you collect 2-3 dumps, preferably from different clients. After the dump you can press "Continue Waiting" and if the watchdog fires again, take a second memory dump before selecting the "Restart Workplace" button. Collect, compress and make the memory dumps available to support staff.
Your observations of the FFDataStorageAndDistribution may be an important clue - please make sure to forward it to support when filing the support case with them.
I also recommend to make extractions from the [Workplace Structure]Web System Workplace:System Event List from the time before the onset of the problem up until and some time after recovery. Copy as ASCII text to to notepad or Excel (do not paste images in Excel).
Examine the following directory tree (and subfolders) in affected servers & clients and collect any file having a modification time more recent than shortly before the time of the onset of the problem (=leave older not changed files). You may also elect (after discovering some recently changed files) to collect & compress the entire folder structure below the path below using the Windows Explorer command "Send to compressed folder". Again, compress and make the files available to support staff.
C:\ProgramData\ABB\Process Portal A
Last, I would like to mention that using CNCP clock synchronizatinn "vertically", i.e. between Windows / AfwTimeService and AC 800M is not reliable until you have upgraded to 6.0.3.3 (or 6.1.0) due to a flaw in the AC 800M Time Adaptor part of AC 800M Connect. The problem is known as Product Issue 800xACON-OL-6000-007. Horizontal sync (controller to controller) via CNCP is not affected. Before upgrading to 6.0.3.3 or 6.1, ABB recommends using SNTP in either direction to keep the clocks in Microsoft Windows and AC 800M in sync.
"CSLib Message Queue Full" indicates that a service or process is subjected to more network application traffic/messages than it can handle.
Given that the service involved is "AfwADServer" (the Aspect Directory) the effects can quickly become system wide (e.g. a transaction need to be acknowledged by all aspect servers).
The workplace watchdog will detect a hung up workplace window after a 30 seconds delay. The small button in the upper right corner of the screen can be used to create memory dumps of the workplace process. For troubleshooting this problem, I recommend that you collect 2-3 dumps, preferably from different clients. After the dump you can press "Continue Waiting" and if the watchdog fires again, take a second memory dump before selecting the "Restart Workplace" button. Collect, compress and make the memory dumps available to support staff.
Your observations of the FFDataStorageAndDistribution may be an important clue - please make sure to forward it to support when filing the support case with them.
I also recommend to make extractions from the [Workplace Structure]Web System Workplace:System Event List from the time before the onset of the problem up until and some time after recovery. Copy as ASCII text to to notepad or Excel (do not paste images in Excel).
Examine the following directory tree (and subfolders) in affected servers & clients and collect any file having a modification time more recent than shortly before the time of the onset of the problem (=leave older not changed files). You may also elect (after discovering some recently changed files) to collect & compress the entire folder structure below the path below using the Windows Explorer command "Send to compressed folder". Again, compress and make the files available to support staff.
C:\ProgramData\ABB\Process Portal A
Last, I would like to mention that using CNCP clock synchronizatinn "vertically", i.e. between Windows / AfwTimeService and AC 800M is not reliable until you have upgraded to 6.0.3.3 (or 6.1.0) due to a flaw in the AC 800M Time Adaptor part of AC 800M Connect. The problem is known as Product Issue 800xACON-OL-6000-007. Horizontal sync (controller to controller) via CNCP is not affected. Before upgrading to 6.0.3.3 or 6.1, ABB recommends using SNTP in either direction to keep the clocks in Microsoft Windows and AC 800M in sync.
Answers
0
This problem don't solve.
If this can be help, According to our experience in one month usually when we use remote desktop on ASCS1 or ASCS2 this problem occur.
If this can be help, According to our experience in one month usually when we use remote desktop on ASCS1 or ASCS2 this problem occur.
Answered 5 months ago
Add new comment