We are having an issue where UCCX has stopped running updated script - old script seems to be stuck.
UCCX System version: 220.127.116.1100-41
CUCM System version: 18.104.22.168000-1
Cisco Unified CCX Editor 8.0(2.0)
2 servers in cluster.
Script is running from subscriber - but has been updated in Publisher (as I always have).
In the past, I've updated scripts using CCX Editor without issue.
Yesterday I uploaded a prompt, modified a script in CCX Editor, then saved it.
Then in UCCX, refreshed the script, refreshed the Application, refreshed everything, noted in Script management that the modified date changed, dialed the number, the prompt has not changed. Still has the old prompt playing.
I've tried creating new variable, setting it in UCCX with checkbox and without checkbox, etc. - No change, issue remains.
Tried to update the Weekday Time of Day and save (to rule out trouble with prompt section) - refreshed all appropriate, times have not changed. Appears that script is just not taking.
If I open the script from UCCX, it shows the updated script.
As a test, I ran Reactive Script from CCX Editor, when the phone is answered, it shows the script from the subscriber IP and the script it shows is the OLD script.
It is as if the replication is failing.
In Tools > DataStore Control Center > Data Stores > Cisco Unified CCX Repository DataStore > the subcriber is showing "Last Update Time" that is not the most recent update.
Clicked on Synchronize Data > waited ages > went back into it - the "Last Update Time" still is the last time I successfully updated the script (a few days ago) - it hasn't synched.
Where am I going wrong? Has anyone seen this issue before, is there a way of resolving it without restarting the box?
Any help appreciated.
Thanks for the reply.
I am definitely making changes on the Publisher (Master) only. We've double checked.
This morning we experienced an unplanned power outage (UPS, generators, et al - thanks to a glitch with the fire suppression system) - our CUCM and UCCX boxes went down hard.
After power was restored and our boxes powered back up, the script updated on the subscriber (which was on the other side of the country and didn't go down) and now it is all good.
I had found the following info on another forum:
" During my installation of the UCCX 8 , i had similar problem .
Actually Any update within the Script wouldn't be updated unless i cycle the Engine in both servers : 1st and HA Node .
So, try to Restart the Engine Process for both Servers , one at a time . ( from the Servicability menu )
i think this may work ( This will take minimum Outage time , even the connected calls wouldn't be affected as the BackUp Node will Take over .. )
As the devices power cycled I didn't have a chance to try this - if the issue resurfaces, I will post the outcome of the above advice.
Will leave the question open for the moment as it isn't really answered (even though it is resolved...by a stroke of luck and bad luck at the same time).
Sometime when you save the script through the script editor, the new script is not updated properly. Therefore save it some other location (may be on
Desktop).Then upload the same script through Script Management and then refresh it in both the servers.
Same for prompts, upload it through prompt management and refresh it.
It will work.
Brett, how did you get on with this?
It sounds like an issue we had a few weeks ago. We ended up updating UCCX to 22.214.171.12404-12 and haven't seen the issue since.
I must confess, since the power outage we experienced (mentioned further up) - I haven't had any trouble
uploading via CCX Editor.
I do recall having one shot at saving to desktop, uploading it and overwriting it - which actually threw an error (which I lost the screenshot of) and then tried again and it worked.
As I have no issue with uploading via CCX Editor - I've not bothered to do it that way again.
Further to this - we have actually updated UCCX from version 126.96.36.19900-41 to version 188.8.131.5204-12.
We had some other issues with reporting and HA over WAN and Cisco advised we needed to bring our version up to this before they could TS further.
This could be assisting with keeping the problem away...but who reallly knows for sure?
Something else that may actually have been the cause or something for others to look at in future - is that it may have been that the Subscriber had become the master without my knowledge. The damned icon looks the same at a glance (same blue square with a white letter) - for the Cisco Unified CCX Engine service in Serviceability (or other services). I can't tell now as it is too late now with it all working as it's supposed to.
Checking the 'mastership' (blue square with white M) next to all the services in Publisher Serviceability Network and Feature services - will definitely be something I will check, if there is a next time, in future.
Be warned - if you don't already know - the 'slaveship' icon is basically idential and tiny but with an 'S'. A very quick glance without conscious observation may appear as though it is Master.