09-13-2011 03:47 PM - last edited on 03-25-2019 10:59 PM by ciscomoderator
Hi Cisco Team,
I did report this a couple of times already as a newly introduced bug with CCA 3.1.X this was not an identified bug in version 3.0 and has only just recently surfaced.
The Issue:
An interesting thing to note with the above, If I build say a 50 line translation profile/rule system it works, but if you open this up on another system that has a fresh install of CCA 3.1.1 it locks CCA up and CCA has a massive memory leak and spikes up to a little over 900mb in memory usage, this was able to be replicated on two systems, the poor netbook just crashes as it runs out of RAM... LOL I thought this issue was restricted to managing builds that were done with CCA 3.0 but it would seem this problem is producible on a fresh 3.1.1 install, still working on verifying this and recording it as well.
Work Around:
Make the required configuration on the CLI following the design flow that CCA uses, whilst it will work CCA will not recognize the configuration, even when it modifies it itself, but if you roll back to version 3.0 it is recognized and you can modify it then using the GUI.
Its nature is a problematic one, which ever way you look at it the management side of this is a nightmare and has a massive downside to it, the issue is reproducible on multiple sites as our target space of late has been to sell the UC-500 series with 100 number in-dials either on BRI or PRI circuits, it gives us a selling edge given how much can be done when you have this type of flex, but this problem pretty much destroys the majority of that flexibility... Which I am sure can be appreciated.
I do have a screen recording of this problem, but sadly this is recorded using TeamViewer and not WebEx so unles syou have TeamViewer to play back the recording, there is no sense in my uploading it... But this should not prevent Cisco from looking into this and identifying the issue internally.
Cheers,
David.
09-16-2011 12:14 PM
Hi David,
Did you get the CCA troubleshooting log on this one? And any recording you have is fine (TeamViewer is freeware). This will allow us to get a good idea what is happening on the UI and your errors. Otherwise, we risk finding problems when reproducing your case, but not exactly the problems you experienced.
Thanks,
Laura
09-18-2011 03:13 PM
Hi Laura,
I don't have any of the debug info as I had to do it via a TeamViewer session to the clients computer, and once the bulk of the work was done that system was decommissioned as they try to find me another suitable way of managing the system remotely.
Attached is the TeamViewer recording, which does not display much but I hope it gives you some indication as to the problem that is happening.
I will however make note of the fact I need to capture these logs, it is a habit I have not got myself into yet and I should **Smacks myself on the hand for it**
I will try and produce a log for you when they ask me to update the DID mapping, which no doubt will be soon, as I think they have not made me shed enough hair yet
Cheers,
David.
05-30-2012 03:18 AM
I too have a simialr issue that has appeared on a system.
A system configured PURELY with CCA 3.2
Created an in-dial on all BRI port to map the first 80 number to extensions in one rule.
I have a running config of the system after this and all was OK and all items showed up ok.
Something happened and now the CCA doesnt show ANY in-dial mapping however they are still there in the running config. In the config the rules now skip every 10-20 item and then they are added at the end.
ie
rule 13
9999 9900 / 300
x
9999 9909 / 309
rule 14
9999 9920 / 320
x
9999 9929 / 329
rule 15
9999 9940 / 340
x
9999 9949 / 349
xxx
rule (lets say about 20) - then fills in the missing bits EXCEPT - see below
9999 9930 / 330
x
9999 9939 / 339
EXCEPT indial 9999 9910 - 9999919
Any ideas how to force add the 9999 9910 - 9999919 range or what do I need to delete to enable the pruging and fixing of CCA?
05-30-2012 03:40 AM
Sorry Tim
I wish I had some good news for you but I don't, if the problem was not
resolved in 3.2 then the same issue still exists.
Here is the catch, although CCA created the translation rules/profiles and
the associated connections to the dial-peers, it only takes one little
stuff up from CCA when it is writing the config for the translation config,
and BINGO! The whole thing has gone the way of the dodo bird, and the only
way to fix this is to delete the translation profile and force CCA to
recreate them from scratch... Not a practice you want to do on every system
that goes this way.
By the way, Cisco were fully aware of this issue and I documented the
daylight out of it, I never actually got a resolution to this problem I
just found ways to work around the issue, albeit very annoying to have to.
Cheers,
David Trad
"Sent from my Acer Iconia A500"
On May 30, 2012 8:19 PM, "focusonit" <
05-30-2012 03:45 AM
How can I delete the entire in-dial rules (Assuming from CLI) so that CCA can recreate them?
The rules are really simple 1-1 mapping + 2 conference in dial extensions.
Regards,
Tim Austin-Mills
IT Engineer
Focus Communications
Phone: 03 8788 3710
Fax: 1800 600 999
Call us today about Telstra’s new 4G network – Australia’s fastest mobile broadband:
05-30-2012 03:50 AM
Tim,
This could be a pain :-\
Try first deleting the site from CCA and recreate it, hoping that CCA
doesn't retain existing information and rescans the whole system and is
able to pickup the configuration...
Crossing my fingers that this may also help you out, the other way is a
real pain and even harder to explain on a forum.
Let's cross that bridge only if we have to.
Cheers,
David Trad
"Sent from my Acer Iconia A500"
On May 30, 2012 8:45 PM, "focusonit" <
05-31-2012 07:42 PM
No luck with creating a new site.
I had a great Cisco engineer help out - looks like it wasnt the first dialplan they had seen messed up
He quickly deleted all the mappings via CLI and then we were good to use CCA again
06-01-2012 11:33 PM
Hi Tim,
I'm stoked that you managed to get support staff to help you out,
especially on the CLI side of things, that is not always something everyone
gets.
To be honest with you I was so no looking forward to having to explain how
to fix this up via CLI, I was just going to organise a time to remote in
and work on it with you.
Unfortunately this is a regular occurrence, so after awhile you get used to
working out what to do... But I'm happy that you are all sorted now
Cheers,
David Trad
"Sent from my Acer Iconia A500"
On Jun 1, 2012 12:42 PM, "focusonit" <
06-03-2012 05:07 PM
I've had this issue also, and have found that just doing a one-to-one mapping of a single extension to a single indial during the CCA startup wizard, then going in to the incoming dial plan, hacking it all out and starting again.
I tried to do it the CCA way with 100 number indial, and had the same issue - it crashed the application and you have to start the wizard again from scratch.
Regarding the CCA config, just make sure you hit Apply or OK after you do anything. It doesn't seem to like being told to remove some CLI and add some CLI in the same process.
Cheers,
Scott
06-03-2012 05:50 PM
Hi Scott,
Next time you are working on this, try and do something for me you will be surprised what you see happen..
Press the "F2" key and bring up the debugging dialogue, watch what CCA does when you try to modify the DID mapping.
No matter how many times I tried to get Cisco Support to understand this they just wouldn't listen it was like it the problem didn't exist and it was something I was doing wrong
So in the end i decided to take the matter in my own hands, wiping out the DID mapping manually via CLI time and time again, I almost perfected it as an art so that way SBSC would never have known I made CLI changes, practicing on my LAB system until I got it right, then fixing up multiple deployments effected with this problem after hours in my own time so the customer was not adversely effected.
Understand though, whilst I criticize Cisco heavily on many things, I still love the products they make, what I call into question is the many silly decisions they arrive at by someone who must not really understand the market or doesn't realize that there are other markets out there outside of the USA, or they just stopped listening to their partners and get a feel from them of what needs to be done, and this was one of Cisco's strong points over their competitors and over the last 2 years that just dwindled away.
Ringing SBSC and getting denied support on such small trivial issues because of not being Express UC certified and which was directly attributed to a problem created by them and their software, was the straw that broke the camels back for me... Even though when I escalated it, it was just fortunate that someone recognized me from these forums and was kind enough to approve CLI changes, but that made me stop and think, what about others who are suffering the same fate, it was this one problem that kicked of a chain of events for me.
Cheers,
David.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: