CPPM Redunduncy

[krasemir]

7 months ago , we installed CS1000E rls 5.0 ,I`ve done the upgrade for the loadware and installed the deplist , Last week while we tested the CPPM redundancy , I`ve rebooted the first CPPM witch was active , suddenly the second one also reboot ,all phones was down , and the strange thing all MGC sequentially rebooted and the system go back to the default loadware . Any explanation for that please and why the second CPPM doesn’t handle the phones ??

 

[DocVic]

Wow, you rebooted an active CPU, in a redundant system!? Were you trying to make the system fail? When you "upgraded, did you load both sides, one at a time? Did you sync when you were done? Did your vendor help you on this project? Or did you just do it on your own?

DocVic
Dedicated to Nortel Products till the end.
Need help? Call Me Now!

 

[krasemir]

As we know in HA system ,one CPPM is Active , the other is in stand by mode , what if there is power cut on the MG in wich stand the active cppm , is it write the other cppm must automaticaly handle the PABX JOB ???
Regarding the "upgraded " : While the system is running , via EM I`ve upload the LW, then upgrade the MGC`s.
Do i have to do it twice for each CPPM ??

 

[bigindian65]

loaware will allow you do do one mgc and reboot, or all mgc's then all re-boot. Sounds like it did what it was suppose to, except didnt write the new loadware, or somehow you missed something and was only 1/2 completed???

Mato' Was'aka

 

[krasemir]

it is write the new loadware, and when i print iss , i saw the new one , and as i said now it go back to the default .
did the way that i used ( via EM ) is correct to install the loadware ? do i have to install them twice for the 2 CPPM ? and how ??

 

[KiwiOz07]

Hi,
I have seen this before when the disk state is not redundant. If you stat cpu in LD 135 you can check.

 

[SSJVirus007]

The only way I see thing went back to default is both your CPU are not sync. Did you check CPU health or CPU status before you restart the active core?

 

[PareshTC]

Firstly, if you initiated a manual reboot of the system, using the RST switch on faceplate, the stadnby CPU wil lrecognise this as a planned reboot and reboot itself. However if you were to simply pull the active card or ELAN cable from active (neither is suggested), it would have simulated failure of the active CPU and the standby would have taken over. This is mentioned in the Whats New NTP.

Regarding your loadware getting rolled back. Can you confirm the disk state was redundant before you started the loadware upgrade? Also after the reboot, which CPPM was active, the orig active or the orig standby?

 

[DocVic]

Call your local vendor today. He can have you fixed up and have this tested for D-Recovery in no time. He's is there to help you, (and make a couple of bucks to support his business and family). Give it a try.

DocVic
Dedicated to Nortel Products till the end.
Need help? Call Me Now!

 

[biv343]

My first guess is that this isn't an HA system, but a geographically redundant system with a SMG/NWAY. It's hard to tell with the details given, but it sure sounds like the MGCs registered to an alternate call server based on the fact that the MGCs downgraded their loadware.

 

[xtract]

Core redundancy works on heartbeat mechanism or what we called health, if the active core's health is down the standby core will detect it and will take over,but if you suddenly rebooted an active core on redundant mode, the standby core will which is in HIRM or redundancy background test will lost connection to HSP and it will reboot itself.