system randomly crashes

[mike101876]

i have had a sight that has been going down randomly for some time (no calls, terminal locked, no DT). when this system goes down the only way that i have been able to bring it back up is to do a man reset (INT will not work). i have noticet two errors that come up alot(berr705) and (XMI000 Msg from shelf 1: XPEC error #0001)

this is an option 61 rel 23
this is a newer prob in past 6 months


has anybody experienced anything like this and if so where do i start.

any help is greatly appreciated.
thanks

 

[tellyteck1]

You start by looking at the errors and looking up what they mean. I am guessing but maybe berr means bit error rate, PEC refers to a car. The problem could be a result of a bad card. Look up errors.AJK

 

[johnpoole]

best thing you can do, to many errors alone can take one down.. could be any common card or just so many misc. errors that the ram get to full to process calls..

john poole
bellsouth business
columbia,sc

 

[mike101876]

is there any possibility that grounding could have any effect on this issue, i have always been told that the options must be grounded to an isolated ground and if they were not you could experience all kinds of crazy problems. i have noticed that at this site that there is one common ground wire for the entire room and found 9 different grounds that have been added over the years. i will translate these errors and see what i can find.

thanks

 

[tellyteck1]

It is always best to have a common ground or then a potential difference is created. The errors will point you in the right direction. AJK

 

[mike101876]

XMI000 shows a lot of info, it mentions replacing the controller, as well as "excessive electrotatic discharges"

however i am having a problem understanding the BERR705 message

BERR0705
EXCH x: y in task n. SR = a, PC = b, Addr = c, SSW = d

i pulled a history but that only goes back to the reset and doesnt show the berr0705 errror but they do show in my OTM alarms (they have occured since the reset)

 

[mike101876]

i also see mention of a patch but no idea of what release
(thread798-587277) it mentions patch # MPLR10994. this was a patch to clear the BERR0705 error

 

[Reusser]

There were several patches for Berr and INI on rls 23 but they are not available anymore - that release is retired.

Most BERR errors are critical and many are hardware caused.

Try splitting cores and just running on one side. If you still get BERR errors switch to the other side.

This will eliminate the CNIs and CP cards but you still have the per sigs and anyother Core card that could be the problem.

If you are getting a lot of XMI000 - I would also suspect that network loop card.

 

[mike101876]

thank you, i will split the cores and let you know.

 

[mike101876]

Good news, after all this the system is up and running fine with no errors, just an FYI, the E-link was taking on a lot of errors and causing all the problems. ince it was disabled everything started working great, at least for now.