ERR0036 Question (Critical)

[clayndwoods]

I keep getting the following ERR036 on my Option 61C on Succession 3.0

ERR0036
Set Based Buffering was unable to put a message into an SSD message buffer.
Action: To fully recover from these errors the system requires a Warm Restart Initialization. The manual INI recovery operation may be scheduled at the customer's convenience.
Severity: Major

This has been going on since Friday. I had some starnge erratic behaviour occur today i.e. ELAN went down, intermittent no dial tone at phones, and the D channel went down on one PRI. I did the Warm Restart Initialization (just because it supposedly needed it) assuming I would clear all the faults at the same time. Although everything returned to normal, I still get that error. Does anyone know what it really means? What is an SDD message buffer and what is it associated with?

I do have a DDCH that goes off into never-never land on occassion and am changing it tomorrow.

 

[janchundia]

post
ld 22
req prt
type parm

I need to check the buffer size in your configuration.

 

[clayndwoods]

PARM
LPIB 500
HPIB 500
500B 160
SL1B 160
NCR 1024
MGCR NULL
CSQI 24
CSQO 24
TUBO NO
CFWS NO
PCML MU
ALRM YES
ERRM ERR BUG AUD
DTRB 100
TMRK 128
FCDR NEW
PCDR YES
TPO NO
TSO NO
CLID YES
DUR5 NO
MLDN YES
MARP NO
FRPT NEFR
DCUS 1
MSCL 100
PMSI
MANU PMS1
PMCR 60
PORT NONE
NDIS 20
OCAC NO
MTRO MR
SBA_ADM_INS 000
SBA_USER 000
BCAP SPEECH
IDLE_SET_DISPLAY TEXAS MEDICAL CENTER
ICON YES

 

[janchundia]

You need to change the buffers because are small size.

Check the planning and enginnering guide for the correct parameters size.

The following parameters are for 61C with Pentium proccesor
HPIB 3500
LPIB 3500
500B 2000
SL1B 255
NCR 20000
MGCR 25
CSQI 255
CSQO 255

After change and dump, you need to reboot the system.

If you buffers are small, then the system don´t have resources for call proccessing.

Regards.

Javier

 

[johnpoole]

javier is 100 percent right, another point when you change the buffers, do a edd and then an ini. the switch does not see that change dynamic.. it falls into the same param as an old config.sys

john poole
bellsouth business
columbia,sc

 

[clayndwoods]

I will try it. Can't hurt. The call registers are dangerously low on this sucker, huh. As well as everything else. This machine started with a few trunks and less than 60 digital phones, and 40 analog line 10 years ago. It now has over 600 lines, and 4 PRI's on it. It's time huh.

 

[johnpoole]

when an upgrade is done (rls) you need to check the buffers to make sure they are maxed out for your new rls. when i look at a switch for maint, i make sure the buffers are right. unlike the old days, you can max them without the performance suffering. they can't go beyond the max allowed. also check the ram to make sure it's been kept up to specs

john poole
bellsouth business
columbia,sc

 

[clayndwoods]

The RAM was upgraded to spec when we did the 25.40b upgrade in preparation for 3.0. I assume it's max'd out as I believe all the memory stick slots were filled at that time. I could be wrong, but I believe otherwise. I believe we have the CPII which is a 68090 processor is it not?

 

[johnpoole]

stat cpu in 135 will tell youif all slots are full ld 22 prt cequ will tell you the size of each stick

john poole
bellsouth business
columbia,sc

 

[clayndwoods]

I believe the error code is crap. I think it's the old VMBA trying to put messages in the buffer for output on aml 8 which is for a recently removed Meridian Mail system. I have to remove all the voice mail system from the station records before I can remove the application. The loops and aml for the old system are already gone. What do you think?

 

[johnpoole]

your probley right, remove all shadows of the mer mail so those codes will not effect buffer space

john poole
bellsouth business
columbia,sc