VS.RAB.FailEstabPS.UuNoReply in Network

[Manmadh]

Dear Expert,

In my network many of the cells are having poor PS accessibility Kpi, More pegging Area's are in VS.RAB.FailEstabPS.UuNoReply.

I have never seen these many failures before.

Kindly suggest solution.

Thanks

Manmadh

[bunellu]

hi Manmadh, check if the RTWP as well increased for mentioned areas.
Description: Number of PS RABs Unsuccessfully Established due to no reply during the RB Setup Procedure.

Normaly can be two cases:
1. UE is not receiving the RRC_RB_SETUP
2. UE received it and send the RRC_RB_SETUP_CMP but RNC didn't received it (lost on air interface)

Check the air interface, RTWP, put the trace on one cell (biggest impacted) and perform the analysis.
Regards

[Manmadh]

Hi Bunellu,

i have analysed as you said, from call faulty logs most of the failures are due to RR_ERR_RNCAP_RB_WAIT_UE_RB_CFG_TIMEOUT.

Please suggest me any parameter optimization for same to reduce VS.RAB.FailEstabPS.UuNoReply to increase accessibility.
Thansks in advance.

[P.T.E.]

Dear,

is this issue on your network is new or old one ? if new check what opt action was taken before this degradation and as mentioned Bunellu check RTWP, User number if increased or not, EcNo distribution ...

BR,

[baby2]

Reset 1st for Huawei if you found some degrade or abnormal

[oswe]

did you adjust antenna ?
I used to have it in rural areas mainly... with huge coverage...

[baby2]

Can you fix it? :o

[qaqa]

one quick suggestion would be you increase your RB setup wait timer, RNC Level, default setting is 5s.
check your RF as suggested by many friends above.
also check your SPU at the RNC. do you have high SPU CPU load for those nodeB with high No_Reply? each NB will be connected to one Sub-sytem of SPU ( you query that through LST UNodeB for HUawei).
also check if your WBBP or LBBP has high CPU usage, it is a counter, > 65% average and >80% max is considered as high.

[ASAP168]

one quick suggestion would be you increase your RB setup wait timer, RNC Level, default setting is 5s.
check your RF as suggested by many friends above.
also check your SPU at the RNC. do you have high SPU CPU load for those nodeB with high No_Reply? each NB will be connected to one Sub-sytem of SPU ( you query that through LST UNodeB for HUawei).
also check if your WBBP or LBBP has high CPU usage, it is a counter, > 65% average and >80% max is considered as high.

Hi qaqa,

How do you check WBBP or LBBP has high CPU Usage ?

Please share, Thanks !

[qaqa]

through the NodeB- Board load counter for UMTS
LBBP is similar.
you also could high alarm at the RNC/NodeB level of high CPU load.

[ASAP168]

through the NodeB- Board load counter for UMTS
LBBP is similar.
you also could high alarm at the RNC/NodeB level of high CPU load.

Hi qaqa,

I found this and thank to you.

[ASAP168]

through the NodeB- Board load counter for UMTS
LBBP is similar.
you also could high alarm at the RNC/NodeB level of high CPU load.

Hi qaqa,

One more think, If one NODE B has Board CPU Overload, do you have any idea to resolve this ?

Thanks,

[ASAP168]

through the NodeB- Board load counter for UMTS
LBBP is similar.
you also could high alarm at the RNC/NodeB level of high CPU load.

and do you have any recommendation for me. Recently, I am having one NODE B has Board CPU Overload every day at peak hour.

Please share thanks,

[agenov]

add one more and setup the cells on different boards


and do you have any recommendation for me. Recently, I am having one NODE B has Board CPU Overload every day at peak hour.

Please share thanks,

[qaqa]

NodeB split is a solution but you need to have more than 1 Channel to get benefit.
reduce footprint could help as well.
which board is overloaded? WMPT or WBBP? which RAN version you have?

[Manmadh]

Hi qaqa,

I have changed RBRECFGRSPTMR to 15 s from SET USTATETIMER .. even though failures are very high.

Thanks & Regards
Manmadh

[bunellu]

Hi Manmadh,
Can you perform one thing.

Deactivate the 2ms TTI under the NodeB license for 30min or 1h depends on the counter measurement availability, then activate 2ms TTI again.
Then check the VS.RAB.FailEstabPS.UuNoReply counter during deactivation period, improved or not. The 10 ms TTI is more reliable in poor UL ch quality but less UL throughput and large TTL (ms).
Check as well the counter which show that during the deactivation period all the users used 10 ms TTI (available in Huawei).
Update us after that with your results.

[oucht79]

Dear
This counter is general counter you have to check High RTWP, High TCP, cell radius counters (overshooting), Poor coverage...etc

thanks

[baby2]

through the NodeB- Board load counter for UMTS
LBBP is similar.
you also could high alarm at the RNC/NodeB level of high CPU load.

@RNC side also ,
if you are BSC6900 , UTRAN team have to manual config cell by cell to board RNC

[brianm]

Hi qaqa,

One more think, If one NODE B has Board CPU Overload, do you have any idea to resolve this ?

Thanks,

You either split the node B and have a 2nd co-located site OR you can replace the WMPT with a UMPT which more processing capacity

[yassanino]

try to encrease total received power in the BTS , and enhance the signaling power in UL , also check your Base Band usage , may be you need extension in hardware , otherwise make trace on your counter to see in which phase you have the failure , but first do a restart and monitor
BR

[Manmadh]

Hi Guys,

My problem solved after enabling.

DRA_BASE_COVER_BE_E2D_SWITCH & DRA_BASE_COVER_BE_E2D_INITSEL_SWITCH.

Success rate has increased up to 4 %.

[brianm]

Hi Guys,

My problem solved after enabling.

DRA_BASE_COVER_BE_E2D_SWITCH & DRA_BASE_COVER_BE_E2D_INITSEL_SWITCH.

Success rate has increased up to 4 %.

Whats the full name of these parameters I cant seem to find them?

Are they NodeB or BSC parameters?

[baby2]

Whats the full name of these parameters I cant seem to find them?

Are they NodeB or BSC parameters?

BSC/RNC parameter

[samuraial]

Depend what RAN version are you using my friend,

On RAN14 I cannot find them also but seems by the name of the switch similar to TTI2ms switches, this one is for E-DCH and DCH, a feature that Huawei had it before i think.

HSUPA E2D feature that included 2 reserved switches but now they have named it as you explained it.
RNC MML：SET UCORRMALGOSWITCH: ReservedSwitch0=RESERVED_SWITCH_0_BIT5-1&RESERVED_SWITCH_0_BIT7-1;

For all that have RAN14 and below should check this feature as it is the same.
Explanation:

E2D algorithm reconfigure HSUPA user to DCH according to the coverage and throughput, make sure the user could transfer data successively.
Pros & Cons
Benefit: decrease call drop, improve user experience in UE power limitation status.
Disadvantage: Impact user experience for high throughout requirement.(as users pass to DCH, but better than fail :) )

B/R

[yassanino]

Depend what RAN version are you using my friend,

On RAN14 I cannot find them also but seems by the name of the switch similar to TTI2ms switches, this one is for E-DCH and DCH, a feature that Huawei had it before i think.

HSUPA E2D feature that included 2 reserved switches but now they have named it as you explained it.
RNC MML：SET UCORRMALGOSWITCH: ReservedSwitch0=RESERVED_SWITCH_0_BIT5-1&RESERVED_SWITCH_0_BIT7-1;

For all that have RAN14 and below should check this feature as it is the same.
Explanation:

E2D algorithm reconfigure HSUPA user to DCH according to the coverage and throughput, make sure the user could transfer data successively.
Pros & Cons
Benefit: decrease call drop, improve user experience in UE power limitation status.
Disadvantage: Impact user experience for high throughout requirement.(as users pass to DCH, but better than fail :) )

B/R


how can that be benifit if u turned off switchs to 2ms TTi , u better rollback and try to tune CPICH to Rab offset, ( increas signaling) ,increase PTX target up by 2dB, also intial an minimaum bitrate in UL/DL , as final you may adjust your RAB definition margin ( get it bigger too) and monitor ( one action after another , never two action at same time)
Br