RUN_RSAIMMERGE_TRANSFER: Long runtime during upgrade to EhP7 MaxDB

[Former Member]

Hi gurus,

We are in downtime phase of our PRD system.

We are in phase RUN_RSAIMMERGE_TRANSFER. We still in this phase during 4 hours. In others RDBMS (Oracle - SQL Server), this phase is very quickly (lower to 1 hour)

Server Name     No. Type PID      Status  Reason Sem Start Error CPU Time   User   Report   Action          Table Sem Sem Sem Sem Sem CurrSem. Mutex Mutex Mutex

<server>_<SID>_00  41 BTC      6870 Running            Yes            14154 DDIC   CL_AIMME Sequential Read E071   0   0   0   0   0        0     0     0     0

We are using SUM 1.0 SP08. Our system has SAP Kernel 7.40 lvl 27 with MaxDB 7.9.08.008, and we go to SAP BASIS 740 SP04. SAP Note 1442150 is implemented by SPS level...., Note 1460004 also implemented

WP log:

A Sat Jul  5 23:23:20 2014

A  AppServerTimeSync, checkLocalTime():

A    Initializing local WP variables:

A    zdate_LastLocalTime with 1404595400

A    zdate_LastLocalBias with 0

A  ***GENER* Trace switched off ***

B  dbrda: Redirect database access is disabled via rsdb/rda profile parameter setting.

E  Enqueue Info: enque/use_pfclock2 = FALSE

M

M Sat Jul  5 23:23:21 2014

M  SecAudit(rsauinit): WP attached to existing shared memory.

M  SecAudit(RsauShmInit): addr of SHM for Audit.. = 7f6fb646d000

M  SecAudit(RsauShmInit): addr of RSAUSHM........ = 7f6fb646e000

M  SecAudit(RsauShmInit): addr of RSAUSLOTINFO... = 7f6fb646e9d0

M  SecAudit(RsauShmInit): addr of RSAUSLOTS...... = 7f6fb646e9e0

A

A Sat Jul  5 23:23:43 2014

A  WP has reached abap/heaplimit = 40000000 bytes

Instance parameters:

abap/buffersize = 900000

ztta/roll_area = 10000000

ztta/roll_extension = 4000000000

abap/heap_area_dia = 4000000000

abap/heap_area_nondia = 4000000000

abap/heap_area_total = 8000000000

gw/max_conn = 2000

gw/max_sys = 2000

rdisp/appc_ca_blk_no = 2000

rdisp/max_arq = 2000

rdisp/max_comm_entries = 2000

rdisp/tm_max_no = 2000

rdisp/wp_ca_blk_no = 2000

rsdb/cua/buffersize = 15000

rsdb/ntab/entrycount = 90000

rsdb/ntab/ftabsize = 90000

rsdb/ntab/irbdsize = 18000

rsdb/ntab/sntabsize = 3000

rsdb/obj/buffersize = 130000

rsdb/obj/large_object_size = 45000

rsdb/obj/max_objects = 40000

rtbb/buffer_length = 60000

sap/bufdir_entries = 10000

zcsa/db_max_buftab = 35000

zcsa/presentation_buffer_area = 20000000

zcsa/table_buffer_area = 250000000

em/global_area_MB = 2048

em/initial_size_MB = 12288

em/address_space_MB = 4096

abap/heaplimit=40000000

Any idea for speed up the phase?

Regards

[former_member188883 · Accepted solution]

Hi Javier,

You have 2 options

1) Try with 740 kernel SP 48 and check the results.

2) As it is known bug for MaxDB, please raise OSS mesage for further investigation and solutions.

Hope this helps,

Regards,

Deepak Kori

[Former Member · Accepted solution]

Hi Deepak,

We have a VH OSS Incident, no additional information.

Yes, we updated the kernel to SP60. The phase finished after 24 hours with this lvl....

I hope that SAP releases the solution for this MaxDB bug in short time.

Regards

[former_member188883 · Accepted solution]

Thanks for update Javier.

You may close the thread now.

Regards,

Deepak Kori

[Former Member]

Hi all,

Finally, the phase finishes after 24 hours.

Regards