Quantcast
Channel: SCN: Message List - SAP Applications on SAP Adaptive Server Enterprise (SAP ASE)
Viewing all 3678 articles
Browse latest View live
↧

Sybase ASE memory configuration

November 10, 2014, 7:01 pm
$
0
0

Dear All,

 

I am trying to re-configure ASE memory on our SAP servers, most of the servers has shared memory values set to unlimited.

 

Current databases were configured with default memory usage, now i am planning to increase current size and configure data cache to use 50 to 55% of the physical memory.

 

Our production system has 64 GB of memory , if i use half of it for database then it will be 32 GB since kernel parameters are already set to unlimited , shall i increase max memory and then followed by data cache and wash size ? Currently only 16k I/O buffer pool is configured for a smaller size, i would like to configure 128k I/O buffer pool with at least 10 GB.

 

Is it a good approach to start with ? I am not a Sybase expert , please help me with correct values

 

Please see current settings below

 

kernel.shmmax = 9223372036854775807

kernel.sem = 1250 256000 100 8192

kernel.shmall = 1152921504606846720

vm.max_map_count = 2000000

 

 

Max Memory

 

Configured Value:  1792000 memory pages(2k)

Default Value:  139264 memory pages(2k)

Run Value:  1792000 memory pages(2k)

Value Range:  139264:2147483647

 

 

General Information

Cache ID:  0 

Cache Name:  default data cache 

Status:  Active 

Type:  Default 

 

Size

Configured Size:  400 MB

Current Size:  400 MB

Configured Partitions:  8 

Current Partitions:  8 

 

Miscellaneous

Configured Replacement:  Strict LRU 

Current Replacement:  Strict LRU 

Buffer Pools:  1 

Overhead:  8 

 

 

Thanks,

↧
Search

Re: Sybase ASE memory configuration

November 11, 2014, 5:05 am
$
0
0

Dear Aravind,

 

What are ASE and OS versions?

 

Mandatory settings as per SAP

Refer SAP note 1539124 to ensure mandatory ASE Configuration are set properly

 

Memory Resources

Max memory = Total OS physical memory - (4 to 16 GB for OS tasks and management utilities) - (memory assigned to local SAP application servers)

Note: In any case, ensure ASE memory does not swap by checking other memory consumers carefully

 

Procedure Cache = 10 to 30% of max memory

It is main working memory of ASE (used for query compilation, small sorts and work tables, hash joins and holds cached query plans)

 

Statement Cache = 100 MB minimun, increase if number of query recompiles caused by plan flushes is high

 

Sort Buffers ~2000 for a single sort.This means a single sort can consume a maximun of 256 MB fro in memory sorting

 

If your system workload involves very large sort operations, increasing this number can  help improving sort performance.

 

HTH

Rajesh

↧
↧

Is there any sybase command or stored procedure to know the status of the job scheduler

November 11, 2014, 5:35 am
$
0
0

Is there any sybase command or stored procedure to know the status of the job scheduler

↧

Re: ISQL login error SYBASE

November 11, 2014, 4:04 pm
$
0
0

Hi Arne

 

What are the passwords for the users sapss and sapsso please? Getting that same problem of a new installation and cannot connect to the adaptive server. Please assist.

 

Regards

Rob

↧

Re: ISQL login error SYBASE

November 12, 2014, 12:18 am
$
0
0

Hi,
Just for information, on SAP BS/BW environment recommendation is to use the DBACockpit (t-code DB02 or dbacockpit) instead of Sybase Control Center. SQL code can also ben run from dbacockpit.

Also to keep in mind that on SAP BS/BW environments we will always need to specify the "-X" flag when using isql. The client password encription is required to be used (ASE configuration parameter "net password encryption reqd" should has a value > 0 ).
Best regards,
Victoria Normand.

↧
↧

Re: Sybase with ECC : Delete / Dump Transaction Log

November 12, 2014, 12:29 am
$
0
0

Hi Danny,
The dump of the transaction log should be smallest than the database dump. That size is very small I agree but the question is: were there some modifications done after the database dump and this dump transaction ? Anyway for more questions on that topic as you said it would be better to open a new thread and I guess this one could be marked as answered

 

Best regards,
Victoria.

↧

Extend DB

November 12, 2014, 7:33 am
$
0
0

Hello,

 

I want to extend database with a new datafile.

 

Each datafile has his own folder.

 

I've tried to create directly the folder in OS, but I want to looking for in path, It doesn't appear.

How can I manage this?

Captura.JPG

 

How can I set the autoextend db?

 

Thanks

↧

Re: ISQL login error SYBASE

November 12, 2014, 7:40 am
↧
Search

Re: Is there any sybase command or stored procedure to know the status of the job scheduler

November 12, 2014, 7:24 pm
$
0
0

Rajesh,

 

Please find the process to get ensured Job_Scheduler up and running.

 

use master

go

select * from sysprocesses where program_name like 'JS Agent%' or cmd = 'JOB SCHEDULER'

go

use sybmgmtdb

go

select * from js_jobs

go

select * from js_scheduledjobs

go

use saptools

go

select * from DBH_CONFIG

go

select * from ATM_CONFIG

go

 

 

 

To Start and Stop the Job Scheduler:

 

use sybmgmtdb

go

exec sp_sjobcontrol @name=NULL,@option="stop_js"

go

dbcc traceon(3605,3641)

go

exec sp_sjobcontrol @name=NULL,@option="start_js"

go

 

Regards

Kiran Kumar A

↧
↧

Re: Extend DB

November 13, 2014, 5:07 am
$
0
0

Hi.

 

The input help shows only the paths that are currently used by the database for datafiles. So for a new path you created, you will need to enter it yourself. It's an input field where you can type in any file system path.

 

For autoexpansion I would recommend to use a DBA Cockpit of version SAP_BASIS 7.02 SP14, 7.30 SP10, 7.31 SP09 or 7.40 SP04. Starting with this version there is a screen Configuration->Automatic Database Expansion. There you can set up rules that define your auto expansion strategy.

 

Best regards,

Jan


↧

R3load error on import with declustering

November 13, 2014, 5:17 am
$
0
0

Hi

We are installing CRM 7.0 EHP3 on Windows Sybase 16.00.2 and choosen declustering and depooling option.

During import Abap phase there are errors on DOKTL and EDI40 tables:

(DB) ERROR: DDL statement failed

(DROP INDEX "DOKTL"."DOKTL~0")

DbSlExecute: rc = 103

  (SQL error 3701)

  error message returned by DbSl:

[ASE Error SQL3701][SAP][ASE ODBC Driver][Adaptive Server

Enterprise]Cannot drop the index 'DOKTL.DOKTL~0', because it doesn't

exist in the system catalogs.

 

(IMP) INFO: a failed DROP attempt is not necessarily a problem

(DB) ERROR: DDL statement failed

(CREATE UNIQUE INDEX "DOKTL~0" ON "DOKTL" ( "ID", "OBJECT", "LANGU",

"TYP", "DOKVERSION", "LINE" ) with sorted_data )

DbSlExecute: rc = 99

  (SQL error 1581)

  error message returned by DbSl:

[ASE Error SQL1581][SAP][ASE ODBC Driver][Adaptive Server

Enterprise]Create index was aborted because pages in the data only

locked

table are not linked.

 

(DB) INFO: disconnected from DB

 

There is a note for this problem:

http://service.sap.com/sap/support/notes/2039613

but our initial R3Load was 742 PL13 and we upgraded to PL23 without any progress.

However in note is statement about R3Load 742 PL10 without "and above" so maybe only this version is OK (don't have access to PL10 to try ) ?

Opened OSS ticket and waiting .....

Regards

MArcin

↧

Re: Is there any sybase command or stored procedure to know the status of the job scheduler

November 13, 2014, 5:48 am
$
0
0

Hi,

 

Addition to the reply from Kiran you can somehow see the status of the JS in Sybase Central and SCC/Sybase (now SAP) Control Center gui's where the Stop or Start button is "highlighted". This gives you a clue if it is stoped or started, but the best way to be sure is to select from sysprocesses.

 

Regards!

 

Per Otto Westheim

↧

Re: ASE Sybase ODBC driver distribution question

November 13, 2014, 12:26 pm
$
0
0

These things need to be inserted into the registry.


Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SOFTWARE\ODBC]

[HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBC.INI]

[HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBC.INI\ODBC]
"TraceDll"="C:\\WINDOWS\\system32\\odbctrac.dll"

[HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBCINST.INI]

[HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBCINST.INI\Adaptive Server Enterprise]

"driver"="C:\\Sybase\\DataAccess64\\ODBC\\dll\\sybdrvodb64.dll"
"Setup"="C:\\Sybase\\DataAccess64\\ODBC\\dll\\sybdrvodb64.dll"
"APILevel"="1"
"ConnectFunctions"="YYY"
"DriverODBCVer"="03.51"
"FileUsage"="0"
"SQLLevel"="0"
"CPTimeout"="60"

[HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBCINST.INI\ODBC Drivers]
"SQL Server"="Installed"
"SQL Server Native Client 11.0"="Installed"
"Adaptive Server Enterprise"="Installed"

*******************************************************************

You also need to add the ODBC directory to the $PATH
The dll and the include directories need to be included in the package.
You also need to have the C:\Sybase\charsets directory included in the package.
The $SYBASE variable must also be set to the C:\Sybase directory.
That is so the charsets directory can be found.

 

Kevin

↧
↧

Re: R3load error on import with declustering

November 14, 2014, 7:33 am
$
0
0

Good to know that there is no support for ASE specific  problems in OSS ......

If any customer will ask for db recommendation i'll have answer .....

↧

Re: Is there any sybase command or stored procedure to know the status of the job scheduler

November 15, 2014, 12:42 am
$
0
0

Dear Kiran,

 

Thank you for your wonderful response.

↧
Search

Re: ASE SP 122 hitting SIGSEGV 11

November 15, 2014, 12:50 am
$
0
0

Dear Francis,

 

I agree with Bret. signal 11 is generally followed by a stack trace.

 

For Example:

00:0006:00000:00926:2014/10/07 00:43:53.34 kernel  Current process (0xa4101c6) infected with signal 11 (SIGSEGV)

00:0006:00000:00926:2014/10/07 00:43:53.34 kernel  Address 0x0x0000000000d915cb (dam_prepare_dump_end+0x890), siginfo (code, address) = (1, 0x(nil))

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel  **** Saved signal context (0x0x000000016366c300): ****

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel  uc_flags: 0x1, uc_link: 0x(nil)

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel  uc_sigmask: 0x416000 0xb 0x1 0x0

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel  uc_stack: ss_sp: 0x(nil), ss_size: 0x0, ss_flags: 0x2

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel  General Registers (uc_mcontext.gregs):

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel       PC : 0x0000000000d915cb (dam_prepare_dump_end+0x890)

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel          RAX : 0x0000000000003fe0  RBX : 0x0000000000000001

00:0006:00000:00926:2014/10/07 00:43:53.36 kernel          RCX : (nil)  RDX : 0x00000001b1983a28

 

Please raise a case with SAP Technical Support Team using your 'S' user ID

 

In general, Signal 11 error is a software bug and SAP should look into this.

 

HTH

Rajesh

↧

What is the best way to (automate)monitor the space growth of the databases in a Sybase server

November 15, 2014, 11:36 pm
$
0
0

Dear Experts,

 

I'm looking to create a stored procedure/SQL Query which the space details of all the databases in a server, with dbfree lessthan 15 % on top of the list

Output should be displayed as follows

db_name    db_size          db_free          db_usage%

 

Any help is highly appreciated.

If you have any better way to monitor the space growth of the databases in a Sybase server, I would love to hear from you.

 

Thank you in advance

Rajesh

↧
↧

Re: What is the best way to (automate)monitor the space growth of the databases in a Sybase server

November 16, 2014, 6:32 am
$
0
0

I would suggest you consider monitoring space usage at the segment level as opposed to the database level.

 

If your data and log segments are on different disks then monitoring at the segment level makes it easier to monitor for, and respond differently to, space issues in the log segment vs a data segment.

 

Consider what happens in the following scenario:

 

- DBA has created the myseg segment and assigned the MYTAB table to the myseg segment

- as rows are added to MYTAB the space in the myseg segment gets used up

- myseg fills up (100% full) but the database shows total space usage of only 40%

- inserts to MYTAB will fail (because the myseg segment is full) even though there's plenty of space elsewhere in the database

 

In this scenario if you are monitoring space usage solely at the database level you will never get an alert (only 40% full) while your application will cease to function (eg, cannot insert rows to the MYTAB table) due to the myseg segment being full.  On the other hand, if you are monitoring at the segment level you should get an alert before the myseg segment fills; said alert should give the DBA enough time to allocate more space to the myseg segment.

 

--------------

 

If you're looking for some sort of real time monitoring/alerting then I'd recommend you place multiple thresholds on all of your segments (data and log) and have the thresholdaction stored proc dump an appropriate message (eg, logsegment @ 87% used) to the ASE errorlog each time it fires.

 

An errorlog monitoring script/program can then generate alerts based on the contents of said errorlog messages.

 

--------------

 

If you're looking for long term (data) space usage monitoring (eg, trending), I'd just run a SQL script on a daily basis, store said results in a table, and generate alerts/reports off of said table.

 

As for SQL queries to obtain space usage data, I would suggest you take a look at the following stored procs:

 

     sp_helpdb

     sp_helpsegment

 

You could either a) parse the output from these stored procs or b) look at the source code for these procs to come up with your own SQL that slices-n-dices the data into a format you can use.

↧

Re: MASTER DB - dump and load (missing configurations)

November 17, 2014, 5:42 am
$
0
0

Hi Hans,

 

At the moment I'm at the step to set up a whole database, therefore I have to load all databases.

But after loading the master, found out that the configs are missing. So if the configs are missing, what else maybe is missing ..

 

 

Regards,

Tobias

↧

Re: What is the best way to (automate)monitor the space growth of the databases in a Sybase server

November 17, 2014, 5:55 am
$
0
0

Free space thresholds are configured within ASE for a given segment.  When a threshold is crossed ASE will automatically fire the user defined/created stored proc associated with the threshold.  What you do within the threshold is up to you (eg, for the logsegment you'll likely issue a 'dump tran'; for non-logsegment thresholds you might auto-extend the segment or simply 'print' a message ... and since there's no client/application front-end the message will go to the ASE errorlog).

 

See the following:

 

     sp_addthreshold

     sp_modifythreshold

     sp_dropthreshold

     sp_helpthreshold

 

     ASE System Admin Guide, Chapter: Managing Free Space with Thresholds

↧
Viewing all 3678 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>