Quantcast
Channel: SCN: Message List - SAP Adaptive Server Enterprise (SAP ASE) for Custom Applications
Viewing all 3587 articles
Browse latest View live
↧

Re: How big to make Large IO cache pools ?

October 3, 2014, 11:36 am
$
0
0

"....So I guess we're reading a lot of pages from the large IO pool but very few of them."

 

If you meant to say that "we're reading a lot of pages into the large IO pool, but actually using very few of them".....the answer is "Yep!"

 

PagesRead is the number of pages actually read into cache....or if you do:

 

select * from master..monTableColumns where TableName='manCachePool'

 

You will see the description is:

 

"Number of pages read into the pool"

 

....now the indicators column suggest it needs to be delta'd - and looking at some code I wrote, it appears I was incorrect earlier and PagesTouched does as well.....so unless you had rollover (PagesTouched hit 2B), then you are in even worse shape than you thought.   Sorry about that - I was answering in a hurry earlier.

 

However, remember that any sample is just a point in time ....there may be cases where the entire cache is used.   Hence you should watch this over a period of a full day or a full week and baseline it.

↧
Search

Enabling SSL for Sybase 15.7 Cluster Edition

October 3, 2014, 1:39 pm
$
0
0

Bonjour,

  I am struggling to enable SSL on my ASE 15.7 Cluster Edition (SP122).

 

So far I have created the certificat, added it to the with sp_ssladmin.

 

I have updated the trusted.txt file in the config directory.

 

I have enabled ssl using sp_configure.

 

I have the licenses correclty installed and ASE_ASM checked.

 

I have update my interfaces file using ssl="cn=sybqa_grp".

 

I do not know if I am missing something here but there are no "config file" for cluster environment so I can't use the "-s" option to enable at startup and I can't find anything the like for the cluster.

 

When I try to start the listener using the following command :

 

sp_listener 'start','ssltcp:nodenade:4124:"CN=sybqa_grp"'

 

I get the following message :

 

Msg 15936, Level 16, State 1:

Server 'sybqa_grp', Procedure 'sp_listener', Line 107:

An error was encountered when starting a listener with protocol ssltcp, host nodename, port 4124.

Msg 15038, Level 16, State 1:

Server 'sybqa_grp', Procedure 'sp_listener', Line 107:

The specified port 4124 is invalid.

DBCC execution completed. If DBCC printed error messages, contact a user with System Administrator (SA) role.

(return status = 1)

 

Anyone having an idea as to why I am getting that error or why is that port not a good one?

↧
↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 6, 2014, 8:00 am
$
0
0

Hi William-Paul,

 

try:

sp_listener 'start', 'ssltcp:nodenade:4124:"CN=sybqa_grp"', "remaining"

 

Think you are missing the engine param.
You can also check the sp_listener status to see if you need to stop or start or if it is live.

 

Regards,
Ryan

↧

Re: How big to make Large IO cache pools ?

October 6, 2014, 9:37 am
$
0
0

Hi Jeff

I am reading your presentation about Cache and I have some questions:



You wrote

“

DO THE FOLLOWING ALL USE LARGE I/O??

Pop Quiz Answers

  1. A.

 

Large scans such as table scans, range scans & index leaf

scans (Yes)

  1. B.

 

Maintenance operations such as dbcc, update statistics

(Yes)

  1. C.

 

Create index (Yes)

  1. D.

 

Bcp

 

in on heap tables (Yes)

  1. E.

 

Select/into (Yes)

  1. F.

 

Insert/select (No)

  1. G.

 

Asynchronous Prefetch”

 

Interesting to me situations C&D.

Could you please explane me how I can use (for example 64kb poll) if my normall page size 8k.

 

I have some cache:

 

 

Cache Name      Status   Type      Config Value      Run Value

default data cache          Active   Default 12288.00 Mb     12288.00 Mb

log_cache           Active   Log Only                 800.00 Mb          800.00 Mb

tempdb1_cache              Active   Mixed     3072.00 Mb        3072.00 Mb

tempdb2_cache              Active   Mixed     3072.00 Mb        3072.00 Mb

                                                                  

                                                                  

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

                            Total    19232.0 Mb 19232.0 Mb                                                                    

==========================================================================                                                           

Cache: default data cache,   Status: Active,   Type: Default                                                            

      Config Size: 12288.00 Mb,   Run Size: 12288.00 Mb                                                      

      Config Replacement: strict LRU,   Run Replacement: strict LRU

      Config Partition:            4, Run Partition:            4

                                                                  

IO Size  Wash Size           Config Size         Run Size              APF Percent

    8 Kb       245760 Kb         9788.00 Mb        9788.00 Mb          10

   64 Kb       40960 Kb          2500.00 Mb        2500.00 Mb          10

                                                                  

                                                                  

==========================================================================                                                           

Cache: log_cache,   Status: Active,   Type: Log Only                                                           

      Config Size: 800.00 Mb,   Run Size: 800.00 Mb                                                                

      Config Replacement: strict LRU,   Run Replacement: strict LRU

      Config Partition:            2, Run Partition:            2

                                                                  

IO Size  Wash Size           Config Size         Run Size              APF Percent

    8 Kb        20480 Kb           100.00 Mb          100.00 Mb           10

   16 Kb      122880 Kb          700.00 Mb          700.00 Mb           10

                                                                  

                                                                  

==========================================================================                                                           

Cache: tempdb1_cache,   Status: Active,   Type: Mixed                                                                  

      Config Size: 3072.00 Mb,   Run Size: 3072.00 Mb                                                           

      Config Replacement: strict LRU,   Run Replacement: strict LRU

      Config Partition:            4, Run Partition:            4

                                                                  

IO Size  Wash Size           Config Size         Run Size              APF Percent

    8 Kb        36864 Kb          2712.00 Mb        2712.00 Mb          10

   16 Kb       24576 Kb           360.00 Mb          360.00 Mb           10

 

 

But when I create index I have next situation

 

 

 

    STEP 1

        The type of query is CREATE INDEX.

 

        TO TABLE

        dbo.tTest

        Using I/O Size 8 Kbytes for data pages.

 

 

What command or syntaxes use for  instructions what in operation bcp & create index use    64 Kb pool?

↧

Re: ASE certification on Windows

October 6, 2014, 11:13 am
$
0
0

ASE 15.7 ESD 4 and higher were certified on Windows 2012, but not explicitly on  2012 R2.

 

http://certification.sybase.com/ucr/platformDetail.do?certId=3610&platformId=133&productId=2

 

The certification group has reported that they've found some problems involving the existing installer accessing the Windows registry, so R2 is not yet certified due to these install issues.  We anticipate (but no guarantees) being able to certify the ASE 16.0 PL05 and 15.7 SP132 releases due to an update in the installer.

↧
↧

Re: Is there a getutcbigdatetime()

October 6, 2014, 11:42 am
$
0
0

Hi Mike,

 

I've opened a new feature change request, CR 771887.

 

-bret


↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 6, 2014, 12:00 pm
$
0
0

Bonjour Ryan, as usual, thanks for your precious time.

 

I finally found out that my certificates were not good and it is the reason why nothing is working.  I spent hours now trying to figure out how to create certificates but no matter how I create them, I always end up with the following error when using sp_ssladmin :

 

Warning: command 'addcert' adds certificate path /opt/sybase/ASE-15_0/certificates/sybqa_grp.crt with its common name field different from server name.

 

Here is the command I user to genrates the certificates :

 

  • /opt/sybase/OCS-15_0/bin/certreq -F /home/sybase/ssl_data/sybqa_grp/csr_server_request.txt -R /home/sybase/ssl_data/sybqa_grp/ca_request.txt -K /home/sybase/ssl_data/sybqa_grp/ca_pkey.txt -P abc123
  • /opt/sybase/OCS-15_0/bin/certauth -r -C /home/sybase/ssl_data/sybqa_grp/ca_request.txt -Q /home/sybase/ssl_data/sybqa_grp/ca_request.txt -K /home/sybase/ssl_data/sybqa_grp/ca_pkey.txt -P abc123  -O /home/sybase/ssl_data/sybqa_grp/trusted.txt

 

  • /opt/sybase/OCS-15_0/bin/certreq -F /home/sybase/ssl_data/sybqa_grp/csr_client_request.txt -R /home/sybase/ssl_data/sybqa_grp/sybqa_grp_request.txt -K /home/sybase/ssl_data/sybqa_grp/sybqa_grp_pkey.txt -P abc123
  • /opt/sybase/OCS-15_0/bin/certauth -C /home/sybase/ssl_data/sybqa_grp/trusted.txt -Q /home/sybase/ssl_data/sybqa_grp/sybqa_grp_request.txt -K /home/sybase/ssl_data/sybqa_grp/ca_pkey.txt -P abc123  -O /home/sybase/ssl_data/sybqa_grp/sybqa_grp.crt

 

  • cat /home/sybase/ssl_data/sybqa_grp/sybqa_grp_pkey.txt >> /home/sybase/ssl_data/sybqa_grp/sybqa_grp.crt

 

  • cp /home/sybase/ssl_data/sybqa_grp/sybqa_grp.crt $SYBASE/$SYBASE_ASE/certificates
  • cp /home/sybase/ssl_data/sybqa_grp/trusted.txt $SYBASE/$SYBASE_ASE/certificates/sybqa_grp.txt


Content of csr_server_request.txt :

req_certtype=server

req_keytype=RSA

req_keylength=2048

req_country=CA

req_state=mystate

req_locality=mycity

req_organization=myorg

req_orgunit=myunit

req_commonname=sybqa_grp_server (I tried with sybqa_grp)

 

Content of csr_client_request.txt :

req_certtype=server

req_keytype=RSA

req_keylength=2048

req_country=CA

req_state=mystate

req_locality=mycity

req_organization=myorg

req_orgunit=myunit

req_commonname=sybqa_grp

 

I don't understand why commonname is always dfferent from the server name :

1> select @@servername
2> go
                               
------------------------------
sybqa_grp                     

(1 row affected)

 

 

Any thoughts!?

↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 6, 2014, 12:42 pm
$
0
0

Ive ran into issues when the contexts are exactly the same it failed.
Have you tried editing the city/org/etc?


I would also use the same CN in both certificates.

Can you use the server name? usually this is an easier setup, but I haven't done this on Cluster Ed.
only ASE EE.

 

Also, this is a warning message if you run the lscert does it display?

If you use sp_listener 'start' and sp_listener 'status' does is this port listening?


Have you restarted the ASE?
Believe this file is only loaded on start.

 

Regards,
Ryan

↧
Search

message: Direct i/o enabled for ufs file

October 6, 2014, 1:56 pm
$
0
0

Hi All,

 

I Have an ASE Server  Adaptive Server Enterprise/15.5/EBF 19395 SMP ESD#5/P/Sun_svr4/OS 5.8/

 

 

I my ASE log I Have many intances of the following message:

 

00:25:00000:00916:2014/09/17 11:13:16.30 kernel  Direct i/o enabled for ufs file '/global/iva/log/dev_iva_log.dat'

00:11:00000:00397:2014/09/17 11:14:52.51 kernel  Direct i/o enabled for ufs file '/global/spm/log/dev_spm_log.dat '

00:57:00000:00277:2014/09/17 11:15:12.83 kernel  Direct i/o enabled for ufs file '/global/bidc/log/dev_bidc_log.dat '

00:10:00000:00258:2014/09/17 11:15:30.22 kernel  Direct i/o enabled for ufs file '/global/sef/log/dev_sef_log.dat '

00:23:00000:00386:2014/09/17 11:16:09.01 kernel  Direct i/o enabled for ufs file '/global/snc/log/dev_snc_log.dat'

00:13:00000:00450:2014/09/17 11:16:14.14 kernel  Direct i/o enabled for ufs file '/global/cam/log/dev_cam_log.dat'

 

 

this message represents a problem?

 

I found in the Sybase solved cases this case related with my issue: Solved Cases (#11783990) but when I try to open it, Does Not possible, Actually I have a user and password for this site

 

 

 

Thank you for your help

↧
↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 6, 2014, 5:03 pm
$
0
0

Ok, it's been a long day ...

 

All attempt to configure the Cluster with a self signed certificate have failed....  I use the exact same process, step for step for a regulat ASE EE instance and it works perfectly but for an unknown reason, unable to make it work for the ASE EE Cluster Edition.

 

I should have specified that first thing but I am running ASE EE Cluster Edition 15.7 SP122

 

I have requested a signed certificate from my corporate security team and it worked at first try.  Probelm fixed BUT and I need your help here, it seems sybcluster can't work through the SSL port but must work from non SSL port (as per Sybase doc) :

 

ERROR - sybqa_grp:sybqa_grp:SQLConnect:270:Login Failure - The user "sa" and the entered password is not authorized to connect to the cluster.

 

http://scn.sap.com/Verify that there is a non-SSL connection for each instance in the cluster and that this connection is included in the interfaces file Unified Agent uses.

 

I find it counter-intuitive having to leave non ssl port open so I can connect with sybcluster, anyone having knowledge of these ports will be able to connect non encrypted.  Is there something I miss here?

 

Further, I am little confuse about "Unified Agent".  To my knowledge, I do not use this "Unified Agent".  Agents are confugure through SCC and I manage the cluster using sybcluster.  Again, am I missing something?

↧

Re: ASE certification on Windows

October 6, 2014, 5:09 pm
$
0
0

Dear, Bret Halford.

 

Yep!! I've already seen that on certification site.

Do you know when will be certified?

 

If you know about that please let me know.

It will be helpful for me exactly.

 

Thanks for your reply.

 

Regards.
jinkyun.

↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 6, 2014, 5:30 pm
$
0
0

This is the information about unified agent.
SyBooks Online on cluster with unified agent and common commands.

 

Thanks,
Dawn Kim

↧

France - French "What's New" session : SAP ASE 16 & SAP IQ - 15/10

October 7, 2014, 2:17 am
↧
↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 7, 2014, 5:53 am
$
0
0

I've heard certreq certificates might not work on that new of an ASE.
They have added openssl to the builds I believe.

Have you tried creating the certificates with this utility?

 

Regards,

Ryan

↧

Re: Enabling SSL for Sybase 15.7 Cluster Edition

October 7, 2014, 8:23 am
$
0
0

@Dawn

 

Thanks, I will read all of this.

 

@Ryan

 

I was not satisfied with the result so last nite I have started all from scratch and remebered something you pointed out about the message being a warning and that the cert being read at startup.  After some cleanup and reboot I finally got it to work with a self signed.

 

At some point you get so deep in the work that you start doing stupid mistakes and don't see clearly ... !

 

Still, I do not understand the point of having non encrypted open ports for sybcluster to work.  When trying to secure your environment you try not letting backdoor open for anyone to use ... !

↧
Search

Re: C# BulkCopy only sometimes supports unsigned bigint fields

October 9, 2014, 11:19 am
$
0
0

In addition to Paul's request ..

 

- is the problematic column created as 'null' or 'not null'? what happens if you change the nullability?

 

- are you using a C# program that utilizes the bulk copy libraries? or are you using the 'bcp' command line tool?  if using C#/bulkcopy-libs, what happens if you use the 'bcp' command line tool?

 

- what is a sample data row that fails during bcp in?

 

- what happens if you use the INSERT command to insert a row, bcp the row out, and then bcp the row in?  success or failure?

↧

ASE MDA columns Hkgc*

October 9, 2014, 1:44 pm
$
0
0

In monOpenObjectActivity I have objects with in the 500K range in HkgcOverflows and HkgcPending.  This has occurred after a restart only a few days ago.  The counters for some objects have obviously wrapped around the 32-bit limit as they are around 2B.  From the scant information I have found so far, that indicates the Housekeeper Garbage Collection queue is building up work and sometimes overflowing its queue.  That sounds like something I would want to try and tune.  Has anyone know if this is something I should be concerned about?  If so, how to tune?  BTW, the queue length per processor in monEngine.HkgcMaxQize = 13360.  I can't find where that is configured if that is what I am supposed to tune.  Thanks!  Doug

↧
↧

Re: ASE MDA columns Hkgc*

October 9, 2014, 2:01 pm
$
0
0

Be aware that one of the *few* counters in ASE 15.7 that isn't too reliable is the HKGCPending - it quite often will be close to 2B - I don't think it is getting cleared/decremented correctly.

 

So, at a high level, each engine has an HKGC queue of 50000 entries.   As rows are deleted or other activity that requires HKGC action, an entry is made in that engine's queue.   This number of entries is not tunable.   If the queue overflows, then those entries are lost.   Typically the problem relates to DOL tables and deletes in which a deleted row is merely "marked" for deletion and it is up to the HKGC to actually remove the row from the page and if the last row, deallocate the page.   You can tune the HKGC by changing the 'enable housekeeper GC' setting - I recommend a value of '5' as a good starting point.

 

This will help, but there still are cases where there could be issues.   For example, a single batch job that deletes a billion rows.   No way they all fit in the queue and no way the HKGC can keep up....result is HKGCOverflows - which points to a likely need to run reorg to reclaim the space.   If 15.7 sp 100+, reorg rebuild online is best option or reorg defragment (sp110??) which is an online option as well.    The problem with other reorg options is that if you read the docs closely, you will note that if they hit a forwarded row, they escalate to a table lock until that extent is finished (reorg forwarded_rows, etc. work on an extent basis for transaction sizes) - so it can literally do hundreds or thousands of escalations to table locks.

↧

Can't kill process created by Sybase Java

October 10, 2014, 12:44 am
$
0
0

Hi ,

I havae some problem with new application worked with Sybase Java.

When I reboot application worked with Sybase Java I  see in sysprocesses next

 

1015171206488927recv sleep  703TestjTDS123AWAITING COMMAND1060529040[NULL][NULL]40960EC2MEDIUMANYENGINE041[NULL]000########111.111.111.111[NULL]

 

 

and when I exec

"kill 1015 " i haven't rezult.

I exec select on sysprocesses  and see that spid.

 

Maybe somebody had similar situation?

how kill this process, because situation repeat only woth process Java.

Version Sybase ASE 15.5 ESD 5.1

↧

Re: C# BulkCopy only sometimes supports unsigned bigint fields

October 10, 2014, 1:06 am
$
0
0

> You mention C# and bcp. I think you are using bcp utility?


No I'm using C#.

I'm using SDK SP122 (I forgot to include this).


See SyBooks Online AseBulkCopy Class.

↧
Viewing all 3587 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>