if you're on 15.7, than use: deferred name resolution
set it to 1 (disabled by default), and no need to create the temp table anymore outside the proc
object resolution will only be performed when the proc is executed, not when it's created
↧
Re: how to use temp table in inside a procedure called?
↧
Re: LIST aggregate function
Bret,
Could you provide us with the status of the CR 570337 or some news about it?
My developers are asking me the same question as Denis.
Vincent
↧
↧
Re: LIST aggregate function
The status is unchanged. Engineering isn't likely to work on it unless they become aware of more interest in the feature.
↧
Re: LIST aggregate function
Thanks Bret!
Could it be possible for us to get the code of list() in Sybase IQ or Anywhere and adapt it for ASE?
If yes, how could we do that?
Vincent
↧
Re: Sybase ASE Small Business editions on vmware 5.1 environments?
It seems that this bug has been fixed in ASE SP100!
Running Adaptive Server on VMware vSphere 5.0
Is it?
↧
↧
Re: Events Cluster Edition
Hi Mark,
Thanks for your collaboration I am going to study this document and I am going to apply your recommendations for the next event.
Cristian,
↧
I'm having problems reaching SCC for over a week now...
For more than a week back, I cannot reach my Sybase Control Center anymore, no matter what browser I use !
Normally I'm using Firefox, and I just thought that the certificate had expired or something, but I keep getting no data back from my SCC server.
I reinstalled SCC without any change...
I also started to tcpdump the traffic on my SCC server just to make sure it wasn't any firewall problem, but the web session actually reach my SCC server, so it must be something else ?
I have now tried with IE, Firefox and Chrome, but neither of them work
I have SCC running on two servers; Production and Test and it's the same problem against both of them
And yes, I have logged a case to SAP/Sybase support, but it's now gone a week and they haven't even picked it up yet :-/
Anyone out there that has experienced this and know how to fix it ?
Please advice
All the best
Mike of Sweden
↧
Physical IO Cost vs Logical IO Cost
As far as I know the ASE optimiser estimates a physical page read as 18 and a logical page read as 2.
This suggests the cost of a physical read is only 9 times greater than a logical read.
In the real world (or at least my world) a physical page is between 2ms and 18ms depending on the SAN tier, what time of day it is, etc
And a logical page is about 0.05ms. So at best a physical page read is about 50 times slower and at worse 300 times slower.
Shouldn't we adjust these values for the world as it is now ? ie logical read should be set to 2 and a physical read should be set between 100 and 300.
Is it possible to change these values to something reflecting our own environments ?
I don't think these timings are that unusual.
↧
Re: Physical IO Cost vs Logical IO Cost
In more recent versions, these ratios (and they are simply ratios, that is, number relative to each other as you point out) are configurable, and YES, you should configure them to the best of your knowledge of your environment:
1> sp_configure 'cost'
2> go
Parameter Name Default Memory Used Config Value Run Value Unit Type ------------------------------ ----------- ----------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- -------------------- ---------- cost of a cpu unit 1000 0 1000 1000 number dynamic cost of a logical io 2 0 2 2 number dynamic cost of a physical io 25 0 25 25 number dynamic
In older versions of ASE, you can use "dbcc tune(wlogical | wphysical , <n>)", but these settings weren't persistent across ASE reboots, so had to be done every time you start ASE.
↧
↧
Re: Physical IO Cost vs Logical IO Cost
Thanks - I see the ratio is now 2 to 25.
As its dynamic I'll see what happens when I change it.
Are these parameters only used when it compiles a proc ?
How does it use these values ? Is it looking in the cache to see what's there ?
↧
Re: Physical IO Cost vs Logical IO Cost
These numbers are used by the optimizer when costing query plans. That can be for proc compiles, SQL statement compiles, triggers, etc. The optimizer uses a great number of factors and statistics to determine how many physical ios, and cache io's a query/statement will need. Then uses these multipliers (along with other stats) to give each possible query plan a "cost". The plan with the lowest cost is chosen as the ultimate query plan to cache and execute.
A very detailed explanation can be found in the product manuals, or some decent google searches.
Here are some starting points:
↧
Extracting the C# ADO.NET drivers without installing
We've been using the SDK SP122 drivers but have discovered they're not thread-safe at all.
We can see this is fixed in SP130 but we have problems getting SP130 to work.
I've downloaded all the SDKs between 122 and 130 but I can't easily install them - I just want to get the dll and test it.
Is there a way to extract the files without an install ?
Alternatively does anyone know when the SDK was first made thread safe ?
Thanks
Also, the install log says
Status: ERROR
Additional Notes: ERROR - Language Modules: FAILED. See log file C:\UBS\Dev\Temp\ASEADOSDK\Install15.7SP125\log\lang.log
Status: ERROR
Additional Notes: ERROR - Open Client: FAILED. See log file C:\UBS\Dev\Temp\ASEADOSDK\Install15.7SP125\log\open_client.log for detail.
But there is no lang.log or open_client.log
How do you get the install to produce the log files ?
↧
Re: I'm having problems reaching SCC for over a week now...
Hi Mike,
what http response you are getting via browser? only "Site not reached"?
Is SCC host pingable and can you connect via telnet on SCC port?
Can you also post the last entries in SCC log? (scc-service.out / gateway.log)
May be your SCC is running out of memory.
Regards,
Jens
↧
↧
Cluster agent for SAP ASE HA
Hi
I have a query on SAP ASE HA. Can someone help me?
We have a plan to use SAP ASE HA on Sun cluster. Does SUN cluster provides cluster agent for SAP ASE or do we get the agent scripts from ASE HA package? I understand there is no ASE agent available on latest SUN cluster ver 4.1. Please advise.
thanks
CS
↧
Re: in-memory tempdb feature
:-( No Reply........
I will post my findings
↧
SYSAM Configuration on 2 (Windows)servers
Hi there,
We would like to install SYSAM on 2 (Windows) servers for failover purposes. Both servers will be online due to other License software.
My question is: How about the on HOST-ID generated licenses file? This will work on the designated HOST but how about the other one?
The licences are for ASE, REP and PowerDesigner Architect.
Please advise.
Regards.
Harry Splinter
↧
Re: I'm having problems reaching SCC for over a week now...
Hi Jens,
when running Firefox at the moment against SCC, I get the message "The connection was reset"
The host is pingable, yes
There is no entry in either of these two logs, they are completly empty both of them
I do get some output however, when I try to connect from Firefox and start a tcpdump session on that port number on my SCC server. So the firewalls are open at least
Regards
Mike
↧
↧
Re: I'm having problems reaching SCC for over a week now...
Hi Mike,
"The connection was reset" tells us that there is a response from the server regarding the port / service.
The logs are empty - strange. Are all logs are empty or only the named ones? When was the last restart of SCC? Have checked system log? On which OS you are running the SCC?
Have you added the SCC server to the trusted zone / local intranet which can be defined in the internet settings on IE? Sometimes there are issues regarding the trust of SSL certificate.
If it is an issue regarding memory leak, you can try to increase its allocation limit in the start script (default scc.sh on unix):
increase the value of variable DEFAULT_MEM_PERM (should be located in line 20-30). Default is 128MB.
Afterwards you have to restart the SCC.
Regards,
Jens
↧
Re: I'm having problems reaching SCC for over a week now...
Hi Jens,
thanks for taking your time !
Not all logs are empty, the repository and agent logs are populated. But shows no error or problem.
The last restart was made by me last week (thursday I think).
Yes I have checked the system logs, but nothing strange there either
I am running on this current Linux version :
Linux version 2.6.32-279.el6.x86_64 (mockbuild@x86-008.build.bos.redhat.com) (gcc version 4.4.6 20120305 (Red Hat 4.4.6-4) (GCC) ) #1 SMP Wed Jun 13 18:24:36 EDT 2012
Yes it has been added to the trusted zone !
I believe myself that this is some kind of SSL certificate issue ? But I have no idea how to verify or trace or chasing down this ? Running against it was no problem before, I have been using it for several years here, monitoring our replication environment. I think "something" has happened in our net regarding the SLL Certificate's ?
Thanks
Regards
Mike
↧
Re: SYSAM Configuration on 2 (Windows)servers
Hi Harry,
When generating licenses for a SYSAM license server, the license need to of served type, and the you have to fill in all 2 or three host-ids for the hosts you have.
there is a Sysam manual on infocenter.sybase.com
Hans
↧