From the error description my first thought was that this error is on a client and might have come from the db server, via TNS. I think there could be an error stack in the alert file of the database, but I don't believe that the TNS client would display more than the single TNS error. I may be wrong though. There may also be more info in $ORACLE_HOME/net/log/listener.log on the server and sqlnet.log (somewhere) on the client.
We put our Solaris 10 kernel variables still in /etc/system. For what it's worth, this is an example of settings set for one of our oracle 10R2 instances on Solaris 10:
* Kernel settings for Oracle set shmsys:shminfo_shmmax=4294967295 set shmsys:shminfo_shmmin=1 set shmsys:shminfo_shmmni=1024 set shmsys:shminfo_shmseg=1024 * set semsys:seminfo_semmap=1026 set semsys:seminfo_semmni=1024 set semsys:seminfo_semmns=16384 set semsys:seminfo_semmnu=2048 set semsys:seminfo_semmsl=300 set semsys:seminfo_semopm=200 set semsys:seminfo_semume=256 set semsys:seminfo_semvmx=32767 * set msgsys:msginfo_msgmap=512 set msgsys:msginfo_msgmax=8192 set msgsys:msginfo_msgmnb=65536 set msgsys:msginfo_msgmni=256 set msgsys:msginfo_msgssz=16 set msgsys:msginfo_msgtql=512 set msgsys:msginfo_msgseg=8192 * set noexec_user_stack=1 * * end snippet Cheers, Tony Mathias Magnusson wrote:
I believe Solaris now has these settings in project and processes.Still should there not be another message that goes along with this one to explain more about the details? I would think this is one of those that is just the top of the error stack and we ned more to understand the real reason.There could also be a hint about it in the massages file in Solaris (/var/admmessages I think) if it really is an OS issue. More than likely it is not there either as it is probably not considered an Os isue by Solaris, just an application resource use issue.MathiasOn Wed, Jul 15, 2009 at 3:12 AM, Tony van Lingen <tony_vanlingen@xxxxxxxxxxxxxxxxxxxxx <mailto:tony_vanlingen@xxxxxxxxxxxxxxxxxxxxx>> wrote:Perhaps the system settings (e.g. out of shared memory or semaphores)? They are in /etc/system. The minimum values should be in the Oracle 10g readmes or else in the platform specific installation instructions. Cheers, Tony nilesh kumar wrote:Hello DBA's Good Evening!! Hope every one is doing great!! Its 10g on solaris. We are getting this error when the Application team try to process data through ETL. But I did not find any errors related to this in the alert log. Has any one also faced this type. Is there any solution to resolve this issue. DPSTN_LTGTN,0: connect failed for env: null, user: , password: ***** Oracle error: -12549 ORA-12549: TNS:operating system resource quota exceeded Is this an error related to the OS issue.? or can we resolve this by resizing the memory? Please advise. Your help will be highly appreciated. Thanks Nilesh
