Rotate Listener log on Windows

Can someone please tell me how to rotate the listener.log without bouncing the listener on a Windows machine.
We have unix kit installed on our servers and I did:
D:\oracle\product\10.1.0\Db_1\NETWORK\log>pwd
d:/oracle/product/10.1.0/db_1/network/log
D:\oracle\product\10.1.0\Db_1\NETWORK\log>ls -lart
total 577743
drwxrwxrwx 0 Jun 24 2005 ../
-rw-rw-rw- 0 Jul 16 16:53:02 sqlnet.log
-rw-rw-rw- 288,468,458 Aug 09 18:38:13 Copy of listener.log
-rw-rw-rw- 0 Aug 09 18:44:11 a.txt
drwxrwxrwx 0 Aug 09 18:44:11 ./
-rw-rw-rw- 0 Aug 10 11:21:33 listener.log
D:\oracle\product\10.1.0\Db_1\NETWORK\log>cat a.txt > listener.log
The process cannot access the file because it is being used by another process.
D:\oracle\product\10.1.0\Db_1\NETWORK\log>
It works well in linux environment but here in windows its not working. Can someone please tell me how to rotate it in Windows environment.
Regards

Hi
I tried the following and it worked on another server.
C:\> cd \oracle\ora92\network\log
C:\oracle\ora92\network\log> lsnrctl set log_status off
C:\oracle\ora92\network\log> rename listener.log listener.old
C:\oracle\ora92\network\log> lsnrctl set log_status on
But name of my listener is LISTENER_1. So these commands are not working here.
Regards
Message was edited by:
Singh

Similar Messages

Example] Shell Script for rotate listener log (V0.2)

create file /home/oracle/sbin/rtlstenlog.txt
#configure for rtlstenlog.sh, OHOME=ORACLE_HOME, LISTEN=LISTENER NAME,NFILE=listener log
OHOME=/oracle/product/10.1.0/db
LISTEN=LISTENER_CTPPDB01
NFILE=listener_ctppdb01.log
and create script file
#!/bin/sh
#V0.2 for every listener name
# refer http://surachart.blogspot.com
source /home/oracle/sbin/rtlstenlog.txt
export ORACLE_HOME=${OHOME}
DATE=`/bin/date +'%Y%m%d'`
DIR==${OHOME}/network/log
ODIR=${OHOME}/network/log/old
FILE=${OHOME}/network/admin/listener.ora
if [ -f ${FILE} ]
then
LOGFILE=${OHOME}/network/log/${NFILE}
if [ ! -d ${ODIR} ]
then
/bin/mkdir ${ODIR}
fi
if [ -f ${LOGFILE} ]
then
/bin/mv ${LOGFILE} ${ODIR}/${NFILE}.${DATE}
${OHOME}/bin/lsnrctl << EOF
set current_listener ${LISTEN}
set log_status off
set log_status on
exit
EOF
/bin/gzip ${ODIR}/${NFILE}.${DATE}
fi
fi
test
[oracle@CTPPDB01 sbin]$ ./rtlstenlog.sh
LSNRCTL for Linux: Version 10.1.0.4.0 - Production on 05-APR-2006 16:13:17
Copyright (c) 1991, 2004, Oracle. All rights reserved.
Welcome to LSNRCTL, type "help" for information.
LSNRCTL> Current Listener is LISTENER_CTPPDB01
LSNRCTL> Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=DB01_VIP)(PORT=1521)))
LISTENER_CTPPDB01 parameter "log_status" set to OFF
The command completed successfully
LSNRCTL> Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=DB01_VIP)(PORT=1521)))
LISTENER_CTPPDB01 parameter "log_status" set to ON
The command completed successfully
ls -la /oracle/product/10.1.0/db/network/log/old/
total 64
drwxr-xr-x 2 oracle dba 4096 Apr 5 16:13 .
drwxr-xr-x 3 oracle dba 4096 Apr 5 16:13 ..
-rw-r----- 1 oracle dba 52564 Apr 5 16:13 listener_ctppdb01.log.20060405.gz
Message was edited by:
HunterX (Surachart Opun)

Hi again,
so far I've constructed this:
cd /input
for file in `ls *log`
do
echo "Copying " $file
cp $file /output/log.$file.`date +%d%m%y`
done
Quite there. I'm cding to directory, i'll focus later how to set some environmental parameters.
Problem with zipping is that I don't know exact logfiles names. Each day there's a new name. So zip -q filename.zip filename.log isn't quite enough and it's impossible to use wildcards. The switch -q and checksums were welcome suggestions.
Thanks again,
Juha
Message was edited by: J.Otava

Is it safe to compress Listener logs while system is running ???

Hi there,
I am new to RAC and am working on 24*7 databases in a RAC CLuster.
There is a very large listener_scan2.log in the /u01/11.2.0/grid/log/diag/tnslsnr/<server name>/listener_scan2/trace folder, the file is about 2.8 G.
I have done some research and have seen that compress -f has been recommended via crontab, this has the advantage of being something that can be automated and run monthly
e.g.
## listener log aging
17 0 * * 0 /usr/bin/compress -f /u01/app/grid/diag/tnslsnr/node1/listener/trace/listener.log
17 0 * * 0 /usr/bin/compress -f /u01/app/11.2.0/grid/log/diag/tnslsnr/node1/listener_scan1/trace/*log
17 0 * * 0 /usr/bin/compress -f /u01/app/11.2.0/grid/log/diag/tnslsnr/node1/listener_scan2/trace/*log
17 0 * * 0 /usr/bin/compress -f /u01/app/11.2.0/grid/log/diag/tnslsnr/node1/listener_scan3/trace/*log
Is this safe to run these while the system and listener are up and running, will the listener create a new listener.log file if the old one is compressed even if listener is not restarted.
Other blogs suggest using lsnrctl and changing the listener log location and then renaming/compress/delete the old log file and then set the listener log location back again ...
Has anyone used this in a Production system ?
I welcome your feedback.
many thanks,
Rob

http://danthedba.blogspot.in/2009/11/manage-listener-log-files-in-11g.html
http://msutic.blogspot.in/2009/06/truncating-rotating-flushing.html
https://kkempf.wordpress.com/2011/06/10/rotating-listener-logs-without-bouncing-the-listener/

Windows bactch job to trim the Listener log

Hi All ,
Can you help me to write a Windows batch job to trim the listener.log on periodically.
I have the below steps, which i do manually but i won't which needs to be put in batch.
copy listener.log listener_<today>.log
lsnrctl set log_file listener_new.log
del listener.log
lsnrctl set log_file listener.log
lsnrctl reload
del listener_new.log
Appriciate the Help.
Thanks in Advance.

Have a look here:
http://www.wikihow.com/Write-a-Batch-File
You have the commands, so you only need the batch file around it
HTH
FJFranken

RMI listener stops whenever the user logs off Windows

Hi guys...
I have a problem whereby the RMI listener stops whenever the user logs off Windows.
I've gone through the net and the only solution was to add "-Xrs" to the end of the "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\IBM Tivoli Identity Manager Adapter" service for the ImagePath statement.
Unfortunately for me, that did not work.
Anybody knows how I can go around fixing this.
Much appreciated.

What's the RMI listener?
Do you mean the RMI Registry?
In which case the solution is to add that same flag when you run the registry.

TNS-12505 in listener.log when opening SqlDeveloper 2.1 (maybe a bug?)

Hi I've a strange problem:
1 download sqldevloper
2 create a connection using Connection type= TNS. Is select an entry in my tnsnames.ora to connect to an istance X (service_name is specified) on server Y. The first strange thing is that every entry in the tnsnames is displayed two times.
3 If I test the connection it works..
4 Close sqldeveloper
5 When I open sqldeveloper (just open the program without connect to anything) i receive no errors but i see in my listener.log (of the server Y) :
04-DIC-2009 11:39:31 * (CONNECT_DATA=(SID=null)(CID=(PROGRAM=)(HOST=__jdbc__)(USER=null))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.168.8.105)(PORT=2285)) * establish * null * 12505
TNS-12505: TNS: il listener non ha risolto SID indicato nel descrittore di connessione
192.168.8.105 is my ip address
I don't understand why it uses SID= instead of SERVICE_NAME=
If i place the mouse cursor over the connection icon (without click) i see an error TNS-1252 with the connection desripter: Y, 1521,NULL (i think that this null is causing all this problems) .
Why? When sqldeveloper opens it tries to test the connection?
Even if it is like this the connection should work: if I try to connect whit that connection everything works fine.
I do something worng? It is a bug?
With SqlDeveloper 1.5 everything works fine. Everything is the same so I guess that is a problem of the new version (2.1).
Oracle client oracle 10.2, SqlDeveloper Version 2.1.0.63, Windows XP Professional Service Pack 3
Hope someone can help me, if something isn't clear I can post more details (just tell me what you need).
Thanks, best regards.
Adriano Aristarco

Hi Adriano,
"network alias"
A network alias that has been found by SQLDeveloper java (as opposed to the jdbc driver) in tnsnames.*.
"connect identifier"
Uses oci 'thick' driver i.e. oracle C code find the tnsnames entry for you - requires additional Oracle client side software such as instant client. Without the additional software you get 'no ocijdbc11 in java.library.path'
Do you have multiple entries due to multiple tnsname.* files (move the inappropriate tnsname.* to a different backup directory). If you picked the 'wrong one' i.e. one with the right name but from an old tnsnames.* file, it might not be able to connect so you get error messages.
You cannot use "connect identifier" unless you have the additional Oracle client side software (I can test the latest instant client to confirm that works if you want to try this option, I do not have the version numbers for this in easy reach).
-Turloch

Errors in alert log and listener log and "alter database mount exclusive"

Hello!
I need a help.
Database 11R2 works under MS Windows Server.
Whwn I start it using Services, according alert log it is started by command "alter database mount exclusive".
Next - alter database open.
After this, it seams that program, which should put data into database, can not work with it, because I see errors in alert log: ora-12537, 12560, 12535, 12570, 12547.
What does itmean and what to do?
This is extract from alert_log
ORACLE_BASE from environment = C:\Oracle
Mon Feb 04 14:54:53 2013
alter database mount exclusive
Successful mount of redo thread 1, with mount id 1458539517
Database mounted in Exclusive Mode
Lost write protection disabled
Completed: alter database mount exclusive
alter database open
Thread 1 opened at log sequence 3105
Current log# 3 seq# 3105 mem# 0: C:\ORACLE\ORADATA\xxx\REDO03.LOG
Successful open of redo thread 1
MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
SMON: enabling cache recovery
Successfully onlined Undo Tablespace 2.
Verifying file header compatibility for 11g tablespace encryption..
Verifying 11g file header compatibility for tablespace encryption completed
SMON: enabling tx recovery
Database Characterset is AL32UTF8
No Resource Manager plan active
Mon Feb 04 14:55:04 2013
replication_dependency_tracking turned off (no async multimaster replication found)
Starting background process QMNC
Mon Feb 04 14:55:06 2013
QMNC started with pid=20, OS id=2860
Completed: alter database open
Mon Feb 04 14:55:11 2013
Starting background process CJQ0
Mon Feb 04 14:55:11 2013
CJQ0 started with pid=25, OS id=2000
Mon Feb 04 14:55:11 2013
db_recovery_file_dest_size of 4977 MB is 0.00% used. This is a
user-specified limit on the amount of space that will be used by this
database for recovery-related files, and does not reflect the amount of
space available in the underlying filesystem or ASM diskgroup.
Mon Feb 04 15:00:29 2013
Starting background process SMCO
Mon Feb 04 15:00:29 2013
SMCO started with pid=32, OS id=3212
Edited by: kogotok1 on Feb 4, 2013 4:54 PM

Thank you.
But in the same time - when I see in alert log those error messages ora -12560, 12537,12535,12570 and so on - clients programs, whiie try to connect, hang up.
For sql plus takes 20 minutes to connect.
lsnrctl status gives the following
LSNRCTL for 64-bit Windows: Version 11.2.0.1.0 - Production on 04-FEB-2013 16:01
:46
Copyright (c) 1991, 2010, Oracle. All rights reserved.
Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=xxx-BD.mosxxx
.elektra.net)(PORT=1521)))
STATUS of the LISTENER
Alias LISTENER
Version TNSLSNR for 64-bit Windows: Version 11.2.0.1.0 - Produ
ction
Start Date 01-FEB-2013 10:22:48
Uptime 3 days 5 hr. 39 min. 54 sec
Trace Level off
Security ON: Local OS Authentication
SNMP OFF
Listener Parameter File C:\Oracle\listener.ora
Listener Log File c:\oracle\diag\tnslsnr\xxx-BD\listener\alert\l
og.xml
Listening Endpoints Summary...
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=xxx-BD.mosxxx.elektra.net)
(PORT=1521)))
(DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(PIPENAME=\\.\pipe\EXTPROC1521ipc)))
Services Summary...
Service "CLRExtProc" has 1 instance(s).
Instance "CLRExtProc", status UNKNOWN, has 1 handler(s) for this service...
Service "XXX" has 2 instance(s).
Instance "XXX", status UNKNOWN, has 1 handler(s) for this service...
Instance "xxx", status READY, has 1 handler(s) for this service...
Service "XXXDB" has 1 instance(s).
Instance "xxx", status READY, has 1 handler(s) for this service...
The command completed successfully
To tell the truth I am confuse - I thought I have only 1 service "XXXDB" and 1 instance - "xxx".
May be I have wrong entries in tnslsnr.ora?

ORA-12535 error but TNSPing Works and listener logs connection

We recently migrated our firewall and then started seeing this problem for users outside of our own subnet. However, although the user gets this error when launching the application, the workstation is able to TNSPing the server successfully and listener.log does log the connection.
I was hoping someone might have some ideas or suggestions as to what might be causing the error despite the apparent connectivity to the database.
Thanks in advance.

Thanks so much for the suggestions.
1. telnet <host> 1521 connects
2. sqlplus <username>/<pwd>@<instance> returns the ORA-12535 error
Finally, the sqlnet.log on the client side doesn't log anything when launching the application so I have no information as to what exactly is occurring at that point. However, the sqlplus command above logged the following:
VERSION INFORMATION:
TNS for 32-bit Windows: Version 9.2.0.1.0 - Production
Windows NT TCP/IP NT Protocol Adapter for 32-bit Windows: Version 9.2.0.1.0 - Production
Time: 26-JAN-2007 08:43:41
Tracing not turned on.
Tns error struct:
nr err code: 0
ns main err code: 12535
TNS-12535: TNS:operation timed out
ns secondary err code: 12560
nt main err code: 505
TNS-00505: Operation timed out
nt secondary err code: 60
nt OS err code: 0
I've been doing more research on this -- please note that I'm not an Oracle dba so I am just getting familiar with lots of things -- and have learned that user connect requests come in through 1521 but are then randomly redirected to other ports. Given the new firewall, I'm wondering if it's an issue that the random ports being selected aren't open for the users outside of our subnet. Does this sound like a plausible explanation? If so, how do administrators generally handle this -- by opening the necessary ports on the firewall? Is there some other standard approach?
Again, thanks so much for any assistance you can provide.

Regarding listener.log file

Hi,
The listener.log file is showing the following code snippet.But when I put this code to my source file (Automation Plugin), it is giving error as 'FullPathToAlias' unrecognised symbol.
// Move this to the top of the routine!
Handle aliasValue = NULL;
FullPathToAlias("C:\Test.jpg", aliasValue);
Do I have to include any header file? Does any one have any idea? I am just wanting to dispaly an image preview on dialog box(Automation plugin).
Please give any idea how can I get the image referrence for the first time from the Hard Disk?
Can anyone help me out.....................?
Thanks and Regards

Mmh, so you're trying to do something like a file-preview?
I don't know if this is possible, mainly because at the moment
I can see no way getting any image data before you've loaded
the image in photosphop (with just using the sdk and not any
mac/win api)
If you're developing only for windows targets, you might want to
try to develop a mfc type plugin like the "mfcplugin" from the filter-
plugin examples. It does give a preview of the result of an filter
applied to an image, so I'd guess you can change this with the help of the win-api to preview some picture file. For MFC there are also some
nice solutions on codeproject you could easily integrate in your plugin:
a thumbnail control:
http://www.codeproject.com/miscctrl/thumbnailctrl.asp
and a picture control:
http://www.codeproject.com/cs/miscctrl/ScalablePictureBox.asp
Otherwise the "dissolve"-plugin from the filters example is for mac and win os's and offers also a preview (of a filter application), maybe this will give you some more insight.

Turning off listener.log?

I have a huge listener.log file (around 9GB) and it is causing storage issues. Is it safe to turn the logging off in the lstnctl? Another alternative is to turn it off, rename the listener.log file and then turn back on and archive the old listener file?
Also, having a file this big on a Unix system, is it possible that turning the logging off will increase system performance?
Thanks,

You can 'rotate' the listener log file by using a script, like
#!/usr/bin/sh
#set -vx
msg_exit()
sts=$1
shift
echo "$@" >&2
exit $sts
TNS_ADMIN=${TNS_ADMIN:-${ORACLE_HOME}/network/admin}
LISTENERORA=$TNS_ADMIN/listener.ora
[[ ! -f "$LISTENERORA" ]] && msg_exit 1 "${LISTENERORA} not found. exiting"
# bepaal de naam van de listener
lsnr_name="`awk '/^LISTENER/{print $1}' $LISTENERORA | tr [A-Z] [a-z] `"
echo $lsnr_name
# bepaal de naam van de log file
logfil="`awk '/LOG_FILE/ {print $NF}' $LISTENERORA`"
logfil=${logfil:-listener.log} # defaults to listener.log when not set
# bepaal de naam van de log directory
logdir="`awk '/LOG_DIRECTORY/ {print $NF}' $LISTENERORA`"
logdir=${logdir:-${ORACLE_HOME}/network/log} # defaults to $ORACLE_HOME/network/log when not set
TIMEST="`date +%Y%m%d%H%M%S`"
# bepaal of de listener draait
if (ps -ef | grep "\"tnslsnr $lsnr_name\"" | grep -v $$ >/dev/null)
then
# listener draait
echo "$lsnr_name $logdir/$logfil"
echo "now stopping ${lsnr_name}"
lsnrctl stop $lsnr_name
echo "mv ${logdir}/$logfil} ${logdir}/${logfil}.${TIMEST}"
mv $logdir/$logfil ${logdir}/${logfil}.${TIMEST}
touch ${logdir}/${logfil}
echo "Now starting listener $lsnr_name"
lsnrctl start $lsnr_name
fi
Sybrand Bakker
Senior Oracle DBA

Listener Log in 11gR2

I have located my listener.log my using lsnrctl stat <listener name>
my particular log is in $ORACLE_HOME/log/diag/tnslsnr/<host name>/<sid>/alert/log.xml
Where is this log file and location actually specified ?
For the same lsnrctl stat command from above, I can see my listener.ora config file. However the above listener log location is not specified in it. In fact LOG_DIRECTORY_<SID> actually points to $ORACLE_HOME/network/admin - however there are absolutely no log files there at all !
The listener.ora does specifiy a listener_ifile.ora, however at 2 bytes in size, this file appears to be empty
Anyone able to make sense of how 11gR2 is now doing this ?
Jim

Try in directory:
$ORACLE_HOME/log/diag/tnslsnr/<host name>/<sid>/traceOr try to use SHOW TRC_DIRECTORY to get listener log directory:
c:\>lsnrctl
LSNRCTL for 64-bit Windows: Version 11.2.0.1.0 - Production on 11-JAN-2012 12:07:07
Copyright (c) 1991, 2010, Oracle. All rights reserved.
Welcome to LSNRCTL, type "help" for information.
LSNRCTL> set current_listener listener
Current Listener is listener
LSNRCTL> show trc_directory
Connecting to (ADDRESS=(PROTOCOL=tcp)(HOST=)(PORT=1521))
listener parameter "trc_directory" set to c:\app\oracle\diag\tnslsnr\xxx\listener\trace
The command completed successfullyEdited by: P. Forstmann on 11 janv. 2012 12:03

Need to Create a New Listener Log file 9i

Hi, I have a issue with a oracle 9i database.
First the listener.log file is 1.2G not good on a 25G server..
How do you create and delete the listener file? Windows 2000 OS...
Second I also have an issue with the archive log files... I have about 150 file 102M each...Again Not good...
Need HELP drowning in a sea of oracle... Given the Shrink command....
At least the Listener issue first then the archive log problem..
Thank you...

carlo_sc wrote:
yes, we normally do a full daily backup and the archive logs are then deleted..But the Job didn't finish
last night. Also we had big Network issue yesterday and I believe that caused the issue with this server
causing an excessive amount of log files to be created...
So with simply thing with lister is just too:
1) stop listener service
2) delete Log file.... (that won't mess anything up on restart ) I'm very skittish about manually deleting stuff from oracle..
Got burned in Sept of last year big timeIt's just a log file. It's for your own use if needed. oracle doesn't use it or require it. You even have an option in the listener to not even write to it in the first place.
3) restart listener service
The archive logs : What if I move some of the files to another location then perform the a full daily backup will the that
solve my archive log issues....
Thanks for all the quick responses
Edited by: carlo_sc on Jan 15, 2010 10:09 AMYou say you already have a backup job that gets the archivelogs and deletes them after backup? And the only real problem here is that it failed? Fix the failure and run it again. You can, in a pinch, manually move the archivelogs, but then YOU are responsible for keeping up with them.
Edited by: EdStevens on Jan 15, 2010 9:32 AM

Does listener log file need to be purged in Oracle 11g DB

We have the Oracle 11g database setup. I have a query as to if it is required to purge the listener log file in Oracle 11g database. I believe that the size of the listener log file is infinite and we need to manually purge the listener log file or is there an automatic log rotation of the listener log file?
I hope my query is clear that if it is required to manually purge the listener log file or is there an automatic log rotation for this file.
Please revert with the reply to my query.
Regards

972145 wrote:
Thanks for your answer but does it mean that there is no automatic log rotation for the listener log file in Oracle 11g database?
Regardshttp://uhesse.com/2011/06/01/adrci-a-survival-guide-for-the-dba/
set the listener home inside adrci and do
purge -age 2880 -type tracepurges all tracefiles older than 2 days I did not tried this but above link used same for alertlog trace file.

Listener log clarifications

hi,
we have bmc monitoring system where we got database down but actually we can see the database is up and listener is working fine...
my question is i got some thing like this below in the listener log..it means there is any problem in network while pinging?
162.53.93.115)(PORT=35133)) * establish * p070 * 0
30-JUL-2009 12:51:21 * ping * 0
30-JUL-2009 12:51:21 * ping * 0
30-JUL-2009 12:51:21 * ping * 0
30-JUL-2009 12:51:21 * ping * 0
30-JUL-2009 12:51:21 * ping * 0
30-JUL-2009 12:51:23 * ping * 0
30-JUL-2009 12:51:23 * ping * 0
30-JUL-2009 12:51:23 * ping * 0
30-JUL-2009 12:51:23 * ping * 0
30-JUL-2009 12:51:23 * ping * 0
30-JUL-2009 12:51:24 * ping * 0
30-JUL-2009 12:51:24 * ping * 0
30-JUL-2009 12:51:24 * ping * 0
30-JUL-2009 12:51:24 * ping * 0
30-JUL-2009 12:51:24 * ping * 0
30-JUL-2009 12:51:25 * ping * 0
30-JUL-2009 12:51:25 * ping * 0
30-JUL-2009 12:51:25 * ping * 0
30-JUL-2009 12:51:30 * ping * 0
30-JUL-2009 12:51:30 * ping * 0
Thanks,
Raman.

It seems someone (something) is tnsping'ing that listener, like this:
C:\>tnsping localhost 5
TNS Ping Utility for 32-bit Windows: Version 10.2.0.1.0 - Production on 30-JUL-2009 20:04:17
Copyright (c) 1997, 2005, Oracle. All rights reserved.
Used parameter files:
C:\Oracle\XE\app\oracle\product\10.2.0\server\network\admin\sqlnet.ora
Used HOSTNAME adapter to resolve the alias
Attempting to contact (DESCRIPTION=(CONNECT_DATA=(SERVICE_NAME=))(ADDRESS=(PROTOCOL=TCP)(HOST=127.0.0.1)(PORT=1521)))
OK (10 msec)
OK (20 msec)
OK (0 msec)
OK (0 msec)
OK (0 msec)
C:\>tnsping localhost 3
TNS Ping Utility for 32-bit Windows: Version 10.2.0.1.0 - Production on 30-JUL-2009 20:04:21
Copyright (c) 1997, 2005, Oracle. All rights reserved.
Used parameter files:
C:\Oracle\XE\app\oracle\product\10.2.0\server\network\admin\sqlnet.ora
Used HOSTNAME adapter to resolve the alias
Attempting to contact (DESCRIPTION=(CONNECT_DATA=(SERVICE_NAME=))(ADDRESS=(PROTOCOL=TCP)(HOST=127.0.0.1)(PORT=1521)))
OK (20 msec)
OK (10 msec)
OK (0 msec)
C:\> (print the tail of the ...\listener.log)
TNSLSNR for 32-bit Windows: Version 10.2.0.1.0 - Production on 30-JUL-2009 20:04:11
TIMESTAMP * CONNECT DATA [* PROTOCOL INFO] * EVENT [* SID] * RETURN CODE
30-JUL-2009 20:04:17 * ping * 0
30-JUL-2009 20:04:17 * ping * 0
30-JUL-2009 20:04:17 * ping * 0
30-JUL-2009 20:04:17 * ping * 0
30-JUL-2009 20:04:17 * ping * 0
30-JUL-2009 20:04:21 * ping * 0
30-JUL-2009 20:04:21 * ping * 0
30-JUL-2009 20:04:21 * ping * 0Better find out how bmc (Patrol?) decides up/down status of the instance.

CryptAcquireContext failing with ERROR_FILE_NOT_FOUND (2L) when user not logged on Windows 8.1

I am having a hard time migrating a C++ CryptoAPI-based application that currently runs on Windows Server 2008 to Windows 8.1. The scenario is:
This application is eventually triggered by WatchDog.exe, which in its turn is triggered when the computer is started by Windows' Task Scheduler.
Task Scheduler uses the following rules to start the WatchDog.exe:
A Administrator User Account;
Run Whether user is logged on or not;
UNCHECKED: Do not store password. The task will only have access to local resources;
Run with Highest Privileges;
Configure for Win 8.1;
Triggered at system startup.
The server sits there, nobody logged, until in a given scenario WatchDog.exe starts the application. Application log confirms that the owner of the process (GetUserName)
is the very same user Task Scheduler used to trigger WatchDog.exe.
It turns out that this application works fine in Windows Server 2008, but in windows 8.1 a call to CryptAcquireContext fails
with return code ERROR_FILE_NOT_FOUND (2L). The odd thing is that the application will NOT fail if, when started, the user is physically logged
on the machine, although it was not the user who started the application manually.
I took a look at the documentation and
found:
"The profile of the user is not loaded and cannot be found. This happens when the application impersonates a user, for example, the IUSR_ComputerName account."
I had never heard of impersonification, so I made a research and found the APIs LogonUser,ImpersonateLoggedOnUser and RevertToSelf.
I then updated the application in this way:
HANDLE hToken;
if (! LogonUser(L"admin", L".", L"XXXXXXXX", LOGON32_LOGON_BATCH, LOGON32_PROVIDER_DEFAULT, &hToken))
logger->log (_T("Error logging on."));
else
logger->log (PMLOG_LEVEL_TRACE, _T("Logged on."));
if (! ImpersonateLoggedOnUser(hToken))
logger->log (_T("Error impersonating."));
else
logger->log (_T("Impersonated."));
err = XXXXXXXXX(); // calls function which will execute CryptAcquireContext
if (! RevertToSelf())
logger->log (_T("Error reverting."));
else
logger->log (_T("Reverted."));
Excerpt with the call to CryptAcquireContext:
// Get the handle to the default provider.
if(! CryptAcquireContext(&hCryptProv, cryptContainerName, MS_ENHANCED_PROV, PROV_RSA_FULL, 0))
DWORD e = GetLastError();
_stprintf_s (logMsg, 1000, _T("Error %ld acquiring cryptographic provider."), e);
cRSALogger->log (logMsg);
return ERR_CCRYPT_NO_KEY_CONTAINER;
cRSALogger->log (_T("Cryptographic provider acquired."));
As the result, I got the log:
[2015/01/08 20:53:25-TRACE] Logged on.
[2015/01/08 20:53:25-TRACE] Impersonated.
[2015/01/08 20:53:26-ERROR] Error 2 acquiring cryptographic provider.
[2015/01/08 20:53:26-TRACE] Reverted.
That seems to show that impersonation is working properly, but still I get Error 2 (ERROR_FILE_NOT_FOUND) on CryptAcquireContext.
Summary:
On Windows Server 2008, the very same application runs properly even without the calls to LogonUser/Impersonate/Revert.
On Windows 8.1, the application, with or without the calls to LogonUser/Impersonate/Revert, will only work properly if the user is logged on (which
is not acceptable).
Any thoughts where I can run to in order to get this working on windows 8.1?
Thank in advance,
Dan

There are a couple of issues.
Based on the parameters being used in CryptAcquireContext(). A profile needs to be loaded and your app has to be running as the same user who created the keyset. (which is why it works when a user is logged on Windows 8.1) Also, impersonation
does not load your user profile, you need to call LoadUserProfile(). It seems like you should be using a machine keyset for your scenario if you want to do this when nobody is logged on.
Take a look at the following KB article for more information.
https://support.microsoft.com/kb/238187?wa=wsignin1.0
thanks
Frank K [MSFT]

Rotate Listener log on Windows

Similar Messages

Maybe you are looking for