Now that the IIS Advanced Logging is dead I am stuck with IIS 10. Is there a way to get the Enhanced Logging to rollover its log every minute? We've used this to determine performance on our servers in near real-time. Now the best rollover option is once an hour. Is there anything I can do outside of writing a custom logging module?
Basically, for performance reasons, logs are buffered and written at these predefined intervals. If these default rollover options are not what you want, you can try to force all entries in the log buffer to be written from the buffer to the log file, here is the command to flush the IIS buffer immediately:
Run cmd as administrator:
cd C:\Windows\System32
netsh http flush logbuffer
You can loop it using a batch file, the following example .bat file refreshes the log file every minute: (the method is mentioned in this thread.)
#echo off
:loop
netsh http flush logbuffer
timeout /t 60 > NUL
goto loop
Related
I am using ftp command lines to send big video files (about 500 Mo) to 6 remote servers.
On 4 servers, everything goes well.
On 2 servers, my script hangs and the upload never completes.
It doesn't hang on the same files on the two servers, so I don't think it's a problem with the files.
Also, on one of the two servers, one video file is first sent successfully, but the script hangs on the second upload.
When I try to abort the script (Ctrl + C or Ctrl + D), nothing happens. I have to kill the terminal to end the script.
This is the code I'm using:
ftp -v -n $server << EOF
prompt
quote USER $user
quote PASS $password
mput *.mov
quit
EOF
What could be my problem?
I'm not entirely sure what is going on here but perfoce will fail 100% of the time when trying to submit a very specific file in our depot. Things I have tried to solve this problem:
It Submit the file on it's own or in a changelist with other files
Submit using parallel sync or without it
Adjust the batch and thread numbers
Submitting via the command line and the visual client
Increasing or turning off net.maxwait
Increasing or turning off netkeepalive.idle as well as interval.
Checking disk space on server( 150GB free)
Deleting the current instance of the file and copying a backup overtop of the current one
The error message I receive are:
From Client side CLI:
write: socket: WSAECONNABORTED
From Server logs:
TCP send failed. write: socket: WSAECONNABORTED Perforce client error:
TCP send failed.
write: socket: WSAECONNABORTED
Perforce is just flat out refusing to let me submit this one particular file, it's a 63MB file, larger than some but smaller than others which have gone through successfully. We are on a 30Mbps connection, so I usually get the failure message after 10-20 seconds. I'm at a loss here so any help or ideas would be greatly appreciated. Thanks!
To connect from windows desktop to unix host, in order of administration on command line level, I like to use MobaXterm in it's free version
The connection is encrypted ssh, identification to Unix host is using user id, passwd.
There is an issue in I can't resolve:
In the "Session Settings" is a field "Execute Command".
This command there is supposed to be executed in the target host when login has been successful finished. The hosts I have to use are may be not the fastest - Or MobaXterm is too fast.
However, the command is issued in a to early stage, the host feels mortally offended, closes connection and throws me out.
It does not matter what the command is - even a sleep is not accepted.
How to tell MobaXterm to act a bit slower?
MobaXterm default behavior for executing a command is to close the session after execution.
To prevent that there is a setting near the "Execute command" field that states: "Do not exit after command ends". Just check that and it will fix the problem.
On occasion Post request sent to our WAS leave the browser (IE) waiting for a response forever (until manually cancelled - e.g. by closing the browser).
Looking at the Trace and SystemOut logs on WAS everything looks fine. in the http_error log we can see the Get and Post requests but not the responses (even for successful requests). Is there a way to see the responses, or at least when they are sent? We need this to ensure that a response is being sent to the browser. Thanks.
I would issue a SIGQUIT signal to the Websphere process in order to produce a javacore, i.e. a threadump, in order to see what my WebContainer threads are actually doing.
Try running a kill -3 <WAS process ID> as root in a unix platform or invoke DumpThreads through wsadmin if running on a Windows platform.
You should enable trace for servlet container. This is done by setting trace string to *=info:com.ibm.ws.webcontainer*=all:com.ibm.wsspi.webcontainer*=all:HTTPChannel=all:GenericBNF=all
Setting up a trace
Does anyone know if nginx supports soft quits? Meaning does it stay running until all connections are either gone or timed out (past a specific time interval) and also not allow new connections during this time period?
For example:
nginx stop
nginx running (2 connections active and blocking any new connections)
nginx running (1 connection active)
nginx stopped (0 connections active)
man nginx
-s signal Send signal to the master process. The argument signal can be
one of: stop, quit, reopen, reload.
The following table shows the corresponding system signals.
stop SIGTERM
quit SIGQUIT
reopen SIGUSR1
reload SIGHUP
Specifically, you want SIGQUIT. In layperson's terms:
stop — fast shutdown
quit — graceful shutdown
reload — reloading the configuration file
reopen — reopening the log files
See also: http://nginx.org/en/docs/control.html for details, and http://nginx.org/en/docs/beginners_guide.html#control for a quick reference.