Keep alive TCP setting for port 3998 to avoid fw timeout?

Hello,
in our new DS6.1 env, replication between the master DS is using port 3998 through several firewalls. If there are no changes to the DB (we are still testing), the connection seems to time out on the firewall and the status of the remote masters in the SMC drops to "stopped" even though they are still running. If one clicks on the DS entry in the SMC, it seems to set up a fresh TCP connection and the status is ok again.
Is there a way to define a keep-alive setting for these connections ?
TIA,
heinz

DS 6.3 has bug 5087249 against it for the missing SO_KEEPALIVE socket option.
This bug was fixed a long time ago in the DS 5.x series which got "unfortunately lost" on the way
to the 6.x release :-(
If you have a valid Sun Support contract you can request a hotfix for this bug 5087249 in
DS 6.3 (or the soon to be released 6.3.1), by raising a support service request.
A similar missing SO_KEEPALIVE socket option bug exists in the DPS 6.3 server,
bug 6782659 is open against this and if you have a valid Sun Support contract you can
request a hotfix for this bug 6782659 in DPS 6.3 (or the soon to be released 6.3.1).
Without a valid support contract, I'm afraid you'll have to wait until the next official patch
release of both these servers (DPS & DS) after the 6.3.1 which does not contain these 2 fixes.
This will most probably be at least one year into the future from present.
The problem with CLOSE_WAITs DPS issue is fixed in the same hotfix for bug 6782659 (which
contains many connection handling bug fixes/enhancement in addition).
Hope this helps.

Similar Messages

How do I change the DNS settings to BC but keep the MX setting for emails?

How do I change the DNS settings to Business Catalyst but keep the MX settings for emails?
I'm basically created anew site using Muse and want to host it on BC and upgraded to the Web basics + to allow for hosting of downloadable pdf's etc.
There are up to 50+ emails address set up on the current set up and I want to just change the NDS settings for the domain to ponit to the new site/hosting an not have to worry about them chaning all the meail settings etc.
Records do I need to change in BC?
Thanks
Iain

Hi,
You shouldn't need to do anything as long as you do not state hosting email with BC. Just ensure to check you DNS in a lookup tool that it's still referencing your external email service.
Otherwise you can set up external email set-up if the option is present.
- http://helpx.adobe.com/business-catalyst/partner/set-email-external-email-provider.html
Kind regards,
-Sidney

Keep Mighty Mouse setting for all network users.

We have intel macs at my school, running parallels for Win XP.
The mouse needs to be set up per user to allow the right click function on the windows side, users do not like using the ctrlshiftclick method.
Is there a way I can set up so that all users who have the mighty mouse has the right click function enabled as the secondary action? I dont have time to log in as each user to set it up.
The users also use PPC eMac, those do not have the fancy mouse, just the standard one.
So far i have tried to copy the mouse preferences: com.apple.driver.AppleHIDMouse.plist. But it does not want to copy to all the users using:
cp -R /Volumes/docs/Mouse\ Prefs/com.apple.driver.AppleHIDMouse.plist /Volumes/docs/students///Library/Preferences/
Some reason it does not copy the file.
Any ideas? Is it the right file?
-Jeff

cp -R /Volumes/docs/Mouse\ Prefs/com.apple.driver.AppleHIDMouse.plist /Volumes/docs/students///Library/Preferences/
The shell will parse this command by first expanding the "*" to every directory under .../students. It then executes that command - not what you want.
You probably want something like this...
cd /Volumes/docs/students
for dir in * do ; cp /path/to/plist $dir/path/to/PrefsDir ; done
This first expands the "*" to all items in students dir. It then executes the part after the "do ;" for each item in the list, setting the variable "dir" to the first one, then second, etc. The "$dir" gets replaced by the value of "dir".
Note that this will not handle any directory names in /students which have spaces in them (because the 'for' treats spaces as argument separators). if this applies, then changes will have to be made.
Before running it on everything, test run using only one student using...
for dir in chosenName do ; cp /path/to/plist $dir/path/to/PrefsDir ; done
Ths will set 'dir' to 'chosenName' and execute the following bit for that value only (just to ensure it puts what you want, where you want it .
Note also that your "-R" option is ignored - this only applies to the copy source if it is a directory (recurive copy), not the target.
-david

HTTP persistent connection (keep Alive) problem

Hi,
I'm trying to deploy a RPC web service and I would like to use the HTTP 1.1 keep alive feature (which should be the default).
I generated the server side and client side code using jdev 903. The AS is OC4j 903.
The problem is that the server is sending a âConnection: close:â header in the SOAP response, despite the client sending a âConnection: keep-aliveâ header in the request. This is obviously causing the client to close the connection and open another for the next message.
The documentation implies that the server will always try to use persistent connections, even so, in the file:
<OC4J_HOME>/j2ee/home/config/http-web-site.xml
there is a config option âuse-keep-alivesâ. Setting this to true has no effect.
Iâm guessing there must be another option somewhere to turn on connection persistence.
Can anyone please help ?

Does anyone know whether Oracle Supports persistent connections ?
I tested this on JBOSS and I seems to get better performance.
cheers.

Where to find the timeout setting for connecting to SAP Bi Universe in Power Query for Excel

Hi All,
I am trying to connect a SAP BI Universe. I am able to login to the environment and get the list of folders. But on trying to use the universe inside the folder, I am getting a time out error. The time out happens exactly at 100 seconds always.
DataSource.Error: SapBusinessObjects: The operation has timed out
Can you let me know if any setting can be done to avoid this timeout and get the universe into power query?

I have the same problem with Version: 2.17.3850.242. It times out after 100 seconds.
In addition, I couldn't report it by sending a frown. Get the following error message
Error Message: Could not load file or assembly 'Microsoft.Office.Interop.Outlook, Version=14.0.0.0, Culture=neutral, PublicKeyToken=71e9bce111e9429c'
or one of its dependencies. The system cannot find the file specified.

Which port to use for the peer-keep alive

Hi All,
We have 2 Nexus 6001s in our data center.
The management port of each 6001 is connected to the other and this link is used as the peer keep alive link.
My colleague is suggesting that we use one of the inline data ports as the keep alive link.
Can you please advise on the pros and cons of using management/inline port as keep live link and the best practise to follow in this case?
Thanks,
Pete

Hi Pete,
Here are the best recommendation in order of preference.
1. Use mgmt0 (along with management traffic)
* Pros: whats good on this option is you are totally separating the VPC Peer keepalive link on another VRF (management) and does not mingle with the data or global vrf..
* Cons: VPC PKL is dependent on the OOB management switch.
2. Use dedicated 1G/10GE front panel ports.
* Pros - can just be a direct link between the N6K pair and not dependent on other boxes.
* Cons - you need extra SFPs for VPC PKL while the VPC PKL traffic join the global VRF.
HTH
Jay Ocampo

Keep alive script for HTTPS service with GET method

Hello,
I tried to implement a keep alive to services that support HTPS connections. Ive created a script that opens a tcp connection on port 443 and tries to do a GET of a test web page. This web page only displays OK.
In a resumed form I used the following script commands:
Socket connect host 192.168.136.134 port 443 tcp
Socket send ${SOCKET} GET /teste.template\n\n
Socket waitfor ${SOCKET} 200 2000
Socket disconnect ${SOCKET} graceful
I was unable to receive the expected 200 response and received a message saying string not found, I used the socket inspect command to inspect the sockets internal data buffer and the result was nothing, it seemed empty.
I tried the same operation using a linux machine that has IP connectivity with the Webservers and I was successful.
In the same CSS I have services configured to respond to HTTP sessions. I tried the following script and everything went well
Socket connect host 192.168.136.151 port 80 tcp
Socket send ${SOCKET} HEAD /index.html\n\n
Socket waitfor ${SOCKET} 200 2000
Socket disconnect ${SOCKET} graceful
When I issued the socket inspect command I could in fact see the sockets internal data buffer content.
The CSS is running WebNS 5.0 build 66
Can someone give me a help?
Thanks,
Ricardo Lourenço

Hello,
There is no direct no direct link between TCP and SSL.
So, in spite of having success with the TCP port 443 connection the CSS will not be able to negotiate de security parameters necessary to establish the SSL session.
This was the reason for not having success through the use of the socket commands, as i refered in my original message.
Thanks for your collaboration,
Ricardo Lourenço

"Keep Alive" parameter for listener?

Hi all,
we have following problem:
From time to time the client (oracle forms 6i patch 18) gets the error message ORA-03114. Te solution is only to kill forms application and start it again.
This hangs probably on the network breaks in the communication (over VPN) with the database.
Is there any parameter for listener like "Keep alive" or else, which can rebuild the connection once it lost?
Best regards

Paste this into the SQLNET.ORA on the application server or the server that is making the SQL calls to the database.
Trace_level_client=16
Trace_directory_client=<path_to_the_trace_directory> # use the full path to the trace directory
Trace_unique_client=on
Trace_timestamp_client=on
Then review the trace or you can try the keep alive in windows
Use the standard TCP KeepAlive on Windows
1. Open the registry under:
HKEY_LOCAL_MACHINE: SYSTEM: CurrentControlSet: Services: Tcpip:
Parameters
2. Add KeepAliveTime : REG_DWORD to the registry
3. Set the value to a reasonable timeout.(decimal and NOT Hex) Like 480000 equals 8 minutes or 120000 equals 2 minutes etc etc.
- Save this
- Reboot

TCP KEEP ALIVE

Hello we have a problem with TCP/IP keep settings.
We have an apllication that after a while (2 hours) doesnt keep its connection,And after that we are taking of timeouts and disconnection from this application.We saw some threads that there a registery TCP/KEEP ALIVE,we already change this registery but
it doesnt help still.Could it be any other solution?Or advice?

I want to configure it to a lower value so that the connection is not closed by routers/firewalls assuming the connection to be idle.For all TCP clients and servers?
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\KeepAliveTime 30000 30 seconds is an absurdly low value. Routers and firewalls will typically close connections after an hour, if at all. So setting it to 30 minutes should be adequate. Setting it as low as you have only exposes you to more network problems: if someone has a cable out for 30 seconds you will get disconnected, even if the cable was put back after 31 seconds and even if your application doesn't actually exchange data every 30 seconds. Don't do this.
Is there a standard way in which the keep alive time can be configured on all OS platforms? No.
If not, is this a standard way of changing keep alive time for java applications on windows?The registry keys might differ from version to version. It's a Windows question, not a Java question.
Is it necessary that the machine be restarted on changing keep alive time or is there a specific time by which the new keep alive time would be picked up during run time or on restarting the java application?Another Windows question.
But I would review your requirement. More likely you should be building application pings into the applications that actually need them.

How do you set up Port Forwarding for ARD 2.2 in AEB N?

Help,
I'm a novice at Apple Remote Desktop (ARD) - not an IT guy, so it has to be pretty basic and detailed.
How do you set up Port Forwarding for ARD 2.2 on the Apple Airport Extreme BS router, 802.11 N. I have one at each end of the internet connection. At one end I have an Airport Extreme N router with 2 macs and eventually 1 windows XP machine (if I can) that I would like to be able to connect to over the interenet (the clients) and at the other end, I have a Mac with ARD 2.2 installed also with an Airport Extreme N router. Note: Both routers use Static IP addresses and all computers use static IP's internally not through DHCP. What are the settings or directions to do this.
I have read and printed out the directions for Configuration of ARD 3.0 that are posted many times in the ARD discusion group, but it uses a Linksys router ( http://www.starkpr.com/ard.htm posted by Dave Sawyer). The Mac router is different, particularly with the place to set a Private IP address. I'm not sure about alot of things, but especially about the Private IP address, what number do I set it to, the one that is in my Network connections list? It automatically changes to a different number in AE N setup for Port Forwarding (by one) as if it is not suppose to the same?????
Are there any directions available that are as straight forward for the Airport Extreme N router, as the one's that are listed here for the Linksys Router's? ( http://www.starkpr.com/ard.htm )
Any and All help will be greatly appreciated.
P.S. I know I should have 3.0 but bought 2.2 just weeks before 3.0 came out and they would not give me an upgrade price, so I'm waiting for 4.0 to upgrade.
Thanks,
Jim

Try the following for each AirPort Extreme ...
AEBSn - Port Mapping Setup
To setup port mapping on an 802.11n AirPort Extreme Base Station (AEBSn), either connect to the AEBSn's wireless network or temporarily connect directly, using an Ethernet cable, to one of the LAN port of the AEBSn, and then use the AirPort Utility, in Manual Setup, to make these settings:
1. Reserve a DHCP-provided IP address for the host device.
Internet > DHCP tab
o On the DHCP tab, click the "+" (Add) button to enter DHCP Reservations.
o Description: <enter the desired description of the host device>
o Reserve address by: MAC Address
o Click Continue.
o MAC Address: <enter the MAC (what Apple calls Ethernet ID if you are using wired or AirPort ID if wireless) hardware address of the host computer>
o IPv4 Address: <enter the desired IP address>
o Click Done.
2. Setup Port Mapping on the AEBSn.
Advanced > Port Mapping tab
o Click the "+" (Add) button
o Service: <choose the appropriate service from the Service pop-up menu>
o Public UDP Port(s): 3283
o Public TCP Port(s): 3283
o Private IP Address: <enter the IP address of the host server>
o Private UDP Port(s): 3283
o Private TCP Port(s): 3283
o Click "Continue"
o Click the "+" (Add) button
o Service: <choose the appropriate service from the Service pop-up menu>
o Public UDP Port(s):
o Public TCP Port(s): 5900
o Private IP Address: <enter the IP address of the host server>
o Private UDP Port(s):
o Private TCP Port(s): 5900
o Click "Continue"
o Click the "+" (Add) button
o Service: <choose the appropriate service from the Service pop-up menu>
o Public UDP Port(s):
o Public TCP Port(s): 5988
o Private IP Address: <enter the IP address of the host server>
o Private UDP Port(s):
o Private TCP Port(s): 5988
o Click "Continue"
(ref: "Well Known" TCP and UDP ports used by Apple software products)

How do you configure the set of port to use for FTP passive data connection

I was able to start the ftp server xml db and connect to it locally. When I tried to connect to it from home, I was able to log in but when i do a "get" command the sessions hangs. I also tried passive mode and do a "get" command and it gives me a connection refused error. I am using the default port 2100. I think this might be a firewall issue with the data connection port. I am not sure which range of ports to open for the passive ftp connection on the server side or where to go and set them for for the ftp server to use? Or this issue might not be firewall related??? Any thoughts. Thanks.

This is a typical problem with VPN/Firewall software and FTP on ports other than 21.
When you open the FTP data connection the FTP Server identifies which port the data will be sent on using the PORT command which is sent on the command port (typically 21 with a normal FTP server, 2100 by default with XDB). THe client then opens that PORT to accept the data (file, output of an ls etc).
As I understand it most VPN/Firewall are configured to monitor 21 for 'PORT' requests. They then understand to allow connection on the PORT number passed on the command PORT. Hence with Port 21 firewall / VPN software is not a problem. However since the XDB command PORT is something other than 21 they do not see / honour the PORT commands from the server and hence the client cannot open the data connection on the specified PORT
As to how to solve it, I wish I knew (After 5 years I still can't use XDB FTP from home to a machine in the office once the VPN tunnel is open :( ). I'm sure it's going to be firewall / VPN software specific and may need the VPN servers to be configured.
The best bet would be to run the FTP server on 21...
Add the following line to your listener.ora file
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP) (HOST = hostname) (PORT = 21))
(PROTOCOL_STACK = (PRESENTATION = FTP) (SESSION = RAW)))
and then restart the listener as root
Start the listener as root.
At the operating system prompt, enter tnslsnr with optional command line arguments. The usage is as follows:
tnslsnr [listener_name] [-user user] [-group group]
where:
Table 4-3 tnslsnr Utility Options
Option Description
listener_name
Specify the name of the listener. If omitted, the default name LISTENER will be used.
-user user
Specify the user whose privileges the listener will use when super user (root) privileges are not needed. After performing the privileged operations, the listener will give up root privileges irreversibly.
-group group
Specify the group whose privileges the listener will use when super user (root) group privileges are not needed. After performing the privileged operations, the listener will give up root group privileges irreversibly.
The listener will temporarily switch to the provided user and group immediately after startup. All subsequent operations will be done with the specified user and group privileges, except the system calls necessary to listen on configured endpoints. The listener will revert to super user (root) for a short period of time to listen on reserved addresses, such as TCP ports less than 1024. After the listener starts listening on all of its endpoints configured in listener.ora, it will switch to the specified user and group irreversibly. Therefore, the listener will give up the root privilege that it initially had. In the current release, -user and -group command line arguments only accept user and group identifiers specified in numeric form.
For example, to execute a root listener called mylsnr and have it use privileges of a user identified

WRT54G v 7.0 -- Setting "Keep Alive: Redial Period" not correct (radio button)

Hi,
I own a Linksys-Router WRT54G v7.0 with firmware v7.00.8 build 008, Jul. 9, 2009.
On the web GUI (Setup/Basic Setup) there's a setting "Keep Alive: Redial Period XX Sec.". As I'd like to stay connected with my provider permanently, I've set the radio button to "Keep Alive: Redial Period" and the according text field to "30" seconds. Then, I've clicked "Save Settings" and got a confirmation message that the settings got saved.
Anyway, even after saving the settings the according radio button is still set to "Connect on Demand...". But that is only true for newer browsers, like e.g. IE8, IE7 or Firefox 8. If you use an older browser, like e.g. IE6, then the radio button is correctly set to "Keep Alive: Redial Period" with a value of "30" seconds.
Conclusion: The settings get correctly saved by the router (I can confirm that), but they don't get correctly shown in the web GUI. Looks like a HTML/Javascript bug in the router's admin web GUI. That's really annoying as you never exactly know the effective settings...
Can someone confirm that bug? Is there a newer firmware than version 7.00.8?
Regards,
Patrick

Try to clear the browser's cache as well.
"Don't fix it if it ain't broken."

Need help w/ setting up ports to run a server for America's Army

Need help w/ setting up ports to run a server for America's Army. I read wat u need to change the ports but i dont understand wat to put. here is wat the site says
Q: How do I run my own server?
A: Quick and dirty server info:
1. Edit RunServer.bat to change the map.
2. Run RunServer.bat
Or:
server.exe LAN MAPNAME.aao (Host a LAN game)
server.exe global MAPNAME.aao (Host a Public game)
Also: When you create a server setup and want to allow other users to join your server, you need make sure the following ports are open for outgoing and incoming traffic in your firewall: 1716 (UDP), 1717 (UDP), 20025-20045 (TCP), and 20047 (TCP). Failure to open these ports will prevent the server from accepting connections from other players or prevent other players from being able to see your server online.
There are several settings that also need to be defined in your server configuration INI file (in the Windows version, these files are located in “My Documents\America’s Army Server Settings\{settings file name}.ini”).
[Engine.GameEngine]
ServerActors=Andromeda.AndromedaMBS
[Andromeda.Andromeda]
GameServerIp=
Make sure that you set the actual IP address of the America’s Army Server under GameServerIp= (for example, “GameServerIp=000.000.000.000”). The supplied address must be your actual internet IP address, if this is left blank or you supply the IP address for your internal network (such as 192.168.0.x), your server will not be able to accept connections from the internet.
If your server.ini file contains the setting shown below, please change the QueryPort setting to 20025. This setting can also be removed, as the default setting is port 20025.
[Andromeda.AndromedaMBS]
QueryPort=20025
Punkbuster user fix correction.
If [Engine.GameEngine] block has been changed to read as below:
[Engine.GameEngine]
ServerActors=IPDrv.AndromedaMBS
Please add the following block to your INI file:
[IpDrv.AndromedaMBS]
QueryPort=20025
(Last Updated: 2006-04-20)

Your images are not stored in the catalog. They are stored in folders on your computer. If you imported images that were already on your computer using the "Add" Option they are still in that same folder. If you imported images from your camera then they are in the folders that you specified when you imported. The catalog points to those images wherever they are located, and records all of the adjustments that you make to the image. When you send an image to Photoshop for further editing and save that image in Photoshop, it is normally saved back in the same folder as the original image.
Images are not "saved" in Lightroom. The basic default workflow in Lightroom is to store all of the adjustments in the catalog, leaving the original image completely unmodified. The catalog becomes the central controlling mechanism. It is a database that contains pointers to where the images are located and a record of all adjustments made to those images using Lightroom. Properly managed, you only have those original master files and secondary files for the ones that you have sent to Photoshop for further adjustment. When you want to provide a copy for someone else, you use the export dialogue for that purpose. I often export JPEG images to share with others or to post on the web. After I have usedthe JPEG for its intended purpose I delete it.

Is it best practice to use a dedicated link between NX7K pair for keep-alive?

I have been using dedicated physical 10G link between pair of NX7k for keep-alive. Is it a best practice? It seems a waste of 10G ports because keep-alives does not need that much bandwidth.
I'm thinking just configure a dedicated VLAN interface in the default VRF, and have it routed through other devices (fir example 6509 core switches) for keep-alive. Has anyone done that before? The goal is not to use dedicated 10G ports for keep-alive.
Thanks a lot.

gwhuang5398 wrote:I have been using dedicated physical 10G link between pair of NX7k for keep-alive. Is it a best practice? It seems a waste of 10G ports because keep-alives does not need that much bandwidth.I'm thinking just configure a dedicated VLAN interface in the default VRF, and have it routed through other devices (fir example 6509 core switches) for keep-alive. Has anyone done that before? The goal is not to use dedicated 10G ports for keep-alive.Thanks a lot.
I agree that using a Tengig port for this purpose is not efficient. The whole idea of peer-keepalive is to have a way to avoid split brain scenarios in case of peer-link failure. For this reason the peer-keepalives should never cross the peer-link as it defeats the original purpose I just cited. Following are some viable options:
You can use the N7K management ports for peer-keepalive functionality. If you have redundant supervisors then make sure you use an external switch (OOB switch for example) to patch the management ports.
You can use dedicated interfaces just like you are doing right now but that would make more sense if you had Gig ports in my opinion.
You can run peer-keepalives inband but you need to make sure this traffic is never routed over the peer-link. It is considered a good practice to use a dedicated VRF for this purpose.
Atif

PPPoe Keep-Alive Program for Solaris 8?

Is there a PPPoE Keep-alive program for Solaris 8? I need this because my DSL provider keeps knocking me off and therefore I lose my connection. I would prefer the program be precompiled if there is one available.
-Thanks in advance

It sounds like you are accidentally running NIS. Somewhere, you told your system that "mydomain.net" was your NIS domain. Look to see if you have an /etc/defaultdomain file. Remove it. You don't have NIS, and you probably have the wrong /etc/nsswitch.conf setup then, too.
Try copying /etc/nsswitch.dns /etc/nsswitch.conf (make a backup first).
Verify that you have 'nameserver' entries in /etc/resolv.conf and a 'domain mydomain.net' entry at the very beginning. This is where you let the system know your Internet domain (well, also in the fully-qualified domain name of your machine in your /etc/hosts file).
In your /etc/hosts file, the first name of your machine should be the FQDN, then a short alias, if you are going to use one, and probably "loghost". You might put "loghost" as an alias for "localhost".
Make sure your 'hostname' reflects the FQDN, or mail might not like it. Try setting the hostname with 'uname -S <hostname>' if it isn't correct.
Good luck.

Keep alive TCP setting for port 3998 to avoid fw timeout?

Similar Messages

Maybe you are looking for