Ping times
Ping times
Hi all,
I've noticed this before, but never could figure it out:
When the computer which runs HostMonitor is just rebooted the average PING time to other computers is about 5ms.
In time this will increase up to about 50ms in about 4 weeks time.
I can see it happen in every history chart from any PING test.
Not shure if it will increase any further because the machine has never been un-rebooted (if that's even a word LOL) for more than 4 weeks.
Is this known behaviour? Could there be something filling up that needs to be cleared every once in a while?
Config server:
2x Dual Core Xeon @ 3.2Ghz
10GB mem
2x 1GB lan (teamed)
W2K8 R2
HostMonitor 8.86 / 460 tests
Other software running on it:
BackupExec 2010 R3
Symantec SEP Manager
I've noticed this before, but never could figure it out:
When the computer which runs HostMonitor is just rebooted the average PING time to other computers is about 5ms.
In time this will increase up to about 50ms in about 4 weeks time.
I can see it happen in every history chart from any PING test.
Not shure if it will increase any further because the machine has never been un-rebooted (if that's even a word LOL) for more than 4 weeks.
Is this known behaviour? Could there be something filling up that needs to be cleared every once in a while?
Config server:
2x Dual Core Xeon @ 3.2Ghz
10GB mem
2x 1GB lan (teamed)
W2K8 R2
HostMonitor 8.86 / 460 tests
Other software running on it:
BackupExec 2010 R3
Symantec SEP Manager
-
- Posts: 2832
- Joined: Tue May 16, 2006 4:41 am
- Contact:
Ping times
We had similar problem. It was connected with Windows Server 2008R2 TCP/IP Window Auto-Tuning feature.
Could you try to disable it ?
You may use Command Prompt started with Administartor privileges and command like the following:
netsh interface tcp set global autotuninglevel=disabled
Could you try to disable it ?
You may use Command Prompt started with Administartor privileges and command like the following:
netsh interface tcp set global autotuninglevel=disabled
Reporting....
Problem solved indeed
Thanks!
Kris
Thanks!
Kris
Sorry, we did not find other possible reasons for such problem.
May be some Symantec sideeffect?
If you cannot disable/uninstall Symantec, probably you can use RMA installed on different (clean) system to perform Ping tests? This is not very good idea for production monitoring because HostMonitor<->RMA traffic is more resource consuming than simple ping/ICMP packets but just for testing...
Regards
Alex
May be some Symantec sideeffect?
If you cannot disable/uninstall Symantec, probably you can use RMA installed on different (clean) system to perform Ping tests? This is not very good idea for production monitoring because HostMonitor<->RMA traffic is more resource consuming than simple ping/ICMP packets but just for testing...
Regards
Alex
-BUMP-
New job, new HM installation
And again I run into the same problem.
Ping response time for all ping tests increases as time goes by.
Already disabled TCP/IP autotuning feature.
I can't believe I'm the only one having this issue.
Config:
CPU: Intel i5 @ 3.2GHz
Mem: 8GB
OS: Win 7 Enterprise UK 64-bit
AV/Firewall: None whatsoever
HM version: 9.59
New job, new HM installation
And again I run into the same problem.
Ping response time for all ping tests increases as time goes by.
Already disabled TCP/IP autotuning feature.
I can't believe I'm the only one having this issue.
Config:
CPU: Intel i5 @ 3.2GHz
Mem: 8GB
OS: Win 7 Enterprise UK 64-bit
AV/Firewall: None whatsoever
HM version: 9.59
Could you try the following:
Step 1: Open regedit
Step 2: Go to the HKEY_LOCAL_MACHINE\\SYSTEM\\CurrentControlSet\\Services\\Tcpip\\Parameters\\Interfaces
Step 3: There will be many sub-keys with names like {068CA6B5-ACDD-4FC1-BCBB-66380B3EAF17}, go through each one of them and search for the one that has a key value called DhcpIPAddress that contains your current IP.
Step 4: Right click on the key and select New->DWORD Value, name it
TcpAckFrequency, then double click it and set it to 1
Step 5: Right click on the key and select New->DWORD Value, name it
TCPNoDelay, then double click it and set it to 1
Step 6: Restart your computer
Regards
Alex
Step 1: Open regedit
Step 2: Go to the HKEY_LOCAL_MACHINE\\SYSTEM\\CurrentControlSet\\Services\\Tcpip\\Parameters\\Interfaces
Step 3: There will be many sub-keys with names like {068CA6B5-ACDD-4FC1-BCBB-66380B3EAF17}, go through each one of them and search for the one that has a key value called DhcpIPAddress that contains your current IP.
Step 4: Right click on the key and select New->DWORD Value, name it
TcpAckFrequency, then double click it and set it to 1
Step 5: Right click on the key and select New->DWORD Value, name it
TCPNoDelay, then double click it and set it to 1
Step 6: Restart your computer
Regards
Alex
Hi
We are using HM v9.58 on Windows 2008 R2.
After reading this topic, I checked at my ping times and realized that the same situation is valid for my ping tests. Then, I applied both solutions (auto tuning and registry keys) and the situation has been corrected.
A history chart sample showing my situation is below:
https://app.box.com/s/q0sc6lf3kchm136clq6r
In this graph, the points which the lines drop off show the times when host monitor restarted. I applied the solutions on 26th October and after this day, the ping times (and graph line) have become steady.
On the other hand, in my case, this situation happens in Host Monitor (Windows 2008 R2) and RMAs running on Windows 2008 R2, but not on RMAs on Windows 2003 SP2 and RMA for UNIX/Linux. Therefore, I thought that the situation may happen because Windows 2008 behaves differently.
Regards,
Oguzhan
We are using HM v9.58 on Windows 2008 R2.
After reading this topic, I checked at my ping times and realized that the same situation is valid for my ping tests. Then, I applied both solutions (auto tuning and registry keys) and the situation has been corrected.
A history chart sample showing my situation is below:
https://app.box.com/s/q0sc6lf3kchm136clq6r
In this graph, the points which the lines drop off show the times when host monitor restarted. I applied the solutions on 26th October and after this day, the ping times (and graph line) have become steady.
On the other hand, in my case, this situation happens in Host Monitor (Windows 2008 R2) and RMAs running on Windows 2008 R2, but not on RMAs on Windows 2003 SP2 and RMA for UNIX/Linux. Therefore, I thought that the situation may happen because Windows 2008 behaves differently.
Regards,
Oguzhan
Thank you for information.
Regards
Alex
Or almost steady. Looks like reply time grows but just a little bit...In this graph, the points which the lines drop off show the times when host monitor restarted. I applied the solutions on 26th October and after this day, the ping times (and graph line) have become steady.
Regards
Alex