Contango's questions

On our Debian servers, I want to ensure that any SSH connections time out and disconnect after 8 hours. This was recommended by our security consultant.

I executed these steps:

# Log in as root then:
sudo vi /etc/ssh/sshd_config
# Use / to search for and change:
ClientAliveInterval=3600
ClientAliveCountMax=8
# Second parameter is number of hours before connection times out and drops.

# Then test config:
sudo sshd -t
# Restart
sudo systemctl restart sshd

However, the tests failed. When I temporarily set ClientAliveInterval=120 and ClientAliveCountMax=2 and restarted the sshd daemon, I would expect the connection to drop after 4 minutes of inactivity. However, it did not.

Any ideas?

I was trying to create a mirrored drive in Disk Management, but I kept on getting this error:

"All disks holding extents for a given volume must have the same sector size, and the sector size must be valid."

Under Windows 8.1, if I format a drive to ReFS, then add a mirror using Disc Management, will it self heal if there is a bad sector on one of the drives?

Or do I have to use Storage Spaces to take advantage of the self healing capabilities of ReFS?

I am using Squid v2.7.

I am getting lots of TCP_MISS messages in the log, and very few TCP_HIT.

What does this mean?

p.s. Here is a sample of the log:

1369840631.853     59 127.0.0.1 TCP_MISS/200 49714 GET http://pagead2.googlesyndication.com/pagead/imgad? - DIRECT/173.194.41.77 application/x-shockwave-flash
1369840631.883    141 127.0.0.1 TCP_MISS/302 1254 GET http://altfarm.mediaplex.com/ad/js/18329-129461-2056-10? - DIRECT/89.207.18.182 -
1369840631.964    163 127.0.0.1 TCP_MISS/200 7939 CONNECT urs.microsoft.com:443 - DIRECT/157.55.231.253 -
1369840631.971    160 127.0.0.1 TCP_MISS/200 7971 CONNECT urs.microsoft.com:443 - DIRECT/157.55.231.253 -
1369840632.093    205 127.0.0.1 TCP_MISS/200 8803 GET http://img.mediaplex.com/content/0/18329/1510_FREE_CrystalClear_Fruitmachine_DCF_300x250_1cc304.js? - DIRECT/89.207.18.181 text/javascript
1369840632.317    193 127.0.0.1 TCP_MISS/200 56608 GET http://img-cdn.mediaplex.com/0/18329/1510_FREE_CrystalClear_Fruitmachine_DCF_300x250_1cc304.swf - DIRECT/46.33.70.42 application/x-shockwave-flash
1369840813.425 287250 127.0.0.1 TCP_MISS/200 12260 CONNECT plus.google.com:443 - DIRECT/173.194.34.70 -
1369840824.390 293221 127.0.0.1 TCP_MISS/200 5204 CONNECT csi.gstatic.com:443 - DIRECT/74.125.226.111 -
1369837426.298      0 127.0.0.1 TCP_MEM_HIT/200 2565 GET http://0.gravatar.com/avatar/0d5183de3952f6c8645c5e6140938ef7? - NONE/- image/png
1369837426.299      1 127.0.0.1 TCP_MEM_HIT/200 1917 GET http://0.gravatar.com/avatar/cbbed29064ada5c3686c4d980e545e67? - NONE/- image/jpeg
1369837426.299      0 127.0.0.1 TCP_MEM_HIT/200 2190 GET http://1.gravatar.com/avatar/9986282128def4361d1971b8bbe3d0d2? - NONE/- image/png

If I type "netstat", I can see a list of IP addresses that my PC is connected to.

If I start with an IP address, how do I find the process that has opened the connection to said IP address?

We have a server with 16GB of RAM. The memory manager is very aggressive in trimming back the Working Set for a particular process, which is resulting in a lot of soft page faults which is slowing the process down and introducing latency.

Are there any settings we can change in the registry to force the Dynamic Memory Manager to be less aggressive with trimming the Working Set?

We wouldn't mind if the server devoted 80% of its RAM to the Working Set of the processes (its only using 2GB at the moment).

I have an Microsoft .NET application that needs to be sped up, so it can have more deterministic round trip response times to incoming TCP/IP internet packets.

Are there any realtime builds of Windows, or realtime OS's that will run Microsoft .NET code?

I guess the subject says it all: would switching to Windows HPC Server 2008 R2 have any effect on average network latency ping times? Even if it shaved a few milliseconds off the time, I'd be happy - but is that realistic?

Every second I get the following fault in the system log in Windows 7. How do I find the source of this error and eliminate it?

- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
- <System>
  <Provider Name="Service Control Manager" Guid="{555908d1-a6d7-4695-8e1e-26931d2012f4}" EventSourceName="Service Control Manager" /> 
  <EventID Qualifiers="49152">7000</EventID> 
  <Version>0</Version> 
  <Level>2</Level> 
  <Task>0</Task> 
  <Opcode>0</Opcode> 
  <Keywords>0x8080000000000000</Keywords> 
  <TimeCreated SystemTime="2011-04-27T09:38:50.676687100Z" /> 
  <EventRecordID>9732594</EventRecordID> 
  <Correlation /> 
  <Execution ProcessID="924" ThreadID="6236" /> 
  <Channel>System</Channel> 
  <Computer>dr-satellite</Computer> 
  <Security /> 
  </System>
- <EventData>
  <Data Name="param1">msrvc</Data> 
  <Data Name="param2">%%2</Data> 
  </EventData>
  </Event>

Currently, there is a 40 second delay when accessing an unavailable UNC network share on Windows 7.

Is there a method to reduce this delay to something less painful, like 20 seconds?

Or, is there some method to eliminate this delay altogether?

Very rarely, our server runs low on RAM (and crashes) due to a certain rogue app that gobbles up all of the RAM. To guard against this happening again, is there any way to configure the built in Performance Monitor to send us a warning email if RAM crosses the 80% use threshold?

Is it possible to run Windows Server 2008 R2 on Amazon EC2? Is it possible to upload your own images of operating systems to Amazon EC2?

I'm using Subversion and TortoiseSVN under Windows, connecting to a Linux webserver running svnserve. I have basic shell access to the Linux webserver.

I have a non-encrypted link working, i.e. I can connect via TortoiseSVN to:

svn://www.mysite.com/svn/myproject/

However, I need to convert this to an encrypted link, e.g.:

svn+ssh://www.mysite.com/svn/myproject/

I can't get it working: any ideas?

p.s. My client will accept nothing less than an encrypted link. Is it necessary? No. Do I need it? Unfortunately, yes.

Is there any service which runs in the background, and does the job of prime95 (i.e. it tests server CPU/RAM for integrity now and again)?

Regards, Shane.

p.s. The reason I ask: we recently had some bad RAM on our server, which ended up slowly corrupting all of our business files. A file would be copied into memory, it would be corrupted in RAM, and copied back onto the RAID 1 hard drive in a corrupted state.

I'm copying 600GB of data from external hard drive A to external hard drive B.

Windows Server 2008 R2 has notified me that 100 files have filenames that are too long (i.e. >255 chars).

Is there a utility that can allow me to search for these filenames, and manually shorten them?

Shane.

We have a contractor working offsite. We would like to set everything up so that his screen is recorded to our Windows Server 2008 R2 machine in real time.

We've been using TeamViewer, but we've run into limitations.

What would you recommend?

p.s. Constraints:

1. The contractor needs to be able to initialize the connection himself, from his end.
2. The recording needs to be in realtime.
3. The video must be recorded at the server end.

p.p.s. I've been a programmer for 15 years. I understand how programmers work, its more to see his coding style and to keep the investors happy, rather than anything else. I trust my contractor.

I'd like to set up some port forwarding in our router, using UPnP.

More specifically, I want to run some program on startup that always forwards port 8443 my IP address (192.168.1.100). Port 8443 is for our SVN server.

How do I do this with Windows Server 2008 R2?

(I don't want to go into the router to set up port forwarding, as everytime I do this, the router resets itself and we lose the internet connection for 5 minutes which plays havoc with everything else on the network)

I'm running Windows Server 2008 x32 under VMware Workstation v6.5.2.

VMwareUser.exe process crashes regularly, and uses 95% of CPU until its shut down manually.

When I shut down VMwareUser, I suddenly lose the ability to copy/paste between the host and guest OS's (I can't live without this so I reboot).

Any idea on how to prevent VMwareUser.exe crashing?

I'm working with a contractor in a different country.

We need some method to add his machine to the local network, using a point-to-point VPN.

Unfortunately, Hamachi, while very easy to use, creates stability problems as he is using Windows 7.

Whats the best method for us to use?

I have Windows Server 2008 running under VMware.

Recently, its started to crash roughly every day, with continuous 100% CPU utilization, and no response in the GUI.

Is there a step-by-step technique to track down the source of this problem?

What logs would I look at?

p.s. The problem appeared around the time I tried to uninstall Acronis, and it blue screened. However, I'm not sure if the current faults are related to Acronis at all.