Chris Dale's questions

I have a server which we are experiencing as really slow. Upon login into the server as domain admin we notice a huge system status bar, filled with invisible icons. If we mouse over the icons, they rapidly disappear. These icons are only present on one of the login sessions (3), as it is not visible on other accounts logging in.

The picture above shows what it looks like when it has occured. It occurs every 10 minutes or so.

Does anyone have an idea on how to discover what is producing this? I have done the following:

• Reviewed event logs
• Payed attention in Process Explorer while experiencing the error

Any other ideas that may not be so error prone? I would've thought the event logs would give me the answer, but any errors there are not related to this it seems.

I have a XP box running some MSMQ public queues. Windows 7 clients cant connect to these, however Windows XP clients can. When connecting from my Windows 7 client with a MSMQ Queue Explorer program, I get the following error message:

When setting up a new XP box, registering it in the domain, installing queue explorer, everything works fantastic. I should also note that my Windows 7 box has no problems connecting to OTHER XP boxes running MSMQ, just some of them (all running latest service packs).

After a LOT of debugging I've seem to isolated the problem down to the RPC calls being made. I see the following problem both when using Queue Explorer and rpcping. The example provided under is by using rpcping with the following command:

rpcping -s xp-host-1 -a integrity -u 9 && rpcping -s xp-host-2 -a integrity -u 9

xp-host 1 has the problem, xp-host-2 is a fresh install without the problem. Inspecting the data from wireshark provides this view. It is clear that the last RPC calls being made succeeds as you will see two additional calls being made to the second host.

Inspecting the packets reveal no distinct differences, however when using Queue Explorer against the two different servers (both XP) we see a difference. It seems to have something to do with authentication / challenge/response or something, but I am struggeling to find out what.

Does anyone have any idea how I can try solve this problem? I have tried to disable the different types of authentication types on my client to no avail. I've tried only lanman, ntlm v1, ntlm v2 and so forth.

Some other facts:

• There is no information in the event logs regarding this
• The MSMQ error codes are not giving a lot of information. Basically they mean "Something went bad, we dont know what".
• If I use the IP address instead of Hostname, MSMQ does not give an error but does not allow me to view public queues.

EDIT 1 more picture:

It looks like it is the rpc__mgmt_inq_princ_name request that creates the problem. On the left side of the picture I try to connect to the host not working, on the right side I connect to same OS which is working.

I have two domain controllers, one where the USN in question is always the same values (current and original), while the other one the original USN is roughly 6 numbers lower than the current value.

What does this mean? How could this potentially affect my applications?

Here is a screenshot of the original and current USN's differentiating:

A client have a couple of racks in a shared server room facility. The racks have wheels on them and they are worried anyone could simply access the server room, unplug all cables and roll out the racks within few minutes.

Does anyone know of any standard security practices one could take in securing a rack in place? Preferably without removing the wheels.

Currently some solutions are to chain them together or chaining them to some infrastructure piece in the room, however I would think that some racks come with ways to secure them from being easily moved?

I am currently experiencing troubles with performance over the last few weeks and I need some aid in how to troubleshoot the Microsoft Messaging Que server.

The IIS and MSMQ resides on the same serevr. There is several application servers which process and return messages to the MQ. The architecture is similar to this:

For some reason sometimes the messages take ~10+ seconds to return to the IIS server and we think it is related to the MQ.

The IIS/MQ server and application servers have no issues regarding network, memory or storage. Everyone seems to be processing fast and without any problems.

I am wondering how I can both sniff messages and debug the MQ by sending my own messages (based on the sniffed data) to try to reproduce the latency troubles. Also if anyone has any good pointers in how to troubleshoot this I'd gladly accept any tips.

Our building is located approx. 100 meters from the explosive charges. They happen several times per day, and really shake the entire building a lot. This is going to go on for many days and the blasts are supposed to get stronger.

Our server rooms are nothing fancy; one of them has all the racks on hard concrete while the other one has a raised floor (the one which allows the cables to go beneath it).

Does anyone have any tips, countermeasures or best practices for us?

Currently we are thinking of the following countermeasures:

• Daily report of the server rooms status lights (HD lights, power supplies and so on).
• Nightly check disk scan on the most important servers
• Order in extra supply of spare harddrives

Edit: Many good answers here! However one needs to be accepted. The highest voted answer at the time of this edit will get their answer accepted.

Does anyone have any good tips on debugging connection leaks on GlassFish?

I am currently debugging a GlassFish server which contains an application that has memory leaks sporadically, eventually leading to having to restart the server to free the pool (pool purging is only available in Glassfish 3 I was told).

When I debug these kinds of errors today I look through the server log around the time when I can see that our monitoring has reported that the pool has grown and see if there is any interesting errors or similar around this point.

Does anyone have any better ways of debugging problems like these?

All the applications are EJB/EAR's which use annotation to inject datasources. The applications then use the datasource to get a connection. Ofcourse every connection is supposed to be closed by the programmers, but I guess someone has forgot it in a case here.

I need to set up ssh sessions between two servers and do not want to make the script fill in username and password everytime.

However I cannot seem to find out where the SSH server is using its config from.

bash-2.05# ssh -V
Sun_SSH_1.1, SSH protocols 1.5/2.0, OpenSSL 0x0090700f

I have certificate files file in both /etc/ssh/ and ~/.ssh/. I can only find one config file for SSH and it is in /etc/ssh/ssh_config, but it contains no data (everything is commented out).

Does anyone know how I can find out where the certificate is stored, or how I can export it so I can transfer it to the other server? I was hoping the config files would give me the answer, but they provide little to no help.

I want to make sure all directories and sub directories are of the group hgdata, but I do not want to change owner's (they are many different owners).

The command chown -R will do the recursive ownership change for me, but how do I use chown without the userid supplied?

On one of my servers we have a lot of scheduled tasks running java programs and what not. The problem is that everytime one of these runs the command window pops up and distracts any work that is being done on the server.

Does anyone know how I can prevent the command windows from popping up?

I have a case where I need to find all connectable shares on my network, and preferably as much information about the share possible. I could do this manually but its quite a big network and it would be too slow.

If I did it manually Im guessing I would do something like

net view
net use //hostname
<browse it manually>

This would however not give me hidden shares so its not a viable option.

Does anyone know of a tool which can help me out in this case? I already tried Sysinternals ShareEnum but it did not work properly. It did a half decent job, but it gave me access denied on tons of shares that was actually open.

Any tips in how I can script this is also appriciated.

Thank you