We are currently looking into buying a SAN, and have some concerns over the speed of disks we are going to put into the array. We would normally have gone for 15k disks over 10k every time, but one of the vendors has told us we wont notice much difference between the 2, and the price difference is quite different.
We are looking at either a HP 2040 or 4400 san, with SAS disks. Will we not see much difference between the disks, and if not why?
I appears we have become a victim of not using the /XJ in a robocopy command, and we have ended up with a mass of nested folders.
We want to get rid of them, but we can use the normal methods of deleting with long files, as we dont have permission to delete the files, and we can seem to set the permissions all the way down as the folder name gets to big!!
Has anyone got a clever way of getting round this, i am thinking we need some sort of script to set permissions (using takeown or similar) until it hits an error, then stop and start again where it failed, so the path gets shallower every-time?
We have an office opening, and we aren't going to have comms into the building when management want the building to open.
Our only option (I think) Is to try and hook up a 3/4g dongle to something to act as the access point, and send all the traffic via that. The model of router we use wont support the usb dongle, so we need some sort of 'bridge'
My idea was to build a Linux box, plug the dongle into that and then via the Ethernet ports plug the router in.
We need the Cisco router in the equation as we create VPN connections over that back to head office. My question is will this work?
We have a really poor business application that uses access databases (I know, fail!) as its back end. There are lots of access db files it uses, all about 4mb in size.
If I copy the files locally to a server, it performs a lot better than if the files are sat on a network smb share, the files are all on the same box, I have just tested by changed c:\files\ to \server\files.
Its probably about 50% faster when pointing locally. The disks are raid 1+0 15k, so they should be fast, running on windows 2008 (not r2) on x86 architecture. There is nothing else on the box, or anyone else accessing it.
Anyone got any ideas why it would be so much slower accessing the files over unc, and any tip I might be able to look into to speed it up?
Cheers Luke
To put our Windows 8 KMS key onto our KMS server running Server 2008 R2, it appears that we first need to remove the existing Windows 7/Server 2008 key.
Is this the case? If so, do the new Windows 8/Server 2012 KMS Keys still allow older operating systems (Windows 7 and Server 2008) to be activated?
We are setting up a new proxy based on squid on an ubuntu server, and would like to have pass through authentication work for the Windows/Internet Explorer client.
We have put the line into the squid.conf for squid_ldap_auth, but this prompts for a username and password in internet explorer. It does work ok once the user puts it in.
Whats the 'best' (standard) way of using pass through authentication?
Cheers Luke
I have recently imaged a machine, and I have got this error when it gets to the installing OS stage -
windows could not display the images available for installation
I've never come across this error before, i've imaged lots of machines in the past but never hit this before, and i cant find much information about it
We seem to have an issue running squid on our windows box. We are using NTLM to authenticate users on our active directory, but it seems t be 'slow' for users that are accessing it over the WAN. Users on the same subnet as the server seem to be ok.
We think it is an authentication issue because when we turn it off it starts to be a lot better. This is the section of the config that sets up the authentication -
#Authentication Params
auth_param ntlm program c:/squid/libexec/mswin_ntlm_auth.exe
auth_param ntlm children 50
# Programs to get the groups out for the user
external_acl_type win_domain_group ttl=120 %LOGIN c:/squid/libexec/mswin_check_lm_group.exe -G
external_acl_type win_local_group ttl=120 %LOGIN c:/squid/libexec/mswin_check_lm_group.exe
Anyone any ideas what might be going on? Looks like we are on version 2.7
Cheers Luke
Possible Duplicate:
How do I deploy MS hotfix KB959628 exe/msp via a gpo easily?
We have a hotfix that we need to put onto our windows xp machines to get Single Sign on for web apps working. I cant add the hotfix to WSUS as it wont appear in the update catalogue.
What is the best way of pushing this out, we have no other 3rd party installers, am i going to have to create a GPO software policy to do it?
EDIT - the file is WindowsXP-KB953760-v2-x86-ENU.exe
Cheers Luke
We have just setup squid as our proxy, and i was going to use Sarg to analyze the log files. I had initially set the Squid logs to rotate everyday so they dont get huge. The problem is i cant see an option in the squid config to read a folder full of squid log files (say *.log).
Is there an easy way to do this or am i going to have to write a bash script or something to process them all into one before i get squid to read it?
Cheers Luke
We have been running MRTG for quite some time to keep an eye on our router usage at remote offices. We now have a need to see roughly how much data (pref in GB) has been transferred over the link.
Can I get MRTG to tell me this info, or am i going to have to use something else?
Cheers Luke
Possible Duplicate:
Office 2010 silent activation after unattended installation
We are looking at deploying Office 2010 over machine startup/shutdown scripts. I have created a msp file and configured config.xml with all the relevant options, and the install from a batch file works fine.
The issue I have is I want to install on a shutdown script, so the users turn their machine of at night and in the morning when they come in office is upgraded.
I have got the it running via a gpo shutdown script, but the issue I have is its not activating the install (we only have a MAK key so i cant setup a KMS server). I asume its because as the install is happening in the shutdown script it cant get out over the web.
The first time the user starts office its running through the process fine and activating the install, but I really dont want the users to have to do this.
Anyone any ideas?
Cheers Luke
We have recently put a new Exchange 2010 server in place, and have moved most of our mailboxes and public folders over to it. Our clients are all still on Outlook 2003.
Most things are working ok, they only issue we have is slowness in public folders. We have quite a few public folders the users work in, where they flag/move/etc. emails.
For example, when they are flagging on an email it can now take about 10 seconds for the flag to actually appear on the message for the user that has clicked it, so they are getting a bit frustrated as this never used to happen.
I have found the article about enabling UDP messages again (http://support.microsoft.com/kb/2009942) which we have done, but it didn't seem to make any difference. Although I'm not sure if we have to restart after applying the reg key.
Has anyone any ideas what might be causing it, or how we can speed it up?
Has anyone ever added a link into the nagios alert messages that allows the receiver to just click the link to put the alert into the acknowledgement state. I've had a look around but cant see anything.
I've got a couple of servers (ftp) where the host is blocking ICMP so its showing as down, but the services (like ftp) are up.
To stop the service complaining about the host being down i have acknowledged the problem. The issue i had was the other night the FTP service died, but we didn't get a notification about it .
How can i still get notifications about service status on a host that Nagios thinks is down, if i disable the active checks will this solve it?
Cheers Luke
We are monitoring quite a lot of devices using Nagios, and the host group page page is starting to get a bit cluttered. Does any one know of a way to force the order of the groups on that page, the main reason being i want the groups with the most hosts to appear first as i think the page will look a bit cleaner
Thanks Luke
we currently use an in house developed login script system. There is a pretty admin GUI tool that exports config files, and a dll called from a vbs login script that processes the config files.
The current system is now getting a bit clunky, it was written in vb 6, runs a bit slow and lacks some functionality the helpdesk would like, so we have been tasked with re-writing it.
I was wondering if anyone else ran any similar in house login applications, had any good ideas about the best way to run them, if there are any apps out there now that do the job etc
UPDATE We are trying to do the normal login script functions, map drives, merge reg keys, copy files etc. We are large organization and a single script would be to large to handle all the different roles we have (they are all different between sites/depts etc).
We need a single point to manage it all from, what we have now does work but needs updating
Luke
I am starting to use nagios to monitor some of our systems, and i was wondering if anyone had ever come across a way to make it monitor directories for files.
I want to look in a directory and say there are x number of files in there that are over x mins old, so warn me.
Cheers
We have recently moved our production terminal servers from stand alone blades into a data centre running on virtual servers(windows 2003).
We are starting to suffer quite major performance issues, and one of the main culprits seems to be IE7, and users launching multiple copies of it instead of using a new window/tab. Its taking up quite a large memory footprint.
Has anyone come across anything we can use to stop them launching multiple instances of the same application?
Ideally you could tell it what to watch for, and when the user tries to open a second copy it will stop it happening, and tell them why.
Anyone know of anything, please bear in mind it is running in a terminal server environment so needs to work per user (each user can have one IE process open)
Cheers Luke
I have a few iis servers (windows 2003) I would like to monitor, and I would like the information presented in graph form if possible.
A couple of the servers i have full access into so I should be able to query using snmp, but some of them all I will have access to is the iis log files (not my kit).
I like some basic traffic info (connections/bandwidth etc) but also maybe something to monitor physical processors/ram usage on the internal ones i have full access to.
Any one point me at a good product
Cheers Luke