Andi

Yeah I guessed that WSUS wouldn't support MySQL. I'll look into anyways of keeping the databases patched automatically rather than manually running updates all the time. Thanks.

Hi all, it's been a while since I posted on here due to moving jobs, but I'll try and get back into it. Currently we only offer the option of SQL databases here but we're thinking of expanding this to allow MySQL onto our network. Does anyone have experience of how this works? Other than just installing it onto the necessary servers? The primary things I'm looking at are: Backups - can MySQL be backed up using Backup Exec (my research so far tells me not directly using an agent but can be using scripts to stop and start the relevant services). Patches - currently we patch our servers using WSUS, presumably this won't support MySQL databases, but is there a similar option available to centrally patch these databases? Security - are there any security issues with having MySQL running on a primarily windows network? Thanks as usual for any advice that can be given.

We're having a funny issue here where users who log on to a station with office 2007 installed are seeing the global address list, but then that same user will login to a station with 2003 installed and they can only see their own contacts. We haven't fully migrated to 2007 yet, as we're doing it slowly to allow staff and students to get used to it first. We're running Exchange 2003 server, and nothing was changed in this when the stations were updated to Outlook 2007. Can anyone help please?

Adobe is the least of my worries though. The appdata redirection is causing Word to hang on our wireless clients. I don't know why, but users that do not have appdata redirected work fine.

This still isn't filtering through and becoming a royal pain now, I'm constantly having to recreate user profiles. The appdata just isn't moving itself back into the roaming profile. Is there a way I can recreate everyone's user profile, so that it ignores any local copy when they login next and creates afresh?

Our citrix presentation server 4 used to map the clients local printer with no problem at all. now seemingly with nothing changed it doesn't. I've been through the troubleshooting steps on the citrix knowledgebase but nothing kicks it back in. Is there anything that I can check out that might disable this function without me changing anything knowingly?

The current setup we have is: Citrix certificate server on a DMZ which passes authenticated users through to a citrix presentation server on our internal network. That all works a treat, but I don't know exactly how, therefore troubleshooting why when I convert the presentation server to a vm image and fire it up live people can no longer login remotely. I can only presume that there's somethign hardware specific that is either on the certificate server or the presentation server. Something like the certificate server looking at the MAC of the presentation server? I'm guessing at what it could be, but after looking through the settings on both servers I can't see anything obvious. It's also difficult for me to find out exactly what the error was when we tried it as the physical server is back up and running again at the moment so that people can use it....

You may actually be a genius! So simple I hadn't thought of it. Now I just hope it works

Thanks, but won't that still apply the change in ownership to the staff folder? I need the script to ignore that folder all together.

I need to change the ownership on a large amount of folders back to the administrator so that I can delete them. They are old redirected appdata folders. However I need the option of excluding the folder 'staff' which is sitting amongst these folders. Is there anything I can do with cacls or similar to achieve this? Doing it by hand will take hours.

OK I've tagged the uplink port with my wireless vlans, and removed the tag from the downlink port. Also I have removed the tag for the downlink port for all the wired vlans. I can now connect to all of my wlans!!! THANKS! Now to test the speed with a class of kids....

I see, I didn't know what the CDP and CUP were until now, I presumed is was some protocol that needed to be tagged. I have CDP tagged for ALL of my vlans!!! I'll try playing with that for the non-working SSID.

I can only see CDP and CUP. Our wireless module is in bay C so could that be the reason for that? It is not tagged on that uplink port, but then neither are the working ones.

When setting the radios to ACS mode it's also asking me to set the power. Currently they all appear to be on 20, is that normal?

OK, set both of those up. Will test and see how it goes. On a related note, have you any clues why when I connect to one of our SSIDS won't find an IP address, despite all the settings that I can see are the same as the ones that work? I tried creating a test SSID and that wouldn't connect either so I've obviously missed a vital step somewhere in the wireless setup (I managed this successfully for the wired portions of our network with this sites help).

Yes all the RPs are licensed.

I haven't seen the option to scan for neighbours, I will look for that now. The APs were all set to auto channel selection, but I have recently enabled advanced mode so that I can select which SSID gets sent through which AP, which I get the impression takes away a lot of the automation. The problem doesn't actually seem to be related to DHCP as the laptops get an IP with nor problems (apart from one VLAN, which I'm working on separately), I put that in my original post as it is one of the things that has changed before the wireless network slowed down.

Last year we had a company come in and setup a wireless network in our science block for 24 laptops. He setup 4 aerials on their own VLAN, secured with WPA2, with the wireless module of the switch acting as a DHCP server for that range, that all worked a treat and the laptops were logging onto the domain with no issues. Recently I've setup a lot more vlans at the school, using Microsoft DHCP server, and HP procurve switches with the IP helper address pointing back to the DHCP server. Also we have added 26 more access points to the network, making a total of 5 SSIDs with their own VLANs. The wireless module still sends and receives information from those APs on VLAN2100 (default for the module). Hardware info: Wireless edge services module mounted in a procurve 5406zl chassis. APs are HP Procurve 200 series. The problem is that the wireless networks have all slowed down since the new ones have added, so much that they are almost unusable. Could it be: The fact that we now use DHCP server and the ip helper-address function on the core switch? The fact that each AP is pushing out the 5 SSIDs I'm at a bit of a loss here, and would like to resolve the issue without calling out a specialist if I can, but I have a feeling I'm a bit out of my depth. I've even gone so far as to actually reading the manual Any help is as always much appreciated.

Resolved the issue, there was a static route on the Watchguard firewall as follows: 172.10.0.0/24 - 10.0.0.1 Removing this route solved the problem. Thanks Watchguard for telling me to put that route there in the first place.

OK, I have changed the default gateway on the computer that will eventually be using the VPN to the firewall IP address. Pings and trace routes still timeout but now I am getting an error message on the firewall log: 2008-10-13 12:03:00 Deny 10.0.0.13 172.10.0.1 icmp-Echo 1-Trusted unknown packet with TTL=0, firewall drop (internal policy) rc="104"

I'm not sure how to answer your first question. The VPN is a Branch-Office VPN, I'm not sure whether that's a name specific to watchguard or whether it's a standard. I guess it's a lan-lan, connected via ADSL lines. Is that what you meant? Good idea about changing the default gateway, I'll give that a try on a machine now and get back to you. Thanks again for all your advice.

Yes the internal vlans follow the pattern: 192.168.1.x 255.255.255.0 192.168.1.254 192.168.2.x 255.255.255.0 192.168.2.254 192.168.3.x 255.255.255.0 192.168.3.254 etc etc We still have some nodes left on the default vlan due to not having managed switches everywhere - the switches and servers are still on the default VLAN also. The default vlan is 10.0.x.x 255.240.0.0 10.0.0.76 The main switch that looks after the vlans and routes is 10.0.0.76 The watchguard firewall (10.0.0.1) takes care of the VPN tunneling. I know that the VPN is working as I can login from the other side of the tunnel with no problems. The IP range the other side of the tunnel is: 172.10.0.x 255.255.255.0

OK, I removed the static route 172.10.0.0 255.255.255.0 10.0.0.1 using 'no ip route' and entered ip route 0.0.0.0 0.0.0.0 10.0.0.1 as suggested. I still can't get through to my vpn. Tracert still stops at 10.0.0.76 (IP of my main routing switch). Any thoughts?

Does anyone know the command to view the current static routes?

Currently there is no route back to the 10.0.0.0 range setup on the firewall, but as the firewall is currently on the same range as the servers I'm guessing it won't need it? I think I'll wait until monday to play with this, 8 years of experience tells me nothing apart from changing this sort of thing on a Friday afternoon is asking for trouble! Do I need to undo the ip route change I just added before I go? Thanks for your help guys!