30 odd year old, Married, IT and coffee junkie
Posted in Uncategorized on October 8, 2014
I haven’t done a post in a while, but this issue had been bugging us for ages, and now we have finally found a fix due to a very helpful forum post.
Basically our set up is as follows;
vSphere 5.1 (Latest patch as of Oct 2014) (VMXNET3 VM adapter)
Server 2012 R2
Exchange 2013 CU1
Outlook 2010 / 2013
We were finding that users were clicking from email to email and there was a lag of 2 seconds or so whilst clicking.
Not only that but also sending emails and whatever else Outlook does.
It didn’t seem to affect Windows 8 users, but did for Windows 7.
Follow this forum post… it worked brilliantly
Posted in Uncategorized on May 16, 2014
I thought I would just update a few of you to the issues we have had installing Office 2013 with KMS.
Firstly, we use Server 2008 R2. We had VAMT 2.0.
I installed the patch so that the server would recognise Office 2013, I installed VAMT 3.1, I then installed the volumelicensingpack patch.
I got a count of 3, but this would never rise.
I should give you a little background to the issue. We use VMware View Virtual desktops, and had 3 gold images. After spinning up the gold images, with 5 clients in each, expecting to see 15 KMS count, we had only 3.
After a brief looking around the internet and 2 weeks of wondering why we had exceeded the licensing period and were still getting the ‘The software licensing Server reported that the computer could not be activated. The count reported by your KMS is insufficient, I have found the issue.
Basically when VMware View spins up new machines and uses the prep method, it does not create a new CMID, so the KMS server believes that all the machines are the same one, despite having a different IP and AD computer account.
The fix is to either spin up a load of gold images, or install it on some physical machines, or laptops or something…
Here is a link to the article
Posted in Uncategorized on July 30, 2013
I’ve recently had the need to route specific VLANs using a switch.
Many would ask, ‘Why not just use a router?’, but obviously the ability to have mulitple VLAN interfaces on a switch is a common occurence nowadays and therefore this is where I will show you how to route a specific VLAN using a switch.
Firstly, out of the box, the Catalyst 3750 acts as a switch. All of its resources are used to be a switch and very little else.
The first job we need to do is to make sure the switch runs as a router. It needs to be able to reserve some of its resources to routing based commands and procedures, so firstly to change the switch to the routing template you issue the following command;
Switch(config)# sdm prefer routing extended-match
Switch# wr mem
Proceed with reload? [confirm]
It may be worth noting that the ‘reload’ command issues a cold reboot tot he switch, so just make sure you aren’t doing this in the middle of the day or something like that, without scheduled downtime!
You can find a more detailed list of the changes made when switching the template at the following URL – http://www.cisco.com/en/US/docs/switches/lan/catalyst3750x_3560x/software/release/12.2_53_se/configuration/guide/swsdm.html#wpxref88774
Once the switch is operational again, the first thing to do is to jump in to configuration mode, and create the route-map. The route-map will contain the next hop for the specific interface / VLAN.
An example of this would be
CAT3750(config)# access-list 10 permit 22.214.171.124 0.0.0.255 CAT3750(config)# route-map pbr permit 10 CAT3750(config-route-map)# match ip address 10 CAT3750(config-route-map)# set ip next-hop 126.96.36.199 CAT3750(config)# int vlan 3 CAT3750(config-if)# ip policy route-map pbr
This would create an access list by which a condition is made, (for example the source IP must match whatever is defined), and also the location. I would generally leave this open to any, just for ease unless it is needed.
It would then pass to the VLAN 3 ID, and look-up the route-map policy to say, you are from here, you should hop to this IP, instead of the default ip route specified on the router.
It’s a great tool for using for when trying to divert some traffic over a slower connection, and more heavy traffic over a better connection.
Posted in Uncategorized on June 12, 2013
Ok, so, in my wonderful hazy ignorance, I recently was doing a clear up of a profile list on a Server 2008 Terminal Server.
On first look, I found the C:\Users\ folder to contain several hundred versions of the same profile (named .001, .002 etc). So I proceeded to delete them all, keeping the most recent one.
On a couple of profiles I must have deleted all the folders because when it came to deleting them it came up with an error that the files were in use and this included the NTUSER.DAT. Anyway, once the user logs on it loads them with a temporary profile.
Now, If i’d of been clever and looked on the net I’d of known that the registry just creates a version of the SID in the profileslist key in the registry, and I could just rename this to .bak and we’d be away. But this is where the similarities end…
In my wisdom I remembered how Windows 7 works, and that you can just delete the registry key HKLM\software\microsoft\windows nt\current version\profilelist\ and then the users SID and all is fine. Wrong, not in this case anyway. Whether this is because it is Server 2008 or not, I don’t know but when the user went to log on I got the error ‘The user profile service failed the logon’ and then the RDP session logged me out.
I’ll keep this short and sweet and push straight to the answer, rather than bore you with the 2 hours of trying different things I tried.
I first, logged the user on to a Server 2008 server, a normal generic one, I then logged them out and logged in myself. I copied their profile from this server back in to the terminal server. I then logged in to regedit and found their SID in the profilelist key. I right clicked and clicked export. I dumped this on the terminal server and logged into the terminal server and doubled clicked the reg file to add the values back in to the terminal servers registry. Got the user to log back in and all was fine!
Thank god for that…
Posted in Work on April 11, 2013
I’ve not updated my blog in a while but thought I would write about how safe it is to consolidate snapshots.
I’d heard conflicting reports of how users had had problems about consolidating snapshot chains within VMs that are being used as ‘Gold’ images within VMware View.
Take the following VM snapshot chain…
VM1 >> Team1 >> Updated Apps >> Updated OS1 >> Added App5 >> Updated OS2 >> You are here
I have found that it is completely safe to click on say ‘Updated OS1’ and click delete. What this will do is consolidate where you updated the OS in snapshot ‘Updated OS1’, in to ‘Updated Apps’.
Basically, the tree moves sideways. All the information from that particular snapshot that you deleted, will consolidate in to the one before.
The last thing you need to do is make sure the tree is up to date when it comes to provisioning more Virtual Machines.
Open VMware View Admin console, naigate to the Inventory > Pools > Select the relevant pool > Edit > vCenter Settings > Snapshot, and just relselect the snapshot you wish to deploy from. It may already be up to date with the latest Snapshot Tree information, but just in case I would do this again.
Hope this helps and removes some fear from removing Intermediary snapshots!
Posted in Uncategorized on May 29, 2012
It would seem as though the iPhone development teams throughout the world are getting wise to people jail breaking their phones.
What I fail to realise is what gain a company like Capital One gets from stopping you checking your credit card balance when your phone is jailbroken.
There is quite an easy fix for this. Search Cydia for a package called xCon. Install it and restart the app which is causing you trouble!
Posted in Work on May 14, 2012
I don’t know if anyone else has had this problem with ESXi 5.0.0 but I am having an issue where the event logs are showing that connection to a Thecus NAS device is dropping every 5 seconds to every 5 minutes.
The VMX configuration files int he inventory change to ‘Unknown’ and then roughly around 10 seconds later they come back. The VM 99% of the time doesnt turn off but I have known it to.
The way that the standalone ESXi server is set up in this configration has worked perfectly with previous versions and I can only put down this ‘latency’ message to fcat that I have installed or upgraded I should say to ESXi 5.
The server has 2 NICs that were teamed to host all the VMs on a 10.0.x.x range with the NAS device also on the same range presenting the storage via iSCSI. I know that VMware recommend that the storage be presented on a separate IP range but this has worked before.
I have now tried having 1 NIC set up for ISCSI vmkernel and 1 for VM Network and binding the sofware iSCSI adapter to the vmkernel port for iSCSI but no luck. Next is to add an ip range for iSCSI after that I’m stuck and it will be a fresh install of ESXi 4.1.0 which I know works. The only issue I might have is that I’ve already formatted the LUN that stores the VMs as VMFS v5 😦