- Well, I have no idea who my Novell rep is...and haven't for several years.
- Plus Attachmate bought them. No idea what Attachmate does or who they are (Yes, I found the website, read, etc, but seriously, no idea still).
- Nor had there been any real contact between our two organizations. Having to negotiate our contract each year was becoming a nuisance. Otherwise our only contact was cutting a check every Sept or Oct. that doesn't doesn't count.
- Constant threats of increased cost so we'd best buy 3 year committment. And were below of the min requirement for our SLA so each year was truly a hassle.
- It was the only Novell product we had left from our transition to MS environment 2 years ago.
- Personnel transition. One of former team members took on a new role with another company. Awesome for him, but put us in a bit of a lurch. Growing people is good, but sometimes they leave. Happens. Part of the deal and I wish him well. How he kept that thing running for 2+ years, I'm not sure but I'll always appreciate it.
- Our back-end database that Zenworks was running on was constantly having to be adjusted. Part of that was the fault of the initial installation. We really had no MS SQL skills and it is a database that needs care a feeding.
- SLOW LOGINS. 37 or so pieces to uninstall zenworks. CASA anyone?
- Upgrading for 10.1 to 10.2 to 10.3 at the client level never worked well in mass deployment. So we only had partial registrations and updates taking place. Ever try to push to clients who don't register with home. Yeah, doesn't work.
- Integrated features sets with our Help Desk and our imaging and deployment.
- We are dropping the renewal on our help desk software and switching this summer
- The help desk has customizable rules, labels, etc that can automagic a lot of our calls
- We already had a lot of the items in place in terms of images, applications (msi), PXE environment
- User profile migration. Not usually a pro or con for us, but if we can do it simply, it will be used. People with a title that includes Superintendent occassionally prevented us from using a re-image.
Now the fun begins. Watching people learn new roles is ...an experience. Trial by fire has been our way. At least we let you learn from a fire hydrant of knowledge. Here comes all the info, fast and furious. We are getting 24 hours of online training. The trainers have been good and knowledgeable. Plus, they haven't wasted our time.
how's this working out btw?
ReplyDeleteIt has gone pretty well. We had personnel transition along with the new product learning curve. We've moved over 12000+ machines to KACE successfully. Our apps we reviewed and migrated as necessary. the biggest hitch is lack of multicast imaging. Trying to throw down 6 sets of 30 machines in a lab at a time doesn't work. The tools for scripting actions are fairly easy to figure out for a variety of support staff. I've been fairly pleased with the progress and capabilities of the product. We are starting to use more and more of the application including software licensing and asset tracking. it is more intuitive than zenworks to me. The integrated help desk is good too.
ReplyDeleteAll in all, the first 60 days were painful. after that, it has gone well.
More detailed update on this process now that I've taken the time to regroup and think on it.
ReplyDeleteThe initial roll-out: It was a bit time-consuming. We had to find all sorts of ways. Use Zenworks, use GPOs, use your desktop support staff (sorry bout that boys and girls). Like anything that you try to hit 10000+ units you will have some wierdness.
Pro-tips: if you go SSL on your communications, don't hose your deployment/KACE guy buy changing about 30 days into it. Get SSL working right off. Encryption works well, just have to reconfigure the client. Nothing more fun than your app guy visiting and saying all 6000 units we've imported aren't communicating. Those oh h@#$ moments in life. Also, on SSL/redirectors and load balancers, be careful. the workstations really really want to talk the KACE directly. Placing your load-balancing appliance with an SSL cert in the middle to save money cost us a bit of time on the backsliding.
Training from Dell/KACE. The guys who did our training were good as was my app guy who was learning from scratch. They record the session. they cover a lot of information. they give best practices. Make sure you are uninterrupted for this. Can't say enough good things about this. Worth it.
Imaging goods and bads. I'm not the imaging person. I just get to hear the good and the bad and the totals done. The imaging works well as a whole. the biggest short comings we had were with the number of simultaneous images we could kick off before the server(s) hosting the images were brought to their needs. Dell's devices allow you to host files on servers besides the K2xxx box with pointers and such so they clients get the info from the nearest servers. We reached speed issues around 30 clients. Luckily, my server people are excellent and could spin another 2k8 server up, add the pointers and we had another box able to server images going quickly. Until then, it was kinda funny watch the WAN guy go "wtf, why is campus A having 780M traffic to campus F servers". Made me laugh at least. No multicast support hurt here. We could have done images to thumb drives, but we chose the vmware server spin ups.
More later...
The next part is pushing apps to your clients once they have the kace client. Again, I'm not the app guy, i just listen to the stories. And i tinker, but luckily, my app guys in my career as director have been nothing short of outstanding. Anyhow, back to the story telling. KACE's policy and roles allow you to deploy applications depending on time-frame (2xxx during images, 1xxx post). I stayed out of imaging portion of the apps. the apps got on the machines when asked. Kace allows you to group machines such as by model, by type, by memory, by version, by OS, by IP range, group users, group apps and apply rules (or deployments) based on these requirements. You don't have to know SQL to do this either! Being a lean group, I like simple rules. (or i'm easily confused, probably depends who you ask). We used machine location based rule deployments such as if lab x - install apps 1-5, if lab y, install apps 1,2,3,5 etc. For our users, we just used some basic group info since you can import users and groups from AD into KACE. (remind me to rant on a missing feature or knob). I guess that all goes to say, KACE has numerous options to distinguish devices and users and apply information appropriately.
ReplyDeleteWhat topic to cover next...