Best posts made by Reid Cooper

Coffee time. Good morning, everyone.

@openit said in Easier guide to setup Salt Stack/Ansible for Windows environment ?:

@aaronstuder said in Easier guide to setup Salt Stack/Ansible for Windows environment ?:

Here is the official docs for Salt:

https://docs.saltstack.com/en/latest/topics/installation/windows.html

Thanks.
So, Salt Master = Salt Stack Server
Salt Minion = Salt Stack Client

Right ?

The terms Salt and SaltStack are generally interchangeable everywhere that I've seen.

I'd vote for an increase on that.

Could always use it as an excuse to re-image the environment as well. Just take backups and do fresh installs.

Sounds like at least a chat with the big boss is in order.

Interesting. What would this do differently than just using sendmail or postfix, though?

Is that really a surprise

@travisdh1 said in Win 10 VDI: with Linux backend management?:

@Reid-Cooper said in Win 10 VDI: with Linux backend management?:

@travisdh1 said in Win 10 VDI: with Linux backend management?:

Doesn't @scale have a VDI provider they work with? Probably a paid solution tho.

I've seen that. Don't know their name. Don't know if it is Linux based or not. But it's some alternative. Web spot maybe?

It has to work with their (highly customized) KVM back end, so should work with at least one Linux distribution.

Does it? My understanding was that it was an app only, no backend interaction.

It's grilled cheese and tomato soup here today.

@scottalanmiller said in The Motivations of Sales:

Sales people are an important part of any ecosystem. They have a lot of value.

Can't live without sales people. It's what greases the wheels of the world. It's how you tell the world that you have a product that they might want. As long as people look for things to buy, someone will look to sell them things.

@cteneyck welcome to MangoLassi.

I'd suggest a script that just does a ping say every ten minutes or something slightly more random like every 9.2 minutes so that you don't hit the same times over and over again to avoid any problems from being on even intervals. Then just log the ping times over a few days or weeks. That will give you a really good idea of how much it varies and if time of day is a factor.

The shirt is cute. That Nintendo Mini, though, is a sweet prize. I've been wanting one of those.

Idea is cool, the web interface for it is the real winner. You can see what it has been doing.

No kidding, I just got on and there are so many unread posts!

It comes down to how much the machines are used. If all of your CPU usage would be below 10% and you have ten to one over subscription, you are still within the 100% mark. In which case you have just barely enough CPU to handle your vCPUs.

@thwr said in What Are You Doing Right Now:

@MattSpeller said in What Are You Doing Right Now:

@scottalanmiller said in What Are You Doing Right Now:

Lunch time here, my dad is making me grilled cheese and tomato soup!

In my top 10 lunches of all time! Good call.

Sounds good. I know onion soup with toast and grilled cheese, but tomato soup... mh, I guess I will try that someday

Grilled cheese and tomato soup is an American classic, like apple pie, Buffalo chicken wings, hot dogs with mustard...

You might want to consider an SLO, not an SLA. SLA means that the team is seen as an outsider to be managed and to have punitive action if the SLA is not met. SLO means it is an objective that everyone is on the same team and working together to make possible. An SLA means you have zero obligation to do better than the SLA. An SLO suggests that you will do your best no matter what and attempt to not go over the SLO numbers.

@stacksofplates said in What Are You Doing Right Now:

Found out today we are getting two new Isilon clusters. Haven't had a chance to play with them before so this will be nice.

Oh, nice!

@kyle said in Helpdesk SLAs:

@travisdh1 said in Helpdesk SLAs:

@kyle said in Helpdesk SLAs:

@scottalanmiller said in Helpdesk SLAs:

@ambarishrh said in Helpdesk SLAs:

@scottalanmiller said in Helpdesk SLAs:

@ambarishrh said in Helpdesk SLAs:

we are a very small team now, 250+ users 3 techs. Screen connect plays a bigger role here!

Idea is to set an SLA and review this on regular basis, find out the gap and either increase headcount or if no budget for additional headcount, then increase the SLA.

With the current tech headcount, the SLA that we are thinking are:

*Resolution includes temporary fix or work-around solution.

New hardware requests usually take 4-6 weeks to deliver from HP (we can keep stock of 5 but the user requests are quite a lot this year and next year as well, but no finance approval to keep more than 5 as of now), so agreed with HR on this timeline.

An SLA means that there is a punishment to you if you do not meet the requirements. What happens to the IT team if they cannot resolve these issues in the SLA time? How do you determine what is within the scope of what IT can have fixed and not?

The purpose for this is also to evaluate the existing issues, see if our small team can cater the requirements and based on the results we either get more people or increase the sla. As of now this will be visible only to IT team and based on our analysis we will release to users on a later stage with SLA that we can achieve

SLAs are not a good tool for that. Just use good reporting for that. SLAs are adversarial, not something you ever want to use internally.

Yet they are integrated in several helpdesk options.

Yes, because help desk software is a generic thing that can be used with the public, internally, or a combination of both. Just because a feature is already integrated doesn't mean it's a good option to use. I think of something like an SLA for leased lines like T1/T3.

I know. I just wish they were more granular in features as there are some hard coded features the are unnecessary and just clutter the classification of tickets. We currently use ManageEngine Service Desk and SLA's are built in and part of the reporting metrics despite the fact we don't use them.

I generally prefer light, simple helpdesks. Unless you are a huge organization, most of that extra stuff is just wasted and in the way.