|
Thanks!
|
#
?
Sep 21, 2018 03:50
|
|
|
|
| # ? Apr 27, 2024 04:51 |
|
|
freeasinbeer posted:In theory I don’t disagree with that premise. Work doesn’t really have the need to see what I am doing. As long as my work product(code in this case) is free of bugs or exploits I think the pendulum is a little too strongly swung towards employers being up in your poo poo. Assume your employer is looking at everything and don't try to circumvent it, but if they're going after people for stuff like personal browsing or how many hours you're active on Skype you probably don't want to work there anyway. This is presuming you're not entry level and can find another job, if that's the case probably just suck it up and use your personal phone off the company wifi. Same deal for breaking poo poo. Do you really want to work somewhere that's going hold you over a barrel because of a legitimate mistake? If you're in fear to the point of lying or covering up your tracks, again, time to check out the job market. On the other hand, just get approval for your changes and you've got a paper trail to cover yourself.
|
|
#
?
Sep 21, 2018 04:11
|
|
|
GreenNight posted:We did the math and on site exchange is cheap as poo poo compared to 365. Runs in a vm and no issues ever. How’d you run the numbers and how many users?
|
|
#
?
Sep 21, 2018 04:23
|
|
|
AreWeDrunkYet posted:Same deal for breaking poo poo. Do you really want to work somewhere that's going hold you over a barrel because of a legitimate mistake? If you're in fear to the point of lying or covering up your tracks, again, time to check out the job market. On the other hand, just get approval for your changes and you've got a paper trail to cover yourself. Having an approved change is not necessarily a bulletproof shield but it goes a long way toward making it about the flawed process that made that change. When someone lies about their actions and is discovered, the failure instantly becomes that person regardless of the technical mistake, design flaw, or the company culture. I watched this happen to someone with a problem that was very clearly architecture plus an easy-to-make commandline mistake.
|
|
#
?
Sep 21, 2018 04:50
|
|
|
Tab8715 posted:How’d you run the numbers and how many users? 400 users and we had softchoice run them. We signed a new licensing contract so we’re done until 2020. We’ll revisit it then.
|
|
#
?
Sep 21, 2018 05:17
|
|
|
We only have 300 users at my company but 0365 makes way more sense than our poo poo exchange environment. Then again, the guys in charge of our VM environment don't really seem to know what they are doing. Maybe exchange could run happily but every non hosted environment i've been in has had more downtime than hosted. gently caress I can't wait to migrate. please god. please make the email outages stop.
|
|
#
?
Sep 21, 2018 05:31
|
|
|
GreenNight posted:We did the math and on site exchange is cheap as poo poo compared to 365. Runs in a vm and no issues ever. I find it unlikely that you are doing the math correctly and/or honestly. I also find it unlikely that you are estimating your costs for failures into it either. There are lots of reasons for doing email on prem and none of them have to do with huge cost savings over any reasonable time table.
|
|
#
?
Sep 21, 2018 05:33
|
|
|
Haven’t had exchange go down since we upgraded from 2010 to 2016. Before that it was when the internet failed and that’s once in a blue moon. It’s been extremely solid for us.
|
|
#
?
Sep 21, 2018 05:37
|
|
|
GreenNight posted:Haven’t had exchange go down since we upgraded from 2010 to 2016. Before that it was when the internet failed and that’s once in a blue moon. It’s been extremely solid for us. My sweet summer child. You have missed the point of what I posted so much that its not worth going into it. I wish you the best of luck with your email.
|
|
#
?
Sep 21, 2018 05:40
|
|
|
Our exchange admin was laid off because we didn't have enough work for him after we finished migrating to O365. That's my O365 cost comparison story.
|
|
#
?
Sep 21, 2018 05:45
|
|
|
Did I mention we're basically 3 offices with a few sales guys eternally? Also that upper management mandated 2.5 gig mailboxes as max size? It works great, never goes down, and licensing is cheaper. I don't see the problem.
|
|
#
?
Sep 21, 2018 05:54
|
|
|
I'm all for getting rid of responsibilities and we're going to be migrating from Exchange 2010 to O365 next year, but I can't say I've had a problem with Exchange in a good 5+ years. And an "Exchange admin?" Is this 10+ years ago or do you have 10k+ users or something? It pretty much "just works" these days.
|
|
#
?
Sep 21, 2018 06:09
|
|
|
GreenNight posted:Did I mention we're basically 3 offices with a few sales guys eternally? Also that upper management mandated 2.5 gig mailboxes as max size? It works great, never goes down, and licensing is cheaper. I don't see the problem. If you have an existing exchange set up thats stable, requires no maintenance, and is approaching the end of its amortization period, that's fine you can keep it. There is no reason for any sort of green field installation of on-premise email in 2018. Email is a solved problem. See also: CI
|
|
#
?
Sep 21, 2018 06:12
|
|
|
Methanar posted:If you have an existing exchange set up thats stable, requires no maintenance, and is approaching the end of its amortization period, that's fine you can keep it. There are definitely reasons but they are not usually technical or financial, but regulatory or compliance. You're way too new and naive at IT to be making authoritative blanket statements like you tend to do.
|
|
#
?
Sep 21, 2018 07:08
|
|
|
CLAM DOWN posted:There are definitely reasons but they are not usually technical or financial, but regulatory or compliance. You're way too new and naive at IT to be making authoritative blanket statements like you tend to do. What regulatory requirements are not yet met by Office 365/google apps? I actually don't know
|
|
#
?
Sep 21, 2018 08:04
|
|
|
You basically get free email and you pay for an office upgrade every 4 years with O365. Why not switch.
|
|
#
?
Sep 21, 2018 12:47
|
|
|
In my experience, companies that consider on prem cheaper are not counting the cost of their hardware. 'we already have it, we're not spending money, it doesn't count!'. But that doesn't take into consideration the need to replace that hardware without downsizing, or preventing the deployment of new projects that could have used those resources instead, or allowing other VMs to get resource boosts to improve performance. Everything on your hardware should be evaluated on percentage of resources consumed versus the cost of the infrastructure, including networking, storage, and bandwidth. That's when o365 and other cloud apps make financial sense.
|
|
#
?
Sep 21, 2018 13:03
|
|
|
Thanks Ants posted:I think maybe the first day is an opportunity to see what the lie of the land is regarding policies and how they are enforced etc. and not just decide to do whatever you want and then detail how much worse you could have made things if you wanted to. This. Also regarding mistakes. I'm on team "tell the truth and resolve,help resolve asap" I still take the words my previous boss told me "I'd rather you try new things and make mistakes than play it safe and do nothing"
|
|
#
?
Sep 21, 2018 13:09
|
|
|
I'm all for offloading everything we can to Microsoft. I'm typing that as I *checks notes* sit on a early morning conference call for a Sharepoint outage. Every department is on the call except for Microsoft.
|
|
#
?
Sep 21, 2018 13:10
|
|
|
The Exchange VM is pretty minimal resources. We have 3 ESX boxes. We're not going to 2. We're not downsizing our EVA to an MSA. Even with our DAG at 3 sites, it's minimal.
|
|
#
?
Sep 21, 2018 13:10
|
|
|
a ticket came in to me, the IT Director. A sales rep demands that I schedule time with him Monday morning to discuss his extensive computer issues that he has supposedly already been having for 10 days, and that he can't put in an email because it takes too long to type, and that he hasn't contacted the helpdesk about at all. I don't think it ended the way he hoped.
|
|
#
?
Sep 21, 2018 13:16
|
|
|
You could have stopped at "sales rep".
|
|
#
?
Sep 21, 2018 14:03
|
|
|
18 Character Limit posted:I watched this happen to someone with a problem that was very clearly architecture plus an easy-to-make commandline mistake. Remember the time someone took down S3 in this exact way?  I am very glad my company has a mature attitude toward mistakes. Unless it was outright malicious or just shockingly dumb, you aren’t going to get in trouble or be looked down on. We’ll talk about what happened, and figure out how to improve the systems and process so that mistake can’t happen again. I don’t really want to work someplace that isn’t like that at this point. because I am bad and make mistakes all the time But don’t abuse that trust and support. Don’t hide your screwup. Own up, learn, and move on.
|
|
#
?
Sep 21, 2018 14:07
|
|
|
I had a coworker who used the wrong script to install firewall rules which inadvertently blocked all traffic for a second. Took the entire network segment down because connections died and had to be rebuilt, including database traffic. We all understood what happened, another team didn't give us all the info we needed about what had been 'pre-migrated' to the segment, but we realized we didn't have a process to double check impact scope before running this change. The team was pretty calm and collected, understanding that mistakes happen. Boss was terse and demanding solutions for a better process, but wasn't angry. Coworker was nearly hyperventilating in the chair next to me, slowly melting away into grief and despair. He was more upset than anybody in the room. Poor dude hadn't made a production impacting mistake before and probably thought he was getting fired. Of course he wasn't, the process was improved, everybody moved on.
|
|
#
?
Sep 21, 2018 14:21
|
|
|
I've made a few mistakes that caused some issues in the past. I remember knowing exactly what I was doing and changing some port tagging on an HP switch that wasn't working right. Then I lost connection to the switch and every printer on that side of the building stopped working. People were giving me funny looks and asking me questions while I ran down to that networking closet to reboot the switch. Luckily I have good management that covers for me/backs me up. The rest of the company sucks major rear end, but our IT dept is great. Being one of 7 guys in a largish local company, we all kind of have to do a little of everything. I'm glad I learned early on that saying "I'm not comfortable doing this. I think I know how to fix it, but I might make it worse because I'm really not qualified on this particular technology." was completely acceptable and understandable. IE knowing what you don't know.
|
|
#
?
Sep 21, 2018 14:24
|
|
|
Internet Explorer posted:I'm all for getting rid of responsibilities and we're going to be migrating from Exchange 2010 to O365 next year, but I can't say I've had a problem with Exchange in a good 5+ years. And an "Exchange admin?" Is this 10+ years ago or do you have 10k+ users or something? It pretty much "just works" these days. This is pretty much how I feel about it. I am confident in my abilities to run an on-prem Exchange setup but...I don't really want to. Most of the stuff we run on-prem now isn't exposed to the internet and it always nags at the back of my brain that my Exchange box is gonna get owned by some hacker and I'm going to be totally hosed.
|
|
#
?
Sep 21, 2018 14:37
|
|
|
Methanar posted:What regulatory requirements are not yet met by Office 365/google apps? I actually don't know Just one example is BC's FIPPA act (other provinces have similar legislation) concerning data sovereignty. O365/Google Apps are not Canada-only, so building and maintaining on-prem email systems is required by law.
|
|
#
?
Sep 21, 2018 14:43
|
|
|
Everybody fucks up. It's just a part of professional development, but you have to actually learn from your fuckups. I've been responsible for at least 3 lengthy outages trying to "fix" something or correct an error. Sometimes poo poo just doesn't work correctly. But I was above board with my management group and made sure I presented them with "how this will not happen again" after every event.
|
|
#
?
Sep 21, 2018 14:52
|
|
|
Methanar posted:What regulatory requirements are not yet met by Office 365/google apps? I actually don't know Not necessarily regulation but they can process FISA warrants and the like without your involvement which can be a pretty big deal depending on your company's work.
|
|
#
?
Sep 21, 2018 15:07
|
|
|
One time I accidentally took our SAP Test server off the domain, didn't realize it didn't have the same local admin setup as all the other servers on our network and had no way to log into it at all. I ended up restoring a backup 
|
|
#
?
Sep 21, 2018 15:13
|
|
|
Methanar posted:What regulatory requirements are not yet met by Office 365/google apps? I actually don't know For government stuff, it has to do with guarantees on where the data will actually be hosted. A lot of government agencies and government contracts require data and hosting remain domestic. I know Clam has mentioned in the past that MS couldn't or wouldn't guarantee data for Canuckistan wouldn't stay in Canada, so government agencies couldn't use O365. e: well, poo poo. Answered more than once above me.
|
|
#
?
Sep 21, 2018 15:17
|
|
|
Same problem with Slack, data sovereignty. Which is ridiculous considering it originated here in Vancouver.
|
|
#
?
Sep 21, 2018 15:21
|
|
|
We spent months questioning o365 due to ITAR requirements. There's a ton of stuff about no foreign persons handling any data.
|
|
#
?
Sep 21, 2018 15:27
|
|
|
Vargatron posted:Everybody fucks up. It's just a part of professional development, but you have to actually learn from your fuckups. This is also an important point. I had a guy working for me at one point who rebooted the wrong server three times in the same year. The first is a right of passage. The second is a whoopsie. The third time I sat him down and told him that if it happened again, I was going to put him on a performance improvement plan. Check the hostname before you reboot, put the hostname in your window title, put the hostname in your prompt, but please do something to reassure me that you understand that you know this is not a mistake you can keep repeating.
|
|
#
?
Sep 21, 2018 15:31
|
|
|
Zorak of Michigan posted:My employer sent me to TOGAF training and I left thinking that I would rather take a career step back than take one forward into a role where they took TOGAF seriously. I don't want to say it was a complete waste of a week, but it was maybe 15 minutes or 2 slides worth of useful information spread across a week's worth of jargon, buzzwords, and repetition. Hmm, that doesn’t sound like I’m going to enjoy it all. I’ll give it a go and worst that can happen is me being bored out of my brains and sideline any ambitions for becoming a solutions architect for the next few years again.
|
|
#
?
Sep 21, 2018 17:14
|
|
|
Zorak of Michigan posted:This is also an important point. I had a guy working for me at one point who rebooted the wrong server three times in the same year. The first is a right of passage. The second is a whoopsie. The third time I sat him down and told him that if it happened again, I was going to put him on a performance improvement plan. Check the hostname before you reboot, put the hostname in your window title, put the hostname in your prompt, but please do something to reassure me that you understand that you know this is not a mistake you can keep repeating. molly-guard is a very nice thing to have installed on important machines. It hijacks any shutdown/reboot requests made over SSH and requires the user to type in the hostname. It's in Debian and related distro repos, IIRC last time it was discussed it wasn't in RH/Cent officially but was easily available.
|
|
#
?
Sep 21, 2018 17:51
|
|
|
I think we're on the right track here, asking "how could this be possible?" rather than "who made an error?" If "be more careful" is the solution, there might be more work to do. I don't know if this is possible, but I'd love it if pasting on the Linux command line automatically prepended a # to every line to make it impossible to accidentally paste and run a command that you don't want to run or something.
|
|
#
?
Sep 21, 2018 17:55
|
|
|
Vulture Culture posted:detect the person's face and re-render it as a digital Picture of Dorian Gray, tailored to each adjudicated soul, more monstrous the nearer the person is to their print quota This is a wonderful concept. Kashuno posted:a ticket came in The trick is always how to mean "gently caress off", get the person to understand that you mean, explicitly, "gently caress off", and not any gentler form of it, but not actually say "gently caress off."
|
|
#
?
Sep 21, 2018 17:58
|
|
|
Zorak of Michigan posted:This is also an important point. I had a guy working for me at one point who rebooted the wrong server three times in the same year. The first is a right of passage. The second is a whoopsie. The third time I sat him down and told him that if it happened again, I was going to put him on a performance improvement plan. Check the hostname before you reboot, put the hostname in your window title, put the hostname in your prompt, but please do something to reassure me that you understand that you know this is not a mistake you can keep repeating. At a past job, we had some bastion hosts you had to jump through to reach production. The NOC team sat on those jump boxes all day, and would loving CONSTANTLY space out and reboot them instead of the actual host they meant to work on. We eventually literally replaced every command we could think of (suhtdown, reboot, halt, etc) with a symlink to a script that said "HEY NUMBNUTS THIS IS THE JUMP BOX. Did you really mean to reboot it? y/n" and then called the real command. Kind of a silly hack but dammit, it cut down on the unintended reboots! Dr. Arbitrary posted:I think we're on the right track here, asking "how could this be possible?" rather than "who made an error?" Yes, exactly. There's been a lot written about the idea that "human error" should be unacceptable as the end of your analysis of a problem. Go back and make it hard or impossible for a human to make that error in the first place. Blameless portmortems, and John Allspaw's blog, are good jumping off points. In our old, dead book club thread Vulture Culture did a good series of posts about Sidney Dekker's book on the subject of human error. He was writing about plane crashes and industrial accidents and poo poo, not touching computers improperly, but the material is extremely relevant.
|
|
#
?
Sep 21, 2018 18:41
|
|
|
|
| # ? Apr 27, 2024 04:51 |
|
|
Kashuno posted:a ticket came in It's always
|
|
#
?
Sep 21, 2018 18:46
|
|