|
Today I came back from retaking a foundation level exam for software testing, and I somehow ended up performing WORSE. The first time I took the exam I got a 57; now after even harder studying (and more time to study) I got a 47. I also realized that I had taken the wrong standard exam for the second time (ASTQB) instead of ISTQB, which I had done coursework on. My father is adamant in forcing me to take a third exam within a week because I NEED THAT CERTIFICATE FOR THAT INTERVIEW still. Do I?
|
#
?
Jul 17, 2018 20:51
|
|
|
|
| # ? Jun 1, 2024 00:30 |
|
|
Kashuno posted:I'm getting terrible flashbacks Could be worse, could have asked to use Access.
|
|
#
?
Jul 17, 2018 20:51
|
|
|
Zil posted:Could be worse, could have asked to use Access. I am fighting to do two major decommissions of Access databases right now. One of them is going smoothly with the dept involved really helpful and interested in utilizing better tools. The other is a huge fight about how I am completely disrupting the procurement process and I need to "get my loving head on straight."
|
|
#
?
Jul 17, 2018 20:58
|
|
|
Kashuno posted:Is there actually a good reason to not just let DNS handle your poo poo most of the time? Like we have some industrial equipment we assign static IPs to because the poo poo will go weird otherwise, but when I started here the guys above me were super serious about every time I spun up a VM it absolutely HAD to have a static IP and would get mad if I forgot/didn't bother to, and now that I'm running the poo poo I can't find a good reason to do so as a general practice. for some reason you have a flaky DHCP server, like off site or something? Anything I can think of is a stretch involving really stupid design decisions.
|
|
#
?
Jul 17, 2018 21:03
|
|
|
Kashuno posted:I am fighting to do two major decommissions of Access databases right now. One of them is going smoothly with the dept involved really helpful and interested in utilizing better tools. Well, they're right, you are completely disrupting their process. The question I'd have for you would be: what are you going to be replacing it with? And how is that new thing better than the old thing? There are 1 billion things wrong with Access, but at the end of the day, their "process" may not encounter any of them. Their process may be smooth as butter, without a hiccup in the last 20 years. On the other hand, if they come and say: I need to do X, make it happen in Access, then sure, tell them to go gently caress themselves. Or if the CEO wants to save money by not having 100 people part of a simple process that can be automated, then again ... gently caress'em.
|
|
#
?
Jul 17, 2018 21:10
|
|
|
Undoing dumb design decisions that were made to cover for even dumber design decisions has been about half my week so far. Doing DHCP over a flaky WAN link? Stop doing it, and/or fix the WAN link.
|
|
#
?
Jul 17, 2018 21:11
|
|
|
Thanks Ants posted:Undoing dumb design decisions that were made to cover for even dumber design decisions has been about half my week so far. Doing DHCP over a flaky WAN link? Stop doing it, and/or fix the WAN link. How hard is it to have an on site dhcp server with a scope that doesn't overlap? It's just a dumb design decision
|
|
#
?
Jul 17, 2018 21:16
|
|
|
I think people get too close to the coal face and become unable to think. If you have poo poo internet at a location and it's going to cause issues with all the cloud applications that you've already evaluated then you fix the shithouse internet problem, you don't compromise every other part of your strategy because the idea of fixing the problem in front of you was something that never came up.
|
|
#
?
Jul 17, 2018 21:26
|
|
|
Volguus posted:Well, they're right, you are completely disrupting their process. The question I'd have for you would be: what are you going to be replacing it with? And how is that new thing better than the old thing? Yeah definitely the back half. It is a person using Access to pull data from an unsupported SQL server that may or may not have accurate or complete data which they are using to base serious financial decisions, and they can't make their numbers tie out with the numbers provided by our FP&A folk, despite their repeated claims they got approval from old management. It has been great.
|
|
#
?
Jul 17, 2018 21:46
|
|
|
Thanks Ants posted:I think people get too close to the coal face and become unable to think. If you have poo poo internet at a location and it's going to cause issues with all the cloud applications that you've already evaluated then you fix the shithouse internet problem, you don't compromise every other part of your strategy because the idea of fixing the problem in front of you was something that never came up. Oh yeah. I worked at a remote office for a few years. When I started, email, DHCP, and DNS were hosted in the home office, over a completely saturated Frame Relay connection. When the HR director came up, did a presentation, and emailed an 8 MB deck to everyone, email was effectively down for hours. Then the fun happened. The home office shut everything down to add a generator to the power circuit. I didn't get an outage notification. The office was full of people working through the weekend on a major pitch. My VIPs really didn't like hearing "LA turned off all our poo poo. No, I'm not coming in. I can't fix it." I will give them credit for taking it though. My advice to not reboot anything that already had an IP address for any reason at all helped.
|
|
#
?
Jul 17, 2018 23:59
|
|
|
22 Eargesplitten posted:Looks like it’s new old stock for one from 2011, which would explain it. EBay for a Dell PowerConnect 6248P. You’ll find them for less than $200. link It’s a managed layer-3 switch with 48 ports that are PoE capable. With the external PSU module it’ll do PoE to all 48 ports. Without it’ll power 24. 4 fiber uplinks in the front and slots for either stacking modules or four 10gig SFP+ fiber uplinks in the back for your Mellanox ConnectX-2 HBA goodness. Drop a ConnectX-2 card into a FreeNAS box and three ESX6 boxes and you have a home lab with a 10gig ethernet backbone. The switch can have problems with stacking, but since you only need one for the home this is a solid workhorse of a switch. I’ve had mine for about ten years now and it’s been rock solid. Also, the 40mm fans can be loud, but you can buy lower rpm fans that decrease the noise significantly without increasing the temps in the unit much. However, lower RPM fans will trip the fan failure light and send traps if you care about such things, so be warned. Agrikk fucked around with this message at 07:32 on Jul 18, 2018 |
|
#
?
Jul 18, 2018 06:24
|
|
|
DHCP chat - Generally I have servers on something static and then user devices/pc's are all done via DHCP. It's the way i've always seen it done and i've had it taught to me, but i cannot for the life of me remember their reasoning, its been years and every place i've worked since has done the same thing. Helps that I currently only take care of a couple servers and a few VMs so its easy to remember. Off the top of my head the only reason I can think of having a static IP is predictability, but that doesn't matter if the people designing whatever your server is being used for account for it rather than calling a specific IP. Our current developer (he's been here 30 years, will probably be wired up to the place when his body finally gives out) mentioned his code calls specific addresses, so gently caress me if i'm ever getting it changed.
|
|
#
?
Jul 18, 2018 12:05
|
|
|
dogstile posted:Our current developer (he's been here 30 years, will probably be wired up to the place when his body finally gives out) mentioned his code calls specific addresses, so gently caress me if i'm ever getting it changed. We had a guy that did that at an old job. WHY THE gently caress DONT YOU JUST USE DNS poo poo they even had some weird thing where you had to let him know if a user's IP changed so he could change something in the code. This also meant if a user had a laptop and a desktop, they had 2 IP addresses and 2 user accounts 
|
|
#
?
Jul 18, 2018 13:13
|
|
|
IPv6 is good because it forces people to use DNS and stop giving a poo poo about what the actual IP address behind something is. Getting people to use IPv6 and give up this idea of addressing things sequentially is another matter entirely.
|
|
#
?
Jul 18, 2018 13:21
|
|
|
dogstile posted:DHCP chat - Generally I have servers on something static and then user devices/pc's are all done via DHCP. It's the way i've always seen it done and i've had it taught to me, but i cannot for the life of me remember their reasoning, its been years and every place i've worked since has done the same thing. The justification I was always given was that if a server for some reason became unreachable by host name, you could still hit it up by IP address. Also the fact that some ancient production software systems didn't function well via hostname (but that was our special snowflake case). Oh and we also had this problem where people would clone VMs with duplicated names and bounce production severs off the domain.
|
|
#
?
Jul 18, 2018 13:29
|
|
|
So I’m looking to replace nginx oss edition in front of my swarm cluster with something else that routes stuff from my F5s to the containers. Looking at haproxy and traefik. Any other recommendations? I’m missing health checks and don’t want to pay 15k per year per nginx instance just for that. It has to play nice with Jenkins and preferably doesn’t need restarts for config changes to take effect (reloads are fine). It’s for about a thousand containers so no insane sizing issues.
|
|
#
?
Jul 18, 2018 16:04
|
|
|
so this is going to probably be really dumb but I am honestly way more new to the cloud than I want to be, and I think it's a good time for my company to explore some cloud options. I spun up an Azure VM, used AADDS to join it to AAD which is the domain $COMPANY.com and our local domain is $COMPANY.net I set up a site to site VPN from Azure to our local network so I can RDP from our local network to the VM. I need to access a file server from the local domain on the Azure VM. I can ping $FILESERVER.$COMPANY.net but I obviously can't reach it if I just do ping $FILESERVER How do I get that to work? Is it due to the DNS suffix being different? Do I have to add that DNS suffix to the Azure VM?
|
|
#
?
Jul 18, 2018 16:04
|
|
|
Edit: never mind
George H.W. Cunt fucked around with this message at 16:24 on Jul 18, 2018 |
|
#
?
Jul 18, 2018 16:14
|
|
|
LochNessMonster posted:So I’m looking to replace nginx oss edition in front of my swarm cluster with something else that routes stuff from my F5s to the containers. Looking at haproxy and traefik. Any other recommendations? A vote for HAproxy. That service is bombproof and scales well. I can’t think of an enterprise customer of mine that doesn’t use it in some fashion in their prod environment.
|
|
#
?
Jul 18, 2018 16:33
|
|
|
dogstile posted:Our current developer (he's been here 30 years, will probably be wired up to the place when his body finally gives out) mentioned his code calls specific addresses, so gently caress me if i'm ever getting it changed. That should be a fireable offense tbh
|
|
#
?
Jul 18, 2018 16:46
|
|
|
Methanar posted:That should be a fireable offense tbh Chances are the dude is the last person with the knowledge of how the old rear end production system functions and can't get fired.
|
|
#
?
Jul 18, 2018 16:54
|
|
|
LochNessMonster posted:So I’m looking to replace nginx oss edition in front of my swarm cluster with something else that routes stuff from my F5s to the containers. Looking at haproxy and traefik. Any other recommendations?
|
|
#
?
Jul 18, 2018 17:01
|
|
|
Vargatron posted:Chances are the dude is the last person with the knowledge of how the old rear end production system functions and can't get fired. He's actually the only guy who knows the system period. He made the whole thing. Boss has a hard on for custom poo poo that we entirely own. Dudes gonna get hit by a bus and then we're all hosed.
|
|
#
?
Jul 18, 2018 17:21
|
|
|
dogstile posted:He's actually the only guy who knows the system period. He made the whole thing. Boss has a hard on for custom poo poo that we entirely own. Dudes gonna get hit by a bus and then we're all hosed. Get out. Now.
|
|
#
?
Jul 18, 2018 17:35
|
|
|
dogstile posted:DHCP chat - Generally I have servers on something static and then user devices/pc's are all done via DHCP. It's the way i've always seen it done and i've had it taught to me, but i cannot for the life of me remember their reasoning, its been years and every place i've worked since has done the same thing. We need to be careful here to differentiate between "unchanging, predictable IP address" and "IP address hardcoded on the device itself" when talking about static IPs. There are plenty of good reasons to ensure that anything providing a network service has predictable IP addresses that don't change semi-randomly because a lease table got corrupted or something was down long enough to expire. Dynamic DNS systems introduce another potential point of failure, some bad software doesn't support DNS, other bad software only looks up a DNS name once and never again until it's restarted, etc. Also as Vargatron noted the ability to connect directly on a known IP address can be useful from time to time if DNS ends up broken for whatever reason. There are almost no good reasons to hardcode that address on the machine itself. DHCP reservations are objectively superior because they make your addressing inherently self documenting and prevent address conflicts. The only devices that should actually have their addresses hardcoded are the DHCP server(s) and anything that needs to be up for DHCP to function properly. I generally consider that to mean the router(s) and in some cases DNS and Active Directory server(s). Anything else providing network services or otherwise benefitting from a fixed address gets a reservation. The couple of static devices also get reservations just for the sake of documentation. On the IPv6 side SLAAC handles this well automatically, on any given network the IP address of a specific interface should always get the same address because the host portion of the address is built from the MAC address of the interface. For those who feel the need to organize their IP ranges explicitly DHCPv6 still allows classic style leases and reservations. tl;dr: Most network-serving devices should have a static IP address, but almost nothing should actually be set on the device to "static IP".
|
|
#
?
Jul 18, 2018 17:40
|
|
|
Thanks for the recs, I was leaning towards HAproxy since it’s been around for ages and always has been bulletproof but didn’t want to narrow my choices to studf I know about. I did look into Istio and Envoy a little bit but that sounds like over the top for my use case (at least right now).
|
|
#
?
Jul 18, 2018 17:53
|
|
|
dogstile posted:He's actually the only guy who knows the system period. He made the whole thing. Boss has a hard on for custom poo poo that we entirely own. Dudes gonna get hit by a bus and then we're all hosed. what in the gently caress
|
|
#
?
Jul 18, 2018 17:57
|
|
|
I loving love it. We have a bunch of poo poo written in QBASIC that management refuses to pay someone to rewrite. Good times ya'll.
|
|
#
?
Jul 18, 2018 18:06
|
|
|
GreenNight posted:I loving love it. We have a bunch of poo poo written in QBASIC that management refuses to pay someone to rewrite. Good times ya'll. Our super critical central app is written in Progress from (I believe 1992). With 25 years of code updates. I think there's a modernization effort underway, but it's taking a long time, for obvious reasons. We're a financial company big enough to have a superbowl ad every year, and an even bigger financial company pays us $texas in licensing fees to use this software. It's insane.
|
|
#
?
Jul 18, 2018 18:34
|
|
|
Sometimes this thread makes me feel like my current situation is super dire. Others times it makes me feel like I'm living in a pod. I guess that's a good balance. Thanks thread.
|
|
#
?
Jul 18, 2018 18:41
|
|
|
Internet Explorer posted:Sometimes this thread makes me feel like my current situation is super dire. Others times it makes me feel like I'm living in a pod. I guess that's a good balance. Thanks thread.   
|
|
#
?
Jul 18, 2018 18:43
|
|
|
Here’s something to make you feel like you’re in a pod: I’ve apparently been rude to enough VPs without realizing it that I now have to run all communication with users by thee assistant manager. And I’m also purely support now, no projects. Guess this tells me I need to kick my job search into high gear.
|
|
#
?
Jul 18, 2018 19:35
|
|
|
22 Eargesplitten posted:Here’s something to make you feel like you’re in a pod: I’ve apparently been rude to enough VPs without realizing it that I now have to run all communication with users by thee assistant manager. And I’m also purely support now, no projects. Are you intentionally being rude or short with people? Like I work with a couple guys that come across in email as giant assholes, and even they haven't gotten to this point....
|
|
#
?
Jul 18, 2018 20:00
|
|
|
22 Eargesplitten posted:Here’s something to make you feel like you’re in a pod: I’ve apparently been rude to enough VPs without realizing it that I now have to run all communication with users by thee assistant manager. And I’m also purely support now, no projects. Maybe try not speaking to people like an rear end in a top hat instead of assuming a new job will fix it?
|
|
#
?
Jul 18, 2018 20:02
|
|
|
22 Eargesplitten posted:Here’s something to make you feel like you’re in a pod: I’ve apparently been rude to enough VPs without realizing it that I now have to run all communication with users by thee assistant manager. And I’m also purely support now, no projects. This is part of the reason I never want to deal with Execs and VIPs. I just *know* I'm going to be too blunt/honest/not gargling their cock enough. skipdogg posted:Are you intentionally being rude or short with people? Like I work with a couple guys that come across in email as giant assholes, and even they haven't gotten to this point.... CLAM DOWN posted:Maybe try not speaking to people like an rear end in a top hat instead of assuming a new job will fix it? Execs are notoriously thin-skinned superior royalty.
|
|
#
?
Jul 18, 2018 20:09
|
|
|
So I wanna throw my Dell Rep out the loving window right now. It turns out they dropped the price of our main monitor 7 months ago and didn't tell us or change it on our premier site in any way. I just ran the numbers and we got overcharged roughly 15 grand.
|
|
#
?
Jul 18, 2018 20:12
|
|
|
22 Eargesplitten posted:Here’s something to make you feel like you’re in a pod: I’ve apparently been rude to enough VPs without realizing it that I now have to run all communication with users by thee assistant manager. And I’m also purely support now, no projects. Who cares, gently caress them. I once spotted an issue before it became a catastrophic failure. I informed my manager and stated it should be fixed asap because it would have a severe impact on critical business processes. Two weeks later my manager told me that the manager of the team who caused the issue thought it was gigantically disrespectful of me mentioning it to my manager and I should apologize to the guy who caused it. I told my manager both of them were insane and I wasn’t apologizing in this universe. He asked me to go talk to the guy, so I did. Turns out he didn’t know why he had to talk with me either and we both came to the conclusion both our managers are loving idiots and he was glad I spotted the issue before it got out of hand.
|
|
#
?
Jul 18, 2018 20:14
|
|
|
LionYeti posted:So I wanna throw my Dell Rep out the loving window right now. It turns out they dropped the price of our main monitor 7 months ago and didn't tell us or change it on our premier site in any way. I just ran the numbers and we got overcharged roughly 15 grand. Sounds about right. Dell has been an absolute clusterfuck over the past 5 years or so. They used to be a lot better to deal with than HP or IBM, but these days it's pretty much a wash. At least in my experience. LochNessMonster posted:Who cares, gently caress them. I hear what you're saying and obviously his place could be toxic, sounds like it probably is, but having to communicate through your manager due to what is being perceived as poor bedside manner is not a good thing. I have never been asked that and I have never had to ask someone to do that. It shows a pretty serious lack of confidence and I would say isn't something someone is going to ask lightly, just because of the extra hassle they are asking to take on. I mean, I'm in pretty much "gently caress everything, everyone is on notice" mode right now and have been for some time at my current place. But it's not good and I would sooner let someone go than tell them they had to communicate with users through me, ignoring things like official communications and the like.
|
|
#
?
Jul 18, 2018 20:23
|
|
|
Darchangel posted:Execs are notoriously thin-skinned superior royalty. That literally doesn't matter. Treat everyone with kindness, patience, and respect. Don't be an rear end in a top hat. That doesn't change with how smart you are, how correct you are, how busy you are, etc.
|
|
#
?
Jul 18, 2018 20:47
|
|
|
|
| # ? Jun 1, 2024 00:30 |
|
|
Darchangel posted:This is part of the reason I never want to deal with Execs and VIPs. I just *know* I'm going to be too blunt/honest/not gargling their cock enough. I don’t get this. It is possible to interact with executives without getting fired or getting your emails proofread. It’s called being nice. It is possible to disagree with an executive with a hardon for technology A while you are CERTAIN that technology B is the correct choice. Just be nice. A little bit of mindfulness and self-awareness goes a long way to not coming off like a truculent rear end in a top hat or a stereotypical nerd who prefers a wiring closet to people. Remembering that executives put their pants (or skirts, or pantsuits) one leg at a time helps, too.
|
|
#
?
Jul 18, 2018 20:49
|
|