|
Memtha wrote: ssh'ing and rdp'ing between severs because it's faster
I was going to suggest that. I've never had a KVM switch that worked all that great even in the VGA days, and knowing how finicky DP/HDMI are, I don't have much faith in KVM switches that support them.
|
|
|
|
|
I still have my Belkin vga. If I wasn't so hopelessly spoiled by 4K I would seriously be buying hdmi->vga and vga-> hdmi adapters to use it again.
|
|
|
|
|
I gave up on KVM for video years ago, but still use KVM for keyboard and mouse. For video I just change the input on the monitor. Which is OK with only two systems (3 monitors) and I would say I generally have to hit the reset button on the KVM about once every couple of months, if that.
|
|
|
|
|
Tempting. Years ago, before I decided to "invest" in a "real" kvm, I had a usb hub plugged into a clicky switch thing designed to share a usb printer between computers, and 3 3-way hdmi multiplexers designed for entertainment centers. Arguably that worked better, but I found it annoying to have to hit 3-4 buttons to switch. But it is very tempting to go back.
|
|
|
|
|
I would get annoyed if the kvm switched all my monitors now, as I frequently have them displaying from two machines, but sometimes go with all three displaying from one. Worst part is if monitors have different ways to change inputs! So I usually buy three of the exact same thing.
A good kvm would deal with all of this but the prices of a good kvm, and the issues of them not working well, has me sticking to my current setup. Until the kvm goes bad (which I dread) I will probably stick to this. I wish I could run 3 machines through a good kvm but I have room to spare so other machines get their own displays and input devices.
|
|
|
|
|
So I usually buy three of the exact same thing.
Same. Black Friday about every five years. Though this time around I am annoyed at a visible contrast difference. Though that's a whole different rant.
Honestly the main reason I don't just go back to input switching is fear of wearing out the little buttons on my monitors.
As far as the kvm going bad, I'm more afraid of the ports it has being outmoded. Again. And what the state of whatever replaces hdmi/dp will be. Already there are talks of thunderbolt/usb-c being used for monitors (already a thing on apple), so I'm afraid to think about a usb-c only kvm that handles multiple monitors and input on a single cable; and what additional compromises they will have to make to do that.
|
|
|
|
|
|
I hate this. Whether or not I start work right away, I really don't like waking up to a code mess - a project that doesn't build or that I'm in the middle of debugging, etc.
I do everything I can to avoid it, but sometimes it happens anyway. Right now I'm facing some indomitable bugs on multiple fronts, multiple projects. In fact, I don't even have a path forward I can make progress on, but for fixing those bugs.
And they are confusing. I've been attacking each of them for days, in the past. I put them down and stepped away for quite awhile too.
That's a tough spot to be in.
But the worst part of all of it is waking up to it. Every day that I do that, I have to be very careful to keep my mood up and not spoil my morning and thus my day, because if I'm not I'll think about it and fret over it even if I'm not actively working on it. The only real out is some sort of mentally immersive activity or meditation that allows me to lose it, but that relief is more temporary than the bugs, so I just try to keep myself occupied until the worst of it passes.
Meh.
I'm guessing I'm not the only one who has run into this. I get the feeling nobody likes broken builds and hanging debugging sessions - especially say, coming into them on a Monday morning (for those of you that do the M-F thing)
Real programmers use butterflies
|
|
|
|
|
I haven't woken up to broken builds but have definitely woken up to debugging sessions for days on end.
|
|
|
|
|
To be honest, I haven't woken up to a broken build in quite awhile, but it has happened when I was in the middle of porting a project.
Real programmers use butterflies
|
|
|
|
|
I see you have your Windows updates turned off. I've had numerous times where Windows has finally decided to reboot my machine to install "critical" updates. There went a 3 week soak session trying to catch a bug. I'm trying to be more generous this year to the clowns, but I swear, whoever made that decision to force reboots needs to meet my 2x4 multiple times.
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
Wow. It's been a long time since I worked on software for which 3-week soak sessions were a thing.
|
|
|
|
|
Some of the systems I work on only turn off once a year. Things like memory fragmentation become very important. It can be a real challenge to track down some customer issues.
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
I used to work on systems like that (telecom call servers). What kind of system are you working on?
I still find it amusing when systems are shut down every week or two for "routine maintenance". 
|
|
|
|
|
Production machines for packaging. Imagine a 200 foot long machine packaging beer or diapers.... you name it. If our product stops running, the entire line goes down. The cost of our machine is insignificant to the cost of the line or the downtime.
So, yeah, like telecom switches and the like. 
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
Interesting. I'd have guessed that the line would be regularly shut down for hardware maintenance, which would reduce the uptime expectations on you. Then again, your software wouldn't want to be the cause of a shutdown.
|
|
|
|
|
Most of the time there is an annual shutdown. In our case, our equipment tends to be wired into the production line, so it's a REALLY big deal to tell the user to cycle power. Many times, our equipment is parked under the line in protected areas. The only way to cycle power is to do it for the entire production line, and there tends to be a LOT of systems connected to that line. Note that if our system isn't working, they aren't making product, I guess that qualifies as mission critical in a light sense of the word.
So, over the years I've learned not to code things in ways we take for granted. For example, malloc's are evil. Skipping the memory leak issue (We'll assume there is a matching free everywhere), this can easily lead to memory fragmentation and a hang down the road. I didn't believe it at first. I'm a believer now.
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
At least it is a "mess" of your own making, one that is probably clean and clear to work with. Right now I am getting ready to deal with cleaning up / fixing a Big Ball of Mud created by another developer. It is definitely a high caffeine morning....
|
|
|
|
|
When it's of your own making, you only have yourself to blame! But cleaning up someone else's Big Ball of Mud was always something that I found delightful.
|
|
|
|
|
Not when that other developer is the "senior" developer still working on that project and is resisting any cleanup (and contributing more to it).
I actually enjoy doing this type of cleanup too, generally. Gives me a chance to stretch my software design muscles to see how best to approach this.
|
|
|
|
|
Gotta love that - especially when your client or supervisor doesn't understand the ramifications of working through a Big Ball of Mud.
TBH the offending code in one of projects is code I didn't write but am using under a compatible license. Except this code is battle tested and I've used it past projects. Somehow the inputs are different I think? but I can't pinpoint where/how. The issue is I don't understand the algorithm being used, and it is very complicated, so the routine is pretty opaque from a debugging standpoint.
So on some level I can relate in my present situation. It's not a big ball of mud, but in some ways it may as well be. 
Real programmers use butterflies
|
|
|
|
|
With four cats in the house, one of which had cat flu as a kitten, we frequently wake up to worse messes than that! 
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Here as well. Our cat here likes chewing on the potted palm fronds and then creating mixed media art expressions on the carpet...
|
|
|
|
|
|
I am in a similar situation now, actually; not the first time, either.
The worse part for me, is debugging sessions that last for days. I take it personal, even if I did not create the mess, and it affects my mood greatly, even though I know better.
I just try to organize all that is broken or not working and create a todo/checklist and work through each one. It helps me psychologically to see some sort of progress.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
