chronid

17 ms·

151.
by chronid 9y ago
Steady state with a lot of requests/s is surprisingly expensive and surprisingly hard to calculate before-the-fact (due to the strange ways ALB is billed, for example).
152.
by chronid 9y ago
Gotta deliver that MVP, right? :)
153.
by chronid 9y ago
Replication is for pussies, and we have no money anyway for infrastructure. YOLO.
154.
by chronid 9y ago
> I don’t operate production systems, but I have helped to design a couple of them. So I understand something about the assumptions you make when building them. Start by operating production systems. You will rapidly discover that patchi
155.
by chronid 9y ago
That's the same in FreeBSD: > (The algorithm used is to try a name server, and if the query times out, trythe next, until out of name servers, then repeat trying allthe name servers until a maximum number of retries are made). So it
156.
by chronid 9y ago
> DNS is supposed to be reliable. UDP does not strike me as a reliable protocol. Tthe glibc behavior is the standard on Linux, and FreeBSD (I don't know about the other BSDs) has the same behavior IIRC. Changing the de facto stand
157.
by chronid 9y ago
good > perfect is somewhat standard thinking in AWS.
158.
by chronid 9y ago
Well, the major companies are the one pushing to pay less and less the outsourcing companies, aren't they? I really want to know what BigCorp will chose the first between a company paying janitors above minimum wage and a company offer
159.
by chronid 9y ago
Data center engineers, IIRC.
160.
by chronid 9y ago
Judging by the reaction "diversity" also seem to boil down "fill the place up with people who think like me, the person calling for diversity". Also, extrapolating anything from a text we cannot see rapidly brings us in
161.
by chronid 9y ago
Nice. I had to move back to Xorg just because of that in 25. Time for an upgrade I guess :)
162.
by chronid 9y ago
Did they fix the copy-paste issues between application with wayland in Fedora 26 (particular between terminal and the rest of the world)? But maybe that is a more of request for the GNOME guys...
163.
by chronid 9y ago
Here: http://blog.darknedgy.net/technology/2015/10/11/0/ For a less politically correct one: http://suckless.org/sucks/systemd For a personal experience: I remember that until
164.
by chronid 9y ago
Breaking expectations is not a reasonable default.
165.
by chronid 9y ago
> A couple months fear of being “witch hunted” is a small price to pay for what appears to be years of harassment women at the company had to endure. Two wrongs don't make a right.
166.
by chronid 9y ago
Patching is not a technical problem, 90% of the time. It's a political one (in medium to big operations). We have let our civilization down? No, our civilization is still asleep believing than when we say that patching is important, th
167.
by chronid 9y ago
I'm sorry: it's your application, your alerts, your responsibility. Not Ops.
168.
by chronid 9y ago
You can use whatever system you want (syslog can drop the logs in a single box, no need for horrible tail-based concoctions), as long as you mantain it and it's not Ops responsibility to fix your AWK scripts that look for events in log
169.
by chronid 9y ago
Ops is maybe trying to help developers and increase security (no login on the boxes needed anymore to look at logs). Better workflow for (certain kind of) alerts, which may then get more complicated. Maybe kibana dashboarding. That in turn
170.
by chronid 10y ago
They are working in tech. For someone else.
171.
by chronid 10y ago
On November 8th, the day of the US election, the whole world got more black and white. People in the center have been called upon to choose sides. In a way, my position is an appeal to preserve some of the gray in the world. A
172.
by chronid 10y ago
I don't think "silicon valley" is worried. I think the body shops are worried, and the writer of the article just conflates the two (because he doesn't really know - or is not interested to know - the difference?).
173.
by chronid 10y ago
You usually don't want to add another dependency you don't control to your system, if you don't have to (in term of time and resources). It's another point of failure, more infrastructure you have to keep alive...
174.
by chronid 10y ago
3am saturday morning. Big critical system goes down, multiple teams involved to fix the mess so conference call needed, I get paged into it too. My phone wouldn't have lasted the 4-5 hours that call lasted without being connected to th
175.
by chronid 10y ago
I feel like we are swapping cause and effect here.
176.
by chronid 10y ago
Heavy discounts are usually given for this kind of issues by fiscal authorities, unless there is proof (beyond doubt) there was unlawful intent to not pay those taxes.
177.
by chronid 10y ago
They will not collect back 100% anyway, maybe maybe 1/3 of the total amount.
178.
by chronid 10y ago
> I’m not sure who said it first, but I’ve heard a number of people say that RSS solved too many problems to be allowed to live. I don't know too, but this is a phrase worth stealing.
179.
by chronid 10y ago
I don't really see why a single standard is necessary (and this thing about self-contained packages makes me cringe a little). Package managers have always cross pollinated and flourished thanks to that - as an example RPM does not s
180.
by chronid 10y ago
I do that too. My following visits are usually due to some random link posted here, reddit, social networks or through friends or colleagues. If I did not do that I would probably have to completely stop commenting on news and opinion piece

More ›