cerved

46 ms·

691.
by cerved 5y ago
while it can be a pain to deal with windows quirks, writing a wrapper script to deal with low level file IO stuff sounds like a good way to open a whole can of worms and irrevocably mess up your whole system
692.
by cerved 5y ago
> I'd be worried about an infinite process loop here. I mitigate this in several ways, first, only one worker can work on a job or child job. At worst, one worker is stuck in a loop. There's also a maximum limit of retries. So
693.
by cerved 5y ago
I think all that is needed is an aha moment
694.
by cerved 5y ago
https://thenewstack.io/git-transitioning-away-from-the-aging...
695.
by cerved 5y ago
That's what makes it good :) I had no idea why it took my brain so long to wrap my head around it. Maybe it was blissful ignorance, never sitting down and making a mental model of it. Always looking for a tldr version of doing things.
696.
by cerved 5y ago
They're not saying the company name because they don't want you going there directly and applying without them getting their hefty bonus
697.
by cerved 5y ago
Right post, wrong platform
698.
by cerved 5y ago
I can't speak to billing since that's not my area but in general I disagree. A process doesn't need a non-trivial flow to need a queue and workers. I've done this with route optimization requests. They have a priority an
699.
by cerved 5y ago
What about in terms of number of requests and requests per second?
700.
by cerved 5y ago
How would double billing occur if the worker dies. The way I would design this, the billing request and bill would be committed atomically such that a request can only be completed with exactly one associated bill. If the worker dies, no bi
701.
by cerved 5y ago
In your experience, roughly at which point did it stop scaling and why?
702.
by cerved 5y ago
no you don't but it's a lot better to do an atomic select/update IMHO. Basically you have status = 0,1,2 for queued, processing, done. You have a filtering CTE that returns the queued rows, sort by id and update status = stat
703.
by cerved 5y ago
It's a matter of information security. Error messages potentially disclose sensitive information on the internal workings of the service. Disclosing that information is a potential security vulnerability. Also note that SE went down ye
704.
by cerved 5y ago
I remember seeing a compiled list of hard drive failure rates for different models, I think it was backblaze. It was awesome
705.
by cerved 5y ago
Since selling ads is how they make money I don't think there's much need to worry about frivolous blocking
706.
by cerved 5y ago
It's pity we're excluded from Freedom Tracking
707.
by cerved 5y ago
You don't even know the guy is employed at Google. Tons of support is outsourced
708.
by cerved 5y ago
This. The discrepancy between GA and our data was always something I had to adress back when I sold AdTech.
709.
by cerved 5y ago
It's not a stretch to imagine Google being able to test their own products and how they interact with each other better than others. Which in and of itself may be considered unfair.
710.
by cerved 5y ago
This is different because Google makes money from AdSense. Like all of it. The guy is trying to run AdSense campaigns. AdSense runs on what, millions, billions of websites. It has presumably very complicated system for detecting malicious l
711.
by cerved 5y ago
I mean you're writing with a chat support agent. Presumably, you're not spending remotely enough money to even be communicating with an agent who is hired by google. Support people are busy, dealing with all kinds of bullshit, wit
712.
by cerved 5y ago
I think you're reading far too much into this. You spoke with a support agent who is checking the status of your AdSense campaign. That system has flagged links on your website as 'malicious'. The agent is simply telling you
713.
by cerved 5y ago
Because they are a support agent who is reporting to the customer a bunch of links that their computer screen is telling them is not allowed.
714.
by cerved 5y ago
ah, yes, CSVs. The format famously preferred and lauded by the web development community
715.
by cerved 5y ago
That sounds incredible inefficient What was the rational for such enormous single payloads?
716.
by cerved 5y ago
possibly Azure?
717.
by cerved 5y ago
That's great, for you. The suit isn't alleging that raising the price to pay for free shipping is illegal. The suit is about controlling the market and hiking up the price to pay for your free shipping across the market. Ie, it&#x
718.
by cerved 5y ago
I'm not familiar with US antitrust legislation but usually these laws stipulate that you can either be "unfair" or have no competition, but not both.
719.
by cerved 5y ago
I thought it was because the default buy box choice wasn't the cheapest
720.
by cerved 5y ago
probably because there's a lot of anal downvoters on HN

More ›