TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

We can't send email more than 500 miles (2002)

1034 点作者 alanpage将近 5 年前

34 条评论

thyrsus将近 5 年前
I&#x27;ve had enough years to become wiser, become a fanatic for configuration management, and get over the embarrassment: I&#x27;m the consultant that screwed things up. Some background: the Stat department was running a variety of systems besides the Solaris workstations, and there was, within UNC-CH, a separate support organization that was cheaper and more comfortable with Microsoft products where Stat was sending their support dollars. When that organization needed Unix support, they called my employer, Network Computing Solutions, and I showed up.<p>There was effectively no firewall at UNC-CH at the time (something something academic freedom something something), and the Stat Solaris machines were not being regularly patched. Uninvited guests had infested them, and it appeared the most likely entry point was sendmail - at the time, it was the most notorious vulnerability on the internet. Since my preference to wipe and reload was unacceptable - too much downtime and too many billable hours - the obvious thing to do was update sendmail. The rest is history.
评论 #23778245 未加载
评论 #23780161 未加载
评论 #23777804 未加载
评论 #23778424 未加载
评论 #23779226 未加载
评论 #23778115 未加载
surround将近 5 年前
Source:<p><a href="https:&#x2F;&#x2F;www.ibiblio.org&#x2F;harris&#x2F;500milemail.html" rel="nofollow">https:&#x2F;&#x2F;www.ibiblio.org&#x2F;harris&#x2F;500milemail.html</a><p>FAQ about the story:<p><a href="https:&#x2F;&#x2F;www.ibiblio.org&#x2F;harris&#x2F;500milemail-faq.html" rel="nofollow">https:&#x2F;&#x2F;www.ibiblio.org&#x2F;harris&#x2F;500milemail-faq.html</a><p>Past submissions:<p><a href="https:&#x2F;&#x2F;hn.algolia.com&#x2F;?dateRange=all&amp;page=0&amp;prefix=true&amp;query=500%20mile%20mail&amp;sort=byPopularity&amp;type=story" rel="nofollow">https:&#x2F;&#x2F;hn.algolia.com&#x2F;?dateRange=all&amp;page=0&amp;prefix=true&amp;que...</a><p><a href="https:&#x2F;&#x2F;hn.algolia.com&#x2F;?dateRange=all&amp;page=0&amp;prefix=true&amp;query=http%3A%2F%2Fweb.mit.edu%2Fjemorris%2Fhumor%2F500-miles&amp;sort=byPopularity&amp;type=story" rel="nofollow">https:&#x2F;&#x2F;hn.algolia.com&#x2F;?dateRange=all&amp;page=0&amp;prefix=true&amp;que...</a><p>(It seems I can’t find a search query which includes both urls)
评论 #23779813 未加载
drtz将近 5 年前
I love that this took a perfect storm of having a statistician and sys admin both bent on finding the cause of a weird intermittent problem in their own way.<p>This could have happened a million times where the story was a lot less interesting:<p>&quot;Hey, I&#x27;m having weird intermittent problems sending email.&quot;<p>&quot;Hmm, we&#x27;re using the wrong version of Sendmail. All fixed, case closed.&quot;
wuster将近 5 年前
Best part of reading this is coming away having learned the existence of units the CLI. How did I spend 20 years on the shell and not have needed or discovered this?
评论 #23777944 未加载
评论 #23777253 未加载
评论 #23780963 未加载
评论 #23779108 未加载
评论 #23785194 未加载
评论 #23777445 未加载
评论 #23777228 未加载
abhgh将近 5 年前
I used to collect these kind of stories:<p>When I flush my toilet my computer reboots: <a href="http:&#x2F;&#x2F;www.techtales.com&#x2F;tftechs.php?m=199712#66" rel="nofollow">http:&#x2F;&#x2F;www.techtales.com&#x2F;tftechs.php?m=199712#66</a> (the first story on the page)<p>If I buy vanilla ice-cream my car wont start: <a href="http:&#x2F;&#x2F;www.netscrap.com&#x2F;netscrap_detail.cfm?scrap_id=501" rel="nofollow">http:&#x2F;&#x2F;www.netscrap.com&#x2F;netscrap_detail.cfm?scrap_id=501</a><p>A specific cargo routing crashes system: <a href="https:&#x2F;&#x2F;www.jakepoz.com&#x2F;debugging-behind-the-iron-curtain&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.jakepoz.com&#x2F;debugging-behind-the-iron-curtain&#x2F;</a><p>Tape-drive failure only within large print jobs: <a href="http:&#x2F;&#x2F;patrickthomson.tumblr.com&#x2F;post&#x2F;2499755681&#x2F;the-best-debugging-story-ive-ever-heard" rel="nofollow">http:&#x2F;&#x2F;patrickthomson.tumblr.com&#x2F;post&#x2F;2499755681&#x2F;the-best-de...</a><p>Interplanetary debugging with the Mars Rover: <a href="https:&#x2F;&#x2F;www.eetimes.com&#x2F;the-trouble-with-rover-is-revealed&#x2F;#" rel="nofollow">https:&#x2F;&#x2F;www.eetimes.com&#x2F;the-trouble-with-rover-is-revealed&#x2F;#</a>
评论 #23783824 未加载
cbhl将近 5 年前
Previously:<p>2018: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=18675375" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=18675375</a><p>2015: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=9338708" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=9338708</a>
评论 #23775707 未加载
评论 #23775673 未加载
abeppu将近 5 年前
The ending of this makes it sound super clean. 3 ms * speed of light =&gt; ~560 miles. &quot;It all makes sense!&quot;<p>But ... isn&#x27;t the speed of light through fiber actually like 2&#x2F;3 of the speed of light in a vacuum? And that fiber isn&#x27;t going to be laid out in a straight line exactly to the destination. So I think really there must have been a fair bit of uncertainty around that ~3ms to abort a connection.
评论 #23776372 未加载
评论 #23776168 未加载
评论 #23777617 未加载
评论 #23776151 未加载
xook将近 5 年前
Truly this is a worldly gem. Thank you for submitting this. :)<p>It&#x27;s easy to forget that, even though transmissions still travel at near lightspeed, it still takes more than an instant to reach its destination, even digitally. I should keep this in mind, I think.
xwdv将近 5 年前
IMO even though this has been posted a bunch of times it’s important to repost these sort of campfire ghost stories so future developers can think more creatively about strange errors.
评论 #23776113 未加载
评论 #23776534 未加载
martyvis将近 5 年前
Only the other week we were doing some testing on a new HCI (hyper-converged infrastructure) I&#x27;m doing the network for. At the end of the test period, we were having some storage sync issues. Everything seemed to PING ok, except my colleague happened to notice large jumbo frames over 8000 bytes were getting dropped. I double checked that we hadn&#x27;t inadvertently changed network configuration. It was only by chance we had another test looking at transceiver signal levels that a customer engineer saw an alarm on RX level. It was then we remembered one test was to remove a module. I then noticed some error counts. We shutdown that particular link until we could visit the site. Sure enough, that fibre wasn&#x27;t quite clicked in anymore. There was enough of a bridge across the fibre air gap for shall packets, but just wide enough so large packets statically couldn&#x27;t be corrected enough to work.
评论 #23779426 未加载
rogierhofboer将近 5 年前
A while back, Andreas Zwinkau collected this kind of stories. Please post him if there is anything not yet listed.<p><a href="http:&#x2F;&#x2F;beza1e1.tuxen.de&#x2F;lore&#x2F;index.html" rel="nofollow">http:&#x2F;&#x2F;beza1e1.tuxen.de&#x2F;lore&#x2F;index.html</a>
noisy_boy将近 5 年前
If somebody is tracking apt&#x2F;yum downloads of packages, they might see a sudden spike for &quot;units&quot;. Just installed and it is a nifty little useful tool.
ryanmarsh将近 5 年前
Love this story.<p>I&#x27;m certain it will continue to be reposted to this website until ceases operations or the heat death of the universe whichever comes first.
PlasticTank将近 5 年前
Love this story, funnily enough it was the first story I ever read on HN on my first day of work as a junior developer.
bobblywobbles将近 5 年前
Can someone please explain to me the POP reference? I do not understand what this author means by that.<p>I also would like help understanding what $ units gives? The command looks to be &quot;units&quot;, but where do the numbers he entered in come from? I would appreciate this extra context.
评论 #23776610 未加载
评论 #23776633 未加载
评论 #23776614 未加载
评论 #23776619 未加载
评论 #23776642 未加载
noisy_boy将近 5 年前
Edit: Found it - its &#x2F;usr&#x2F;share&#x2F;units&#x2F;definitions.units (on Pop!_OS, so probably same on Ubuntu&#x2F;Debian).<p>The FAQ[0] mentions:<p>&gt; units on SunOS doesn&#x27;t know about &quot;millilightseconds.&quot;<p>&gt; Yes. So? I used to populate my units.dat file with tons of extra prefixes and units. And actually, I think I was using AIX to run units; I don&#x27;t know if it knew about millilightseconds. Take a look at the units.dat shipped with Linux these days. It definitely knows about millilightseconds.<p>I tried locate for units.dat but couldn&#x27;t find it. Anyone knows where is it? Not keen on running a system-wide find.<p>[0]: <a href="https:&#x2F;&#x2F;www.ibiblio.org&#x2F;harris&#x2F;500milemail-faq.html" rel="nofollow">https:&#x2F;&#x2F;www.ibiblio.org&#x2F;harris&#x2F;500milemail-faq.html</a>
评论 #23778275 未加载
评论 #23777000 未加载
davidkuhta将近 5 年前
Hat tip to @nfriedly from back in 2011:<p>FYI units on OS X doesn&#x27;t recognize millilightseconds, but you can do this:<p><pre><code> You have: 3 lightyears &#x2F; 365 &#x2F; 24 &#x2F; 60 &#x2F; 60 &#x2F; 1000 You want: miles * 559.21802 &#x2F; 0.0017882113</code></pre>
评论 #23776713 未加载
评论 #23775928 未加载
评论 #23776385 未加载
评论 #23776199 未加载
评论 #23776071 未加载
rcarmo将近 5 年前
This story makes me smile every time it comes up. It&#x27;s fascinating how many arbitrarily coded limits we keep breaking as we make our tech go faster without re-assessing the original assumptions :)
ip_addr将近 5 年前
Does anyone know of stories similar to this? I would like to read more.
评论 #23776925 未加载
评论 #23776532 未加载
评论 #23778954 未加载
评论 #23778513 未加载
评论 #23777207 未加载
MasterYoda将近 5 年前
Haha, this was interesting. I posted the same story with more or less the same title a couple of month ago [1], and no one saw it and no comments. This time it got almost 1000 points and a lots of comments. What I think is interesting is how the same thing can get so different traction. Wonder what factors it is that makes a thing get traction and not?<p>[1] <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=22164691" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=22164691</a>
henpa将近 5 年前
Since I&#x27;ve worked with linux email servers (sendmail, qmail, postfix, exim, etc) practically my whole profissional life (since around 1997, but used BBS since 91 - I&#x27;m 41 now), this story really amused me and got my attention! I <i>love</i> this kind of email debugging! LOL
walrus01将近 5 年前
In network engineering circles, this one is second only to the RFC for IP over avian carriers.
评论 #23778197 未加载
darekkay将近 5 年前
There is a popular GitHub repository for similar debugging stories: <a href="https:&#x2F;&#x2F;github.com&#x2F;danluu&#x2F;debugging-stories" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;danluu&#x2F;debugging-stories</a>
doebi将近 5 年前
Soundtrack for this story: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=tbNlMtqrYS0" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=tbNlMtqrYS0</a>
bawolff将近 5 年前
I&#x27;ve heard this story before, but i didn&#x27;t realize it was as recent as 2002. It feels like something from a much earlier bygone era, like the early 90s
评论 #23779537 未加载
评论 #23778458 未加载
failrate将近 5 年前
In a way, Hacker News acts like a Leitner deck.
kinduff将近 5 年前
This is one of my favorite bugs ever. Loved to read it again.
panosfilianos将近 5 年前
Debugging software is not unlike detective work.
dang将近 5 年前
Anybody know the original year?
optimaton将近 5 年前
This is wild!
netmonk将近 5 年前
So old! DUPE
simonebrunozzi将近 5 年前
I suggest to change &quot;mail&quot; to &quot;email&quot; in the title.
评论 #23776834 未加载
staycoolboy将近 5 年前
The one thing that throws a wrench in this story for me: Lattes.<p>Latte&#x27;s in 1994? In North Carolina? No way. Maybe on the West Coast, but I moved to Cali in 1989 and they were a rarity until the mid-late 90&#x27;s. There were only 425 starbucks in the US in 1994 (from their site). The &quot;fancy coffee&quot; craze was just a blip on the radar in the mid 90&#x27;s but gaining momentum.<p>;-)
评论 #23777337 未加载
评论 #23777187 未加载
评论 #23780286 未加载
评论 #23778344 未加载
评论 #23785093 未加载
coronadisaster将近 5 年前
just get it in the pipes
评论 #23776548 未加载