16

I'm in IT and I've repeatedly heard stories of a server that was left stuck in a space that had a wall built in front of it, and was only discovered years later when the wall was torn down.

This 2001 article from The Register gives an example of such a claim:

According to a report by Techweb it was only then that those campus techies realised they couldn't find the server. Attempts to follow network cabling to find the missing box led to the discovery that maintenance workers had sealed the server behind a wall.

In one variety of the story, a fan starts to squeak and someone punches a hole in the wall to discover it.

I'm skeptical due to the fact that heat that would build up in a wall would fry the server.

Has a server (or servers) been walled in and discovered, still running, years later?

Oddthinking
  • 140,378
  • 46
  • 548
  • 638
Citizen
  • 264
  • 2
  • 9
  • 3
    This needs a [notable claim](http://meta.skeptics.stackexchange.com/questions/864/faq-must-all-questions-be-notable) in order to be on topic: "I've repeatedly heard stories" doesn't suffice. Note that "server" could mean many different things; some machines acting as servers might dissipate just a few watts. It'd be a fun back-of-the-envelope computation to look up the R-value of drywall and figure out how much wall area you would need to dissipate how much power with how much of a temperature rise. – Nate Eldredge Apr 13 '16 at 23:00
  • 1
    @NateEldredge Added a citation, would that suffice? – Citizen Apr 14 '16 at 00:08
  • Dead End: The [Techweb link](http://www.informationweek.com/wire/story/TWB20010409S0012) for the story is now broken, and not in the Wayback Machine. – Oddthinking Apr 14 '16 at 01:01
  • @Oddthinking I just clicked on it and took me here. http://www.theregister.co.uk/2001/04/12/missing_novell_server_discovered_after/ – Citizen Apr 14 '16 at 01:03
  • The date on the article at The Register is 4/12. That article appears to have come from [this article](http://www.informationweek.com/server-54-where-are-you/d/d-id/1010340?), dated 4/09. A number of links give dates of 404. (Oh right. That's not a date.) Note how the closer one gets to the source, the closer one gets to April 1. – David Hammen Apr 14 '16 at 01:25
  • @Citizen: I think we are at cross-purposes here. You provided a link to the Register, which mentions, but doesn't link, to a Techweb article. I hunted down that Techweb article as it may answer the question. However, when I found a link to it, it had been taken down and now gives 404 error. – Oddthinking Apr 14 '16 at 01:26
  • @Oddthinking thanks for improving my post. The link was weird. I wonder why it works/ed for me? Anyway, thanks for taking the time. – Citizen Apr 14 '16 at 01:48
  • Sorry to harp on it, but are you saying my link to TechWeb is working? – Oddthinking Apr 14 '16 at 04:04
  • 1
    @Oddthinking Yes, The original link worked for me and has always worked. I indicated that in my first reply to your first comment. It has always worked for me. – Citizen Apr 14 '16 at 04:30
  • I first heard this story about 1990, and the computer was an IBM S/38 at, I'm thinking, a bank. The computer (a server) continued to operate normally, and the error wasn't discovered until an IBM systems engineer came to do scheduled maintenance, about 6 months after the construction ended. (The story was repeated within IBM because the computer continued to function just fine without any attendant, a fairly impressive feat at that time.) And just because the room's entrance has been walled off does not mean that the ventilation has been cut off. – Daniel R Hicks Dec 22 '18 at 12:37
  • Actually, thinking some more about it, I'm remembering hearing the story in "the old Wells Building" used by IBM Rochester, and that likely puts it earlier than 1990. And the computer involved may have been a System/34. – Daniel R Hicks Dec 22 '18 at 13:16
  • A total disregard for space conservation makes me skeptical. Who builds like that? By being so careless with faux walls they decrease square footage by large amounts? –  Dec 22 '18 at 17:38
  • 1
    @fredsbend - Not at all unusual when old buildings are "remodeled" several times in a row. – Daniel R Hicks Dec 22 '18 at 20:45

1 Answers1

16

The Register article was most likely a feed from this very brief and unsourced InformationWeek article, dated two days before the article in The Register.

The story has the ring of urban legend to it. People who work behind the scenes have sometimes gruesome stories about lost equipment, lost machinery, and even lost people showing up years after the fact.

This particular story is almost certainly an urban legend since Judson Knott, director of Academic Computing Systems at the University of North Carolina, Chapel Hill disavowed the story a little over a month after it first appeared.

"I believe that this is what is commonly known as an urban legend. If it were a true story and Server 54 belonged to my organization, I would identify and fire the system administrator responsible for losing a server for four years. We run a first-class IT operation here in Chapel Hill and it is embarrassing to be associated with this kind of story."

pericles316
  • 22,676
  • 2
  • 84
  • 161
David Hammen
  • 14,500
  • 9
  • 57
  • 51
  • As can be seen by the quoted text, the source for The Register article was TechWeb, not InfoWeek. It is worth noting that the disavowal was not issued by UNC, but by Sun Microsystems, which explains the rather stilted plug. – Oddthinking Apr 14 '16 at 04:03
  • 2
    @Oddthinking -- TechWeb and InfoWeek are now one and the same. Trying going to www.techweb.com. It redirects to informationweek.com. – David Hammen Apr 14 '16 at 05:01
  • Note that, in the (somewhat different) account I heard, the equipment was not "lost", since it was still functioning just fine. It's just that no one knew where it was. – Daniel R Hicks Dec 22 '18 at 20:47
  • Please don't change "have sometimes gruesome stories" to "sometimes have gruesome stories". They stories exist of lost equipment, machinery, and even people exist. (Whether they're true or not is a different question.) The stories of missing people can be gruesome, but not so much the stories of the lost equipment. I'll accept "have sometimes-gruesome stories" (i.e., hyphenated) if the community thinks that that is an improvement, but "sometimes have gruesome stories" is a disimprovement. – David Hammen Dec 28 '18 at 17:32