BOFH: Baitin' switch

The old ones are always the best


Episode 13

"OK, let's just take a look then," our recent office addition says, clicking on the network management tool.

"Ah, there's your problem - your port is only set to 10 megabits per second, half duplex. If I just change that to auto you'll notice a short outage while your machine's network interface readjusts itself to 100 meg full duplex and then everything will be working tickety-boo for you.. No, No, don't mention it - and have a really great day."

. . .

"Did you know that all the data ports on that switch are hard configured to 10meg half duplex?" Graham the temp asks.

"Yeah," the PFY says, feigning interest. "It's probably something the previous admins did."

"But didn't you install those switches?"

"Yes - from the old building - but we didn't configure them."

"Really? You didn't take the changeover as an opportunity to revise the configs?"

"Not at the time, no," the PFY says, never one to seek out work unnecessarily - a stickler for the adage 'if it isn't broken, who gives a crap'.

I live my life by that.

"So these are configured the same way as they were..."

"In 1997, yes."

"You don't have a switch replacement policy?"

"Of course we do," I say. "If it's on fire it can go on next year’s replacement schedule!"

"But what happens meantime?”

“Oh, we’ve got a cupboard full of 10 meg PoE hubs that are still working perfectly!” the PFY says. “Which is why we don’t like to set the switch speed too high - they might notice.”

“You mean PoE switches?”

“No I mean PoE hubs. I tell you, they are as rare as rocking horse shit,” the PFY says. “But we’ve got stacks of them!”

“So there’s no QoS on them then?”

“No chance.”

“Doesn’t it affect the VOIP phones?”

“You’d think so wouldn’t you - but what we do is when we install a hub we slap it on a VLAN which has all the non-voice services deprioritised.”

“But doesn’t that make the servers and internet appear slow?”

“It sort of does, but with the desktops configured the way they are...”

>ring<

“Hello, Graham here, how can I help?”

“It’s still as bad as it was before. Would you hold please? >tappity< Yes, well it looks like your machine is requesting 10 meg half du.. can I get back to you? Thanks >click< You’ve set everyone’s desktop machines to 10 meg half duplex?!”

“Bingo,” the PFY says.

“And if he had his way,” I say, gesturing at the PFY, “We’d have upgraded to coax.”

”But you’ve got high spec copper to the desktop.”

“Uh-huh.”

“And multi-gigabit core and distribution network hardware.”

“Uh-huh.”

“But prehistoric access switches???”

“Indeed,” the PFY says. “Back in the day some beancounter configured a cable warranty period as an active switch hardware lifetime figure in the asset depreciation schedule. Those ancient 100 meg switches are all about halfway through their book-rated lives.”

“Now we’ve done what we could about this,” I add. “My assistant here managed to back a van over a couple of boxes of them during the building move - which was quite tricky as they were being carried by the movers at the time - but we have stacks of them to replace.”

“So in the meantime you just... manage expectations?” Graham asks, seeing a glimmer of the light.

“Manage user expectations and actively assist in switch failure,” the PFY counters.

“By assisting you mean?”

“We mean backing the cage nut screws on the switch off so far that the slightest touch (by a beancounter wondering why his port speed is 10meg half duplex) will cause the switch to fall from its lofty perch.”

“Onto the floor,” Graham nods.

“No,” the PFY says, “onto the switch below it - with the loose cage nut screws - which falls onto the switch below it - with the loose cage screws - and so on and so on until all the switches fall...”

“Onto the floor,” Graham nods.

“No,” the PFY says. “onto the beancounter. Because after all, they started it.”

“I see,” Graham says. “And so I’ve pretty much put a spanner in the works.”

“Not entirely,” the PFY says. “Just ring the beancounter back and tell him it’s all working perfectly.”

“But won’t he...”

“Think you’re lying? Of course he will. Then he’ll get the key that he thinks we don’t know he has and just check the comms room for himself...”

“Several crashes and a small cry later we’ll have an insurance claim ready to be processed,” the PFY finishes.

“I see. But isn’t it a bit... cruel?”

“You only say that because you’ve never seen a cupboard full of aging 12-port 100 meg access switches wheezing away,” I say, gesturing at the access comms room. “Just take a look for yourself and see what we mean.”

. . .

“I see,” Graham says, as the warm cloud of heated components envelops him. “These things are ancie-"

>CRASH< >CRASH< >CRASH< >CRASH< >CRASH< >CRASH< >aggg<”

“Two in one day!” I say, shaking my head.

“Two?” the PFY asks.

“Well we’re ringing that beancounter back, aren’t we?”


Other stories you might like

  • AI tool finds hundreds of genes related to human motor neuron disease

    Breakthrough could lead to development of drugs to target illness

    A machine-learning algorithm has helped scientists find 690 human genes associated with a higher risk of developing motor neuron disease, according to research published in Cell this week.

    Neuronal cells in the central nervous system and brain break down and die in people with motor neuron disease, like amyotrophic lateral sclerosis (ALS) more commonly known as Lou Gehrig's disease, named after the baseball player who developed it. They lose control over their bodies, and as the disease progresses patients become completely paralyzed. There is currently no verified cure for ALS.

    Motor neuron disease typically affects people in old age and its causes are unknown. Johnathan Cooper-Knock, a clinical lecturer at the University of Sheffield in England and leader of Project MinE, an ambitious effort to perform whole genome sequencing of ALS, believes that understanding how genes affect cellular function could help scientists develop new drugs to treat the disease.

    Continue reading
  • Need to prioritize security bug patches? Don't forget to scan Twitter as well as use CVSS scores

    Exploit, vulnerability discussion online can offer useful signals

    Organizations looking to minimize exposure to exploitable software should scan Twitter for mentions of security bugs as well as use the Common Vulnerability Scoring System or CVSS, Kenna Security argues.

    Better still is prioritizing the repair of vulnerabilities for which exploit code is available, if that information is known.

    CVSS is a framework for rating the severity of software vulnerabilities (identified using CVE, or Common Vulnerability Enumeration, numbers), on a scale from 1 (least severe) to 10 (most severe). It's overseen by First.org, a US-based, non-profit computer security organization.

    Continue reading
  • Sniff those Ukrainian emails a little more carefully, advises Uncle Sam in wake of Belarusian digital vandalism

    NotPetya started over there, don't forget

    US companies should be on the lookout for security nasties from Ukrainian partners following the digital graffiti and malware attack launched against Ukraine by Belarus, the CISA has warned.

    In a statement issued on Tuesday, the Cybersecurity and Infrastructure Security Agency said it "strongly urges leaders and network defenders to be on alert for malicious cyber activity," having issued a checklist [PDF] of recommended actions to take.

    "If working with Ukrainian organizations, take extra care to monitor, inspect, and isolate traffic from those organizations; closely review access controls for that traffic," added CISA, which also advised reviewing backups and disaster recovery drills.

    Continue reading

Biting the hand that feeds IT © 1998–2022