PaperCut Blog

Tech & DevInfrastructure

Two tales why health monitoring is scarily important

Two tales why health monitoring is scarily important

Take it from a support manager who’s seen some grisly stuff: when it comes to protecting your print infrastructure, implementing health monitoring should be your top priority.

Sceptical? These chilling tales should scare some sense into you.

*wolves howl, printers jam in the distance*

First up, what is health monitoring?

Health monitoring provides real-time updates on the status of your printing network so you can anticipate, manage, and eliminate issues (or avoid them altogether).

It’s also baked right into PaperCut MF and NG, so there’s no excuse to not have it up and running.

Especially when you know what frightening things can happen if you don’t…

#1: The tale of the paralyzed university

“Can ANYBODY tell me how many of my 620 devices are online now?!”
– Sysadmin, moments into meltdown

A major university, this customer prints upwards of 250,000 pages a day. That’s over 10,000 pages an hour, or 170 pages every minute.

So when printing was slowed dramatically for an entire afternoon, and 100,000 users ended up losing access altogether, the knock-on was nothing short of horrifying.

And yep, not one team member could answer the sysadmin’s seemingly simple question. Eep.

How the issue came about

Reading logs in real-time (a big no-no for print servers), using a pseudo Cloud solution with some questionable protocols…

The customer’s missteps were numerous, and took things from “hmm, that’s interesting” to “please send help” pretty quickly.

They also began monitoring with a third party tool (something we advised against) partway through the chaos, which only made things worse.

So the site server was losing the app server, no one knew why, and any realisations came too late. It was the ultimate hotpot of helplessness.

How health monitoring would’ve helped

With a PaperCut-endorsed health monitoring tool in place, every symptom would’ve been flagged well in advance.

Are servers talking well? Are jobs slowing? Are “online” printers actually frozen up? These are just some of the questions health monitoring would’ve asked, answered, and communicated – with time to spare.

It also would’ve prevented the desperate and ill-advised search for a bandaid fix, leading to hurtful third-party monitoring.

#2: The tale of the sleeping car sales site

Our next victim, a popular online destination for car sales, experienced a number of spotty connections between their central app server and numerous rural site servers.

The kicker? It all happened overnight when no one was watching. So when morning rolled around, so too did refreshed logs and zero evidence of any technical tomfoolery.

And you guessed it: everything came back with a vengeance.

How the issue came about

A few wobbly links caused the site servers to think the app server was coming and going, scaring them into an offline/online loop.

Then came the repeated attempts to sync and re-sync endlessly, which brought the system to a crawl, then to its knees.

Again, it was a case of limited knowledge leading to a haphazard and damaging remedy.

How health monitoring would’ve helped

Notifications, people! With health monitoring, the client would’ve been alerted to the irregularities as they popped up, not hours later as they woke up.

More specifically, PaperCut health monitoring could’ve pinpointed which crucial logs to grab before they disappeared, making the cleanup a lot easier.

How to start health monitoring

1. JUST DO IT (please)

Implementing this potentially network-saving tool is easy:

  1. Flex your sysadmin skills and get started yourself
  2. Talk to your ASC, and ask them how to get the ball rolling
  3. Ping us, and we’ll help you figure out the best way to begin monitoring.

2. Use a tool endorsed by PaperCut

Nagios, PRTG, and Zabbix all play really nicely with PaperCut, but if you have something else in mind, run it by us to see if it’s suitable. We’ve also got some handy starter guides for PRTG and Zabbix.

Lastly, our friends over at Selectec have done a great writeup on Grafana and Prometheus, a couple of other excellent tools we endorse.

3. Leave your logs alone

As the saying goes: let sleeping logs lie, or you might get bitten. So never touch them. Ever. Not even in your most rebellious sysadmin dreams.

Comments