Tag Archives: monitor

darkstat: Making an exception again

Last time I saw darkstat, I had some misgivings about including it among console-based software. I am having those same thoughts again, now.

2014-08-30-6m47421-darkstat-01 2014-08-30-6m47421-darkstat-02

The problem is, I have entertained a lot of programs in the past that lean more toward the graphical, even if they do a lot of their core work at the command line. Looking back, I probably shouldn’t have fudged with things like lilypond or even mplayer or inkscape, because in retrospect, they just aren’t intended to run at the console alone.

darkstat falls into the same bracket. It’s a fantastic network traffic monitor, has a great array of graphs and charts, but is really intended as a browser-based interface to your network.

True, it can dump your traffic in hex, to the console.


But that’s a rather humdrum task for a program that’s doing lovely things within X. If there is another mode or another way to see darkstat in action, other than pointing Firefox at, I haven’t found it.

Please don’t get me wrong: darkstat does a fantastic job and if I had a proper server running (god forbid) and needed to connect remotely and see what’s happening to it — within the last few minutes, few hours, days or even this month — darkstat would be at the top of the list.

But if I ever repeat this little adventure and cruise through console software again, darkstat won’t be included. No more exceptions. ๐Ÿ˜

cv: The coreutils viewer

The other C-program I have today is a coreutils viewer, appearing as cv on Github. No, just for the record, I don’t dredge Github looking for new material. As it would happen, most titles these days are submitted from readers, and thrown into the hat.

This one was relayed to me by e-mail, and I realized later that I actually had it on my list as coreutils-viewer, so it’s possible I copied that name from elsewhere on the Internet. Regardless. …


Some other tools to amplify the output of core utilities — like pv or Advanced Copy — attempt to integrate themselves into the command, or pipe through it. cv, as you can see above, takes the sidelong approach by checking for running instances of dd, cp, mv, grep and a bunch of others, and showing their progress as a summary.

It’s an interesting solution to the long-standing issue of less-than-communicative programs, like cp. And goodness knows those have bothered me for quite some time.

cv has a few options to keep you busy; it has a monitor mode with -m, that will loop until processes finish, and another monitor mode with -M which will loop indefinitely, allowing you to keep it on screen as a kind of coreutils process monitor. I like that.

And there is a filter option with -c that lets you trim the display to only one particular process. Not much more than that, but simple is best. ๐Ÿ˜‰

I think I shall keep cv on board for a little while longer. I like the idea of having a continuous monitor of coreutils processes, even if I am quickly approaching a system made up completely of monitors, and nothing actually doing any work. ๐Ÿ™„

aajm: Conspiracy theories abound

There’s definitely something kooky going on with ls vimwiki/ | shuf -n1, because the next title picked is aajm. Between that, aview and bb — and maybe even asciijump from yesterday — there’s some sort of slant toward aalib-based software here.


Not that I’m complaining, it’s just to see three of those titles surface in the space of weeks suggests a lack of randomness. Then again, how can I be sure it’s really random … ? ๐Ÿ˜•

What you see in the gif there was deliberately slowed; I know when I first showed this some years ago, it was a bit of a drag for a Pentium machine. But on this 1.5Ghz Pentium M, releasing the -d flag sent it spinning into an ASCII blur, and making the results unviewable.

So if you try it on anything newer than a K6, you’re probably going to need to adjust the delay. Don’t say I didn’t warn you. ๐Ÿ™‚

What I failed to mention last time, is that aside from my usual suggestion for a practical use for aajm, the author has a short list of other ideas — including some very interesting ones, like tying aajm to system load values.

I tried it (look for the -jln flags) and it does make for a very unusual system monitor. My fear though is that on a slower machine, using this as a system monitor might increase the load, which would cause the animation to move faster, which would burden the machine more, which might increase the load, which would cause the animation to move faster, which would … ๐Ÿ™„

The only other thing I feel is worth mentioning, is that under X, you need to specify the curses driver, or it will latch on to a separate window with a different driver. Not that it will make a big difference, but it does cause a little inconvenience.

aajm is worth tinkering with, and a valid addition to a kind of screensaver thingy. If you take it in a different direction and use it as a system monitor or something else, it’s done double duty. Conspiracy theories aside, it’s good to have around. ๐Ÿ™‚

iftopcolor: Please do not feed the bloggers

If you think you can win my affection just by taking time-honored mainstays of the Linux software landscape and injecting them with color … well, you can.


neosrix left a note late last year about a colorized iftop — rather cleverly called iftopcolor — and I should have known better than to wait this long to check it out.

As best I can tell, it’s feature-identical to the standard iftop utility, but adds blue and green indicators for up/down transfer speeds.

Which in turn makes iftopcolor much more readable than its progenitor. Don’t get me wrong: The original is an excellent program. But just by virtue of its layout and the way the animation works, it’s particularly difficult to read — especially when it’s running wide open.

If you keep an eye on transfer stats, iftop alone is useful, but iftopcolor is a giant leap forward in ease-of-use. If you have a machine with multiple interfaces, I daresay it will be a lifesaver.

And I feel gratified to see a useful tool like iftop get a healthy swath of color. :mrgreen: Now ladies and gentlemen, you know the rules: Please do not feed the bloggers. ๐Ÿ˜‰

nmon: A monitor for everything

I could swear I have seen nmon before, and yet it’s not in the list of applications I commented on already, and it’s not on the old blog. I must have dreamed it. O_o


nmon picks up the reins of collectl, giving a variety of monitors that are all visible at your cue, on a long list of subsystems.

nmon works as a full-screen application though, with one-key commands that toggle boxed displays down the screen. If you find you don’t want a particular monitor or if you feel you need more space, you can add or subtract monitors as nmon runs.

Keypresses are easy-to-remember mnemonics, too. c for CPU, d for disks, n for network, k for kernel statistics … and so forth. And if you can’t remember one, press the h key for a help panel.

There are also “short” views of some monitors, that restrict themselves to active processes or active disk access, rather than listing a lot of deadbeats. It will save space, I promise.

nmon does just about everything right in my book, with the exception of no apparent provision for scrolling. It does a beautiful job providing a customizable panel of monitors, but you’ll end up turning off one or two so you can view one or the other.

Plus, the order of the panels seems fixed, so it’s not a matter of stacking them in the most convenient order for you; whether you start the CPU monitor or the network monitor first, they’ll always appear in the same order.

That’s a terribly small complaint though, and hardly worth voicing. nmon does a great job putting a lot of otherwise esoteric system information at your fingertips, with enough controls and options to keep even me busy.

Let me dig around here and … see if I have any … ah, here we are: For color, flexibility, usefulness and presentation, I hereby award one highly coveted K.Mandla gold star for nmon: โญ ๐Ÿ˜‰

ttyload: Choosy text-only geeks choose ttyload

Start up ttyload and you’ll see hints of things like tload, nload and even htop, in a way.


As far as system monitors go, ttyload really only relays the load averages over time, but the scrolling effect from right to left, and the addition of on-screen timestamps makes it a very enjoyable tool.

I can’t say if ttyload was taking a cue from nload or tload, but it’s clearly a step up. And if you’re confused by the colors, note that the asterisks change color where lines intersect, which is how the legend is useful.

Other than that, there’s not a lot that ttyload does. You might find it useful as something to run on the bottom-most desktop layer, or in a separate panel in your terminal multiplexer.

For my own part, I prefer to use it as one of many nifty screensavers. ๐Ÿ˜‰

A note or two for Arch users: At the time of this writing, the AUR package was pointing at the wrong source, and was out of date. That being said, the 0.5.3 version wouldn’t build for me, but the 0.5.2 would, but only if I built it by hand and without the PKGBUILD on that page.

Of course, the Debian versions were fine. ๐Ÿ˜€

slurm: Within one standard deviation

Abel sent me an e-mail a few weeks ago, reminding me that slurm belonged in the S section, and we discussed why slurm isn’t really a favorite for either of us. Here’s what it looks like, if you’ve never used it:


slurm does a lot of things right, but still falls short on a few other points. And when you stack it up against things like speedometer or ifstatus, it’s just riding in the middle of the pack.

There’s the good …

  • Color, and color themes, which is a nice touch.
  • Flexible, in terms of interface as well as refresh rate and graph modes.
  • Not tied to superuser, so you can start it without needing administrator access. And yes, for a normal desktop user like me, that is a feature.

… and the bad …

  • Sits in a fixed space, which is a huge disappointment if you’re using a tiled window manager or just like to resize windows from time to time. You’re either losing space that’s unused, or clipping the display.
  • No adjustable graph heights, so there’s a “bounce” effect if the display rate changes drastically.
  • No keystroke to redraw the screen. If you do decide to shrink the terminal, then resize again to a larger dimension, the graph labels don’t redraw and now there’s just data out there, flapping in the wind.

I think you get the point. It’s not terribly flawed, just missing out on some important points.

Abel said he (?) had seen slurm mentioned a few times on other sites and thought it was getting undue attention. I’d have to agree with that. It’s not a bad program, and it does a few things right. But overall it’s just sitting somewhere around one standard deviation.

P.S.: It’s worth mentioning that the link to the slurm “home page” is a github repo, I believe because Hendrik Scholz’s original version at www.wormulon.net/slurm disappeared long ago.

scanlogd: Nothing to see here

I’m a little behind the power curve today, after spending the day with some curious computer issues, which may or may not be related to hardware upgrades.

Also a curious issue: scanlogd. The curious part being, I am afraid I don’t have anything to show for it.

kmandla@j05sdg1 ~ $ sudo scanlogd
kmandla@j05sdg1 ~ $ 

If something beyond that is supposed to happen, I can’t see how it comes about.

I’ve been working with the Debian version, which installs and starts without issue, but doesn’t … really … seem to show anything for the (miniscule) effort involved.

I read through the man page, and maybe that lack of output is OK. It seems it shouldn’t really do anything unless someone attempts to scan ports on that machine — in which case it should just make a note of the attempt somewhere in /var/log.

Which is very very unlikely to happen, given that scanlogd is just wallowing around on my home network.

I will leave it to the more knowledgeable server managers to see if it is of any use. As for me, as a lowly desktop user, this doesn’t seem to have a function. ๐Ÿ˜ฆ

pppstatus: By virtue of past performances

I’ll mention pppstatus very quickly today, for three reasons. First, it wouldn’t run on my machine, most likely because I don’t have a ppp device for it to poll.

Second, I have mentioned it in previous posts, alongside such hits as ethstatus and others.

And third, I have used it in the past, so I know it’s functional. In fact, I still have one or two decent screenshots of it.


Mind you, that’s not a recent image.

pppstatus is not in Arch, but it is in Debian … sort of. It’s in Squeeze and Sid, but for some reason isn’t in Wheezy. Must be a packaging issue.

pppstatus is basically a network monitor for a ppp device. I know that’s a little behind on the times, and I take that into consideration when I draft little passages about software.

And I’m sure there are still one or two folks out there who use connections of that type. Maybe not 24/7, but I’m sure it happens.

And you know me: If I wasn’t sure it had at least worked in the past, would I list it here? Actually … don’t answer that. ๐Ÿ™„

oping: And even better, noping

In strictest terms, oping is satisfactory as a tool to replace the standard ping command found most everywhere else.

That’s all fine and dandy, but the real catch is noping, which is bundled along with it.


Just to be clear, oping is useful in its own right, and seems to do the job of ping quite well.

noping, on the other hand, was obviously designed for people like me.

A fullscreen ping monitor, with scrolling results, color highlighting and a small text box at the bottom showing cumulative statistics. Very neatly, very nicely done.

I don’t know that I’ve seen much in the way of an actual ping monitor, if you don’t count ekgping. I suppose netselect might count too, in its own way.

And while ping tools as a whole are probably taken for granted, it’s nice to see the principle carried out to a logical end, and with such panache.

I can find no faults here either. A coveted K.Mandla gold smilie for noping: ๐Ÿ˜€

P.S.: oping is in the repositories for Wheezy; in Arch it appears as liboping.