Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

GUI-Based Asset-Tracking Tools For a Datacenter?

timothy posted more than 4 years ago | from the your-gigantic-whiteboard dept.

Hardware 113

toruonu writes "How do you keep track of what's in your datacenter, where it is, what it's connected to and what is it doing right now? I mean I have built a datacenter from scratch over the years and I have machines from Sun, IBM, HP, Supermicro. I have machines that are simple workernodes and machines that are heavy grade storage consolidation machines. Then there are tens of switches, some for interconnect, some for management and don't get me started on the UPSs etc. So how does one keep any kind of decent track of such a system as the current form of twiki pages with various tables just doesn't cut it anymore and I'm looking for a freeware solution that could actually show me a visual representation of the various nodes in the racks, their connections and dependencies. Just to give a simple example, if I'm going to disconnect UPS #3 right now and swap switch #5, which machines should I even consider taking offline?" (The best-looking such system I've seen was being used at OSCON at a display booth for the Open Source Lab, and I think it was home-grown. Anyone who can shed light on that system?)

Sorry! There are no comments related to the filter you selected.

A spreadsheet (3, Funny)

Dice (109560) | more than 4 years ago | (#31945274)

Rows for hosts, columns for PDU, switch and console ports. Additional rows for asset tag information, unit manufacturer, model number, serial number. Last row for notes on the system, e.g. any historical hardware issues that may be relevant.

Re:A spreadsheet (4, Insightful)

MisterFuRR (311169) | more than 4 years ago | (#31945298)

Rows for hosts, columns for PDU, switch and console ports. Additional rows for asset tag information, unit manufacturer, model number, serial number. Last row for notes on the system, e.g. any historical hardware issues that may be relevant.

Because *that* scales well...

Re:A spreadsheet (1)

houstonbofh (602064) | more than 4 years ago | (#31945528)

Rows for hosts, columns for PDU, switch and console ports. Additional rows for asset tag information, unit manufacturer, model number, serial number. Last row for notes on the system, e.g. any historical hardware issues that may be relevant.

Because *that* scales well...

Modded troll by someone who has never had to work with a messed up and out of date spreadsheet designed for a 10 node system that has now grown to well over 100 nodes.

Re:A spreadsheet (0)

Anonymous Coward | more than 4 years ago | (#31947548)

Warning! By submitting this post, all moderation points in this discussion will be discarded.

Re:A spreadsheet (3, Insightful)

MBGMorden (803437) | more than 4 years ago | (#31945308)

And gets limited to a single user, or passing around the file and worrying about who has the latest copy, etc.

Spreadsheets make piss-poor databases. You could code up a simple app + database system to do that in less than 2 days.

Re:A spreadsheet (1)

Colin Smith (2679) | more than 4 years ago | (#31945950)

You could code up a simple app + database system to do that in less than 2 days.

Go on then. Show us how good you are.


Create, Read, Update and Delete (1)

itomato (91092) | more than 4 years ago | (#31946342)

CRUD. []

1. Download, extract.
2. Twiddle bits
3. Create table space
4. ??
5. Profit

Total time depends on your skills, but really, most of the real work has been done.

Re:A spreadsheet (0)

Anonymous Coward | more than 4 years ago | (#31946546)

Give me full requirements and $100/hr, and I'll give you a Delphi or .NET app for windows. Alternatively I can do a web version.
This problem just begs to be developed in-house instead of trying to shoehorn a 3rd party solution into it.

Re:A spreadsheet (1)

MBGMorden (803437) | more than 4 years ago | (#31946874)

If you're willing to pay me for my time, sure I'll develop it. Unless you honestly think I'm going to donate 12-16 hours of development time to some random guy on Slashdot for the heck of it?

The reality is it takes all of 20 minutes to setup a database that keeps track of as much info as the GP's spreadsheet idea does. If you can't wrap an interface around that in 2 days then I just pity you.

A full fledged application? Yeah, it'll take a bit longer. But something small in place of a half-assed spreadsheet? It's pretty simple.

Re:A spreadsheet (0)

Anonymous Coward | more than 4 years ago | (#31948936)

Not everyone on Slashdot is a coder or gives 2 shits about databases. Probably why the submitter is asking for a solution.

I have done many jobs in the past - even some light perl scripting where needed - and I honestly would rather have someone else do this kind of project than me. It's just not up my alley - and the submitter probably doesn't have the skillset either.

Re:A spreadsheet (1)

Dice (109560) | more than 4 years ago | (#31947504)

It works fine for a single digit thousands of hosts and three sysadmins, which is what I use it for. Concurrent write access isn't really an issue since updates are fairly infrequent and it's obvious who should have the write lock on the spreadsheet, that being the guy in the datacenter who's installing or removing equipment.

An app would be nice, but it wouldn't provide any real benefit over the spreadsheet model until it was extended to touch on other areas of datacenter operations. Something like RedHat's Satellite (or the Open Source Spacewalk) would be an example of an application which provides extra functionality such as server provisioning, configuration management, and centralized updates.

Re:A spreadsheet (0)

Anonymous Coward | more than 4 years ago | (#31945344)

Or you can just diagram the racks on a spreadsheet, color code them, and note on each one what the device is named, PDU its on, serial number, weight, whatever you want. Each row represents 1U. And you can type the name of the device right on in the cell that represents the device and put the other info in a note on the cell.

This is how the world's largest retailer keeps track of the racks in their two data centers. Seriously.

Re:A spreadsheet (1)

Monkeedude1212 (1560403) | more than 4 years ago | (#31946846)

The best part about this is that you can translate Serial numbers and service tags into quantifiable units and create some pretty good line graphs and pie charts.


martiniturbide (1203660) | more than 4 years ago | (#31945372)

Not free or open source. I think IBM Tivoli Application Dependency Discovery Manager or TADDM (what a name) can help you discover the machines and dependencis you had there. [] But i don't know if it will help ou with your example of the UPS. And it will cost you the left nut and half of the other.

Advanced mind maping software? (4, Informative)

sznupi (719324) | more than 4 years ago | (#31945380)

Might be close enough, I guess...since that's essentially what you want to do, map some aspects of the "electronic mind" that's under your care.

Wikipedia seems to hava a list perfectly adequate as a starting point []

Re:Advanced mind maping software? (1)

andr00oo (915001) | more than 4 years ago | (#31947320)

Mindmap? Its a nice idea, but Mindmaps are really a hierarchy and datacentres aren't really. Lets say you started with your Datacentre in the middle and you had a bunch of branches for the racks and a bunch of branches for the switches. The servers in the racks need to connect to the switches as well. The model doesn't fit that well. Andr00oo

Re:Advanced mind maping software? (1)

sznupi (719324) | more than 4 years ago | (#31947968)

You mean datacenters have several interconnected hierarchies? Mind maps always could model that - apart from the "iconic" tree/star you could set particular shapes or colors for each box, you have another two hierarchies right there. Two trees can sensibly branch from opposing sides of the screen and connect to common boxes. Or the map can simply have a bit of depth, with you choosing which group of connections to show, and which to make more "translucent".

I do not know if available free software can do it - for my needs, if I feel like a mind map of sorts (or more preciselly "doodle" ;p ) is in order, large piece of paper works fine...which gives almost all of the above, only the last one has a different form (and I don't need extensive modification, reproduction, etc.). But I would be somewhat surpised if it can't, perhaps worth checking, that's all.

Re:Advanced mind maping software? (2, Interesting)

andr00oo (915001) | more than 4 years ago | (#31948598)

You might be right, but my concept of mind maps (and I shamelessly copied the wording from Wikipedia) is that a mind map "is based on radial hierarchies and tree structures denoting relationships with a central governing concept". You might be thinking of a Concept Map: []

PS. (1)

sznupi (719324) | more than 4 years ago | (#31948108)

And how could I forget to mention that, at least to some degree (where it makes sense), distribution of boxes on a mind map can follow quite closely physical arrangement of you datacenter.

Racktables (4, Informative)

MBGMorden (803437) | more than 4 years ago | (#31945384)

Not sure if it meets his needs exactly, but I've used [] in the past and it's worked well for keeping track of a small-ish datacenter (about 400 sqft with 7 full size racks and a couple dozen servers).

Bright Computing Cluster Manager (2, Informative)

middlemen (765373) | more than 4 years ago | (#31945418)

A couple of weeks ago there was this company called Bright Computing, that was pitching their software called Cluster Manager, which looked very cool as they had an excellent interface and could keep track of everything from machines, to routers, switches, power usage etc. Something to look into. []

I use Nagios (4, Interesting)

ircmaxell (1117387) | more than 4 years ago | (#31945424)

I use Nagios for that kind of thing. Don't get me wrong, it isn't "perfect" at it, but it does a decent job once setup. If you use parenting in the configuration files, you can click on "network map", and immediately see each hosts' dependencies. And IIRC there are comment fields that you can write misc information (such as rack position, switch position, model, make, etc)... And it's free...

Nagios? Never heard of OpenNMS, I take it (4, Informative)

spun (1352) | more than 4 years ago | (#31945540)

It does automated network scanning for assets, asset tracking, alerts and notifications, SNMP data collection, reporting, and yes, it has a .svg based network mapper that you can customize with your own graphics. It comes with MIBS for hundreds of devices, but you can easily import your own MIBS for unsupported devices. It's open source, of course. Nagios is just a bunch of disassembled parts. You have to wire it together for each device. Adding new devices is a pain: you have to install the Nagios monitors on each new device. Nagios does not speak SNMP! OpenNMS does speak SNMP, and it will autoscan networks for devices, and devices for capabilities. Adding thousands of devices at a time is a snap. Plus, OpenNMS uses a modular architecture that scales well. We use it on a network consisting of over 2,000 clients at 50 offices, 30 IBM Blade servers hosting hundreds of VMWare virtual hosts, and innumerable network devices, printers, etc.

Re:Nagios? Never heard of OpenNMS, I take it (0, Redundant)

quantumplacet (1195335) | more than 4 years ago | (#31945640)

I'll second this. Once upon a time I spent weeks trying to get a Nagios setup working the way I wanted until I eventually abandoned it. Recently discovered OpenNMS, and while it's far from perfect it is a huge step up from Nagios. One thing I will warn though, it's a bit of a beast. Full SNMP collection on a few hundred devices requires some decent processing power and a boatload of I/O.

Re:Nagios? Never heard of OpenNMS, I take it (1)

ircmaxell (1117387) | more than 4 years ago | (#31945748)

Wow, does that ever read like a sales pitch...

I've heard horror stories setting up Nagios, but to be honest, my setup was a breeze. I did the smart thing, and wrote templates for each kind of device I'd be using, and then setup Puppet to push new Linux servers into Nagios (and set them up properly) automatically. For my windows computers, AD automatically sets everything up for me (with the exception of adding the server to Nagios, which is done via a shell script). Granted, I have a small system now (only about 50 devices and 500 services), but I can provision new hosts/services in at most a few minutes as needed (Most of them are less than that, since it's automated). And for the record, Nagios does do SNMP. In fact, it's monitoring my phone system right now via nothing but autodetected SNMP. I'm not saying OpenNMS is bad at all. I'm just saying my experience with Nagios is nothing compared to what I've read others experience.

Re:Nagios? Never heard of OpenNMS, I take it (1)

mu51c10rd (187182) | more than 4 years ago | (#31946226)

I switched from Nagios to OpenNMS as well. With OpenNMS, I ran the yum install, put in my IP ranges and community strings, and that was it. It discovered, monitored, and graphed Cisco, Windows, and Linux boxes common services after 2 minutes of configuration (literally). No horror story about Nagios, OpenNMS just scaled better and required no shell scripts or setup really.

Re:Nagios? Never heard of OpenNMS, I take it (1)

dissy (172727) | more than 4 years ago | (#31947028)

And for the record, Nagios does do SNMP

What the parent was referring to is the fact Nagios itself internally does NOT do SNMP, nor see any SNMP data what so ever outside of a preprocessed result code.
Nagios uses external programs (Mosty perl, but anything that can spit out exit codes and stdout text will work) which are what uses SNMP.

All of my SNMP monitored devices are polled by perl scripts that came with Nagios. All Nagios sees from them is a return code with one of three states, not the SNMP data.

While most people tell you to setup MRTG or Cacti for pretty SNMP graphs, Nagios can not do that same thing by itself, only due to not having SNMP support internally.

And in a way this makes sense, as it is a status monitor. It was designed to only provide a simple summary that is boiled down to 'up','warning', and 'critical'.

OpenNMS does poll devices itself and store the actual results, which are parsed and Those results also stored.

While it is a minor thing to nitpick on, it does affect how you use each tool, and even which tools in your suite.

For the record, I run Nagios and MRTG at work, however I've been wanting to look at an OpenNMS system as soon as I can re-purpose a machine for it that is capable.

Re:Nagios? Never heard of OpenNMS, I take it (1)

PhrstBrn (751463) | more than 4 years ago | (#31949252)

Nagios + Check_Mk + PNP + NagVis has worked well for us as a platform.
  • Nagios for notifications
  • Check_MK is a monitoring plugin for Nagios. It does a bit better job than something like NRPE, it creates only one active check per server, and all the other services become passive checks. Increases performance, and does a lot of auto configuration for you.
  • PNP4Nagios for RRD graphs/trending. It integrates well with nagios for performance data.

  • NagVis is what the OP might be looking for. It allows you to upload pictures, and put some XY coordinates on it for each server. So you can upload a photograph of your server rack, and put little check boxes for each server transposed on the picture (healthy/not healthy) with labels and everything too.

Pay for it. (4, Funny)

NeutronCowboy (896098) | more than 4 years ago | (#31945426)

This problem has several paid solutions, all of which work fairly well, and make maintaining a data center the job of one person, instead of 20 people looking at a spreadsheet and log files. I haven't found an open-source package that is nearly as competent as the integrated solutions offered by HP, IBM and others. Warning: sticker-shock is included. Bonus: PHBs like looking at pretty pictures, and all the commercial tracking software produces pretty pictures. Your PHB looks like a super-hero to his PHBs, and you become an invaluable asset to your PHB for making him look good.

Re:Pay for it. (1)

CyprusBlue113 (1294000) | more than 4 years ago | (#31946566)

I finally just wrote my own system for tracking all the connections in the datacenter at my last job, unfotunately that former boss did not wish to OSS it, nor sell the product to outsiders. I wish I had it at my current job.

That being said, it is actually suprisingly simple to write your own system, the only difficult part at all is creating the different templates for all the different objects you use, other than that, its a very simple database of objects, and connections.

Human input (2, Interesting)

spaceman375 (780812) | more than 4 years ago | (#31945440)

I don't know your solution, but I can tell you it will involve automatic network mapping and polling of services. You need to find a solution that relies on human input as little as possible. Otherwise documentation gets out of date, no longer trustworthy, leading to lack of incentive to update it, ... With a big budget, I'd go for RFID on everything, with local readers doing triangulation. That's the only way to really track physical objects. Add that to the maps that network discovery makes and you've got what you need.

OCS inventory and GLPI (1)

Colin Smith (2679) | more than 4 years ago | (#31946180)

GLPI is the asset management DB and OCS Inventory, inventories your inventory.

There are 2 sides to asset management.

What should be.

killer app needed. (1)

demigod (20497) | more than 4 years ago | (#31945488)

I've been trying to find something for over a year to do just that.

Nothing meets all my needs yet.

Right now I'm using racktables [] ,Open-AudIT [] and some stuff I wrote to fill in the gaps.

The real problem of course is getting all the techs to actually update stuff when they move it.

Re:killer app needed. (1)

ircmaxell (1117387) | more than 4 years ago | (#31945868)

I was very intrigued by Open-AudIT. So much so, that I downloaded it. I looked at the source. And then I promptly deleted it. The few files I looked into could be used in a book for how NOT to program PHP. Too bad too. It looked like it had potential...

Re:killer app needed. (1)

Lennie (16154) | more than 4 years ago | (#31947236)

Yeah shocking.

I do think if someone took the time to clean that up it would be a big improvement.

The advantage is, almost anyone could do it, with some directions, it just takes time.

Rackview (2, Informative)

Anonymous Coward | more than 4 years ago | (#31945490)

The system used to visualize the OSUOSL lab is called Rackview.

Rackmonkey (3, Interesting)

tigerknight (305542) | more than 4 years ago | (#31945494)

This is what I use to keep track of the racks in my work's facility. It allows you to put in a whole lot more than just simple rack location. It's a wonderful tool.

Re:Rackmonkey (0)

Anonymous Coward | more than 4 years ago | (#31945874)

Another vote for RackMonkey.

We keep track of 400+ servers with it. creates decent reports for the executroids. Decent GUI, simple enough to fob off on the Tier 1 guys to populate/maintain the data.

Asset Tracking (1)

RobertM1968 (951074) | more than 4 years ago | (#31945510)

Many of the older IBMs (and probably the newer ones?) came with (by default or as an option) asset tracking capabilities - I know all of my smaller ones had them by default (including the RFID unit) - though my bigger ones do not come with them (I guess because you'd need a few people or a forklift to move them). You may want to look into that for your IBM servers. It included (among other features) RFID tracking to know when/if a machine was moved, moving, being removed, etc.


somethingtoremember (1530149) | more than 4 years ago | (#31945512)

Pen and paper has been proven to be have the highest ROI, ease of use, enterprise scalability, and most importantly, security, of any inventory solution. It is also an extremely dynamic application, allowing an unprecedented degree of freedom in the form of sketches, diagrams, language flexibility, etc. Also, since manufacturers conform to the process of using ink and paper to print ID tags and model numbers on their equipment, this application will seamlessly integrate. It is extensible with self-adhesive stickers. This Off-The-Shelf software is available at any corner store. Its product lifetime is in the span of hundreds of years. There is no competition that can match it.


Em Emalb (452530) | more than 4 years ago | (#31946112)

NOOB! Everyone knows PENCIL and paper is so much more efficient than pen and paper.

Sheesh. Pen and Paper. LAWLS.


Anonymous Coward | more than 4 years ago | (#31946480)

Easier to wipe your ass with, as well.

We wrote our own... (2, Interesting)

Fallen Kell (165468) | more than 4 years ago | (#31945522)

Seriously, we looked all over for something, but nothing fit. Grant it, we did this back in 2000-2001 timeframe. We setup a mysql database, and wrote up a website with php which was the interface. We scanned in floor plans of the buildings and setup an image clickmap for all the cubicles/locations on the floor plans and had them all point to a unique location_id. The location_id's were one of the keys in the datbase to track the hardware.

So you could litterally navigate to a particular building/floor, and then click on the cube/location and it will then show a list of all the equipment in that area. You can add new hardware to that location or click on a piece of hardware and view its information (CPU type/speed, hostname, IP address, MAC address, RAM, etc..) and if it moved to a new location, you click on a "move" button, and it opens up the list of buildings/floor plans and you simply navigate and click on the place where it moved to, and then update the database record.

We add some more sophisticated features like barcodes to cubicles and to the systems themselves and you can go around with a barcode scanner hooked up to a laptop and simply scan the barcode on a cubicle, and then scan all the barcodes of equipment in the cubicle and it would automatically associate all that equipment with that particular location (and if it was a new piece of equipment, it would open the form to add the equipment into the database).

Re:We wrote our own... (1)

Da Web Guru (215458) | more than 4 years ago | (#31945928)

Another vote for the home-brew option.

I built the web interface that we are using at work to track server information. Due to the fact that it was built completely from scratch (using Perl, that language that so many people claim is dying a slow painful death), we were able to customize it to integrate into a number of other systems such as equipment locations, IP address management, Nagios service monitoring (currently monitoring over 13000 services), server bandwidth usage tracking, internal issue tracking, etc. We are using it to track nearly 10000 servers across 6 facilities in 4 different cities.

Re:We wrote our own... (1)

Lennie (16154) | more than 4 years ago | (#31947218)

The problem with home-brew is, we are all inventing the same wheels, over and over again. And what for ?

Re:We wrote our own... (1)

Da Web Guru (215458) | more than 4 years ago | (#31948254)

The problem with home-brew is, we are all inventing the same wheels, over and over again. And what for ?

Because we were told to "make it work". We weren't given a budget so we couldn't go out and buy something off the shelf. And we didn't see an open source tool that could combine all of those functions into a single easy-to-use interface. Regardless of what existing tool we started with, we would have to hack on several other systems just to make it work. Rather than reverse engineering 4-5 different software tools and hacking them all together into one interface, then figuring out how to connect it to our billing and customer support systems (also built in-house by other teams), it made sense to just code it up from scratch. This way we could customize it to support the requirements of support, sales, and management staff, as well as support however many facilities, locations, and staff members (with varying levels of access) as necessary.

Next billion dollar idea... (0)

Firemouth (1360899) | more than 4 years ago | (#31945542)

What we'll do is make a device that's about 6" long, maybe 1/4" in diameter. At one end we'll have some kind of a mechanism to enable us to create markings on a surface. Next, we'll come up with thin sheets of material that we can mark up. Just spitballing, maybe we'll call it... pencil and........ paper. Yes, pencil and paper. That was the easy part... the hard part is figuring out where to put all these pieces of paper once they've been marked...

Altiris (0)

Anonymous Coward | more than 4 years ago | (#31945556)

Symantec bought a company called Altiris a couple years ago and Altiris had a pretty good asset/CMDB solution where you could define relationships like this and run reports to see what impact would be seen if you were doing maintenance on a particular piece.

More details at

Numara Track-It! (1)

zero_out (1705074) | more than 4 years ago | (#31945574)

I used Numara Track-It! [] for a few years while doing desktop/server admin. There are some nice auditing tools in there which grab all the hardware and software info automatically. It is highly configurable too. I'm sure there are better solutions out there, but that is what I used.

The Dude (0)

Anonymous Coward | more than 4 years ago | (#31945632)

Though it's not an adequate tool for a wide array of CMDB purposes, if all you need is something to keep track of servers and network layouts, I'd recommend "The Dude"...

MAM (1)

stephen.schaubach (156426) | more than 4 years ago | (#31945666)

I work for a descent size bank and we tried Mercury Application Mapping (bought out my HP). It uses nmap which security freaked out over. Once it goes out and finds everything, it draws lines to and from each component. It does this by looking at certain config files in each app (web, app, db, etc.) which was cool but permissions were a hassle. In the long run it took quite a lot of effort to get anything out of the package and we eventually scrapped it completely. It costs big bucks or at least it did for us and not it sits somewhere not being used...that never happens of course.

Whiteboard (0)

Anonymous Coward | more than 4 years ago | (#31945736)

Seriously, last datacenter I worked, we had 2 whiteboards along one wall with the whole thing charted in multicolor dry-erase... it worked very well for our day to day needs. Of course there were visio diagrams that got updated regularly in case of accidental erasure of the whiteboard, and everything was tagged with labels indicating what it was and what it was connected to.

Gotta love the captcha: voltage

1000 sq. ft? (0)

Anonymous Coward | more than 4 years ago | (#31946648)

Seriously, how many machines can you track on a whiteboard? Go-ups, or come-downs, *maybe*..

Mish-mash of green, black, and red Expo by three different hands?

What happens when Curly and Larry are screwing around and randomly remove data, only to be discovered by Moe, come action time?

Hilarity ensues!!

GLPI (1)

frodo from middle ea (602941) | more than 4 years ago | (#31945738)

GLPI [] , works well for us, and it's GPL, what more can you ask ?

you f41l it (-1, Troll)

Anonymous Coward | more than 4 years ago | (#31945754)

there are Java IRC client Where it belongs, From t4e FrreBSD were compounded come Here but now Product, BSD's The curtains flew During this file the fruitless

Wiki + Dia + MySQL (1)

raddan (519638) | more than 4 years ago | (#31945810)

We use a wiki (XWiki) with various tables to match machines to replacement FRUs, diagrams done in Dia (or OmniGraffle), and a homegrown inventory DB in MySQL to track parts & other equipment availability. There's no magic solution. You just need to pick a tool and make your staff stick to it. Here, what works well is distributing responsbility for various subsystems to different admins, so that no one admin is overwhelmed with administrative work. You need an FC controller board, you go talk to the person who 'owns' the SAN subsystem parts-- if they're out, no biggie, because it's all in the wiki. Just update the quantity in the DB and let the person know in case they need to order more stock.

What the OSUOSL uses and beyond (4, Informative)

ramereth (752738) | more than 4 years ago | (#31945828)

The project you saw at OSCON was called RAIV (Rack And Inventory Viewer). Unfortunately it ran into a dead-end and is currently vapor-ware. Currently we're using an internal CakePHP webapp for basic inventory and customer tracking, but its very buggy and lacks many features.

We are in the midst of working on a completely new project that will cover many of the problems mentioned in this article and beyond. Think of it like an open source datacenter management webapp and backend. Its still in the planning stages, but the intent is to have a plugin based system where you can use the inventory plugin, DNS/DHCP plugin (to replace maintain), virtualization management (deployment and console access), etc all in one interface. The idea is to create an admin interface and a customer interface so that they can access and see information about their environment. We're far from having a demoed project but we hope to have something soon.

Re:What the OSUOSL uses and beyond (1)

timothy (36799) | more than 4 years ago | (#31947882)

Interesting! Thanks for the reply.

What I liked, being a visual person, was the way that things on-screen were easily matched w/ their physical counterparts. I'm in charge (for good reason) of *nobody's* data center, but I was taken by that system's use of graphics. I would imagine not being impressed by abstract representations and cute names if a certain machine is on the fritz at 3:31 a.m.


Re:What the OSUOSL uses and beyond (2, Informative)

size1one (630807) | more than 4 years ago | (#31948158)

RAIV wasn't graphical at all, so you must be thinking of the virtual server room tour we put together 2 years ago.

The code for that is here: []

however theres no documentation, and likely doesn't work with the latest version of openlaszlo. It wasn't tied into our inventory system at all. I manually merged several different sources into a single xml file. We've since moved away from laszlo in favor of html+css+javascript+svg. Eventually we might rewrite this, but have no concrete plans right now.

Re:What the OSUOSL uses and beyond (1)

charlesnw (843045) | more than 4 years ago | (#31951278)

Have you ever heard of OpenQRM? It sounds almost exactly like what you mentioned. :)

Not free, but still awesome (0)

Anonymous Coward | more than 4 years ago | (#31945838)

I realize it's not free, but I've found that to get a proper scope/view of what's going on in a large datacenter that implementing a full ITIL solution tends to be the best bet. Here where I work we use Servicedesk Enterprise by ManageEngine/Zoho. If it absolutely must be FOSS, I can't think of a better way to handle it than using Nagios with some of the varied plugins to do auto-tracking/creation of nodes, and something to allow you to do click-based management (click a node, get an SSH window?). (0)

Anonymous Coward | more than 4 years ago | (#31945902)

Rackmonkey (0)

Anonymous Coward | more than 4 years ago | (#31945944)

Rackmonkey is a nice open source package that does some of what you ask for...

Re:Rackmonkey (1)

modir (66559) | more than 4 years ago | (#31946470)

I would have given you some points if I could. RackMonkey is really a good solution. Although I don't know if it scales to the point where questioner needs it.

We would have taken this software but there was only one problem. We have several devices (like e.g. firewalls) which are only "1/2 U". Meaning we have two devices next to each other. And this could not be represented at the time when we evaluated the software.

Re:Rackmonkey (1)

tigerknight (305542) | more than 4 years ago | (#31948524)

There are a few non-standard racks in my facility too, in that case we added notes to describe what is what. Irritating that it marked me as 'anonymous coward'... I'm logged in!

RFID IT Asset Tracking Solution (tested great!) (1)

siteforce (93054) | more than 4 years ago | (#31945952)

I looked at this solution -> We will probably get it:

IT Asset Management (ITAM) is becoming an increasingly important, yet difficult responsibility as the number of IT assets increases exponentially. Trying to keep track of mobile assets adds to the complexity. Surprisingly many organizations still use time-consuming manual pencil and paper, or barcode processes to track IT assets. Not only are these processes extremely labor intensive, they are often prone to human error. RFID technology can be leveraged not only to automate the physical inventory process, but also to provide real-time tracking and security of IT equipment.

Data center staff are typically not employed to maintain IT asset inventories, yet they are ultimately responsible for accounting for hardware and the data they contain.

Specialized RFID tags enable small devices such as blade servers to be read without pulling them out the racks. Data center assets can be tracked by walking around with handheld RFID readers and by placing fixed readers at entry/exit points and other strategic locations to track the movement of equipment.

Some IT vendors such as Dell, HP, IBM and others are beginning to offer services to tag equipment with RFID tags prior to shipping to the customer. This not only eliminates the need for customers to tag equipment, it can help automate the internal receiving, provisioning, and deployment processes.

Visio + Custom DB (1)

dtecmeister (1171031) | more than 4 years ago | (#31946158)

We took measurements of everything and put it in visio to scale. An overhead view and at least a front view of each rack works well. That's for the visual part. For the connections, use a database so it can be easily queried and traced. Full dependency reports may take some SQL wizardry, but can be accomplished if the DB is carefully designed.

This great idea keeps coming back (1)

Whuffo (1043790) | more than 4 years ago | (#31946230)

Automated inventory programs seem like the answer to a prayer. Unfortunately, the reality is that they're kinda disappointing. The big problem with this stuff is that it isn't psychic. Some very important pieces of information (like the physical location of the machine) can't be automatically determined. I see someone getting ready to reply with something about IP addresses; that's not as useful as you might imagine - IP addresses tend to change over time. The best you can do with IP addresses is determining what facility the machine is currently in - if you assign different subnets to each facility. That doesn't get you the detail you want, but at least you can see from the machine's IP address that it's in the data center subnet. That tells you it's in the data center, but not precisely where it's located.

There's also the machines that are shut down or off the network for whatever reason; the automated tool won't find them. And you'll find that these inventory tools can't even reliably tell what services / programs are running on a particular machine - they can only identify the ones that they know about. There's a lot of programs / services in use that will show up as *unknown* if they show up at all.

The big issue: there is much info that you need that can't be discovered by your magical inventory tool. Those things can only be discovered by a human being at the machine so they can write it down and then type it into the inventory program. You might be better off just to stop at the "write it down" stage and keep paper records instead. In either case, the success of the process depends upon the human beings collecting all the manual data accurately and updating those records each and every time they add / move / change a machine or the software running on it as well as make any changes in the network topology that might change the network address of a machine.

I've been the guy that got stuck making something like this work several times; management really wants this information (in colorful presentations) and your accounting department needs to know when you scrap a machine so they don't keep paying property tax on it. What I've discovered time and time again: everyone who touches those machines in any way needs to log everything they do to any machine or the network. In the real world, those folks will be in too big a hurry, forget to log the change, or log the change inaccurately. And the inventory programs are somewhat less than perfect: every one I've tested would miss or incorrectly indentify machines and software. Even better, I've seen phantom machines show up in the inventory program listings. A very popular gotcha: to work around the problem of turned off machines the inventory program will hold onto that listing even though it doesn't detect that machine - it might come back online in an hour or two. What happens in the real world is someone (or something) renames the machine. Now the inventory program see it and thinks it's a new machine and adds it to the database. The old listing is still there and you've got a phantom machine. Keeping track of software licenses sounds good and most of these programs promise to do that. But far too many are unable to tell an Excel viewer from Excel.

Here's what you need to know about these automated inventory tools: if you currently keep accurate paper records and everyone faithfully updates them with every change, then an inventory program can provide some value - as a way to verify the paper records and generate colorful presentations for management. But if (like most companies) you don't have paper records or they're badly out of date or incomplete because people change things and don't record the changes then an inventory program will not work for you; garbage in, garbage out. Remember that those same people will be updating the inventory program with the data it can't find by itself and if they weren't doing it before the program, they won't do it afterwards either. Also, you can not trust the output of the program - it will be inaccurate in some degree no matter how hard you try. To verify the data you need those paper records to check against so you'll want to keep the paper records current, too. After a year or so and when you're sure that the program is giving accurate data you can discontinue the paper records. Remember, a bad inventory is worse than no inventory...

My experience: every month, 5% of the machines would have a significant change that the inventory software couldn't detect. Good luck with your grand experiment; try to leave yourself an "out" when the results aren't what management is expecting.

Use the network (1)

Junta (36770) | more than 4 years ago | (#31947208)

So if you know the network topology to anchor to and enforce discipline in wiring, you can derive specific location from the network (i.e. this is the way xCAT correlates physical location to a logical entity). If you use near-rack edge switches, even if you are weak in your discipline, it at least narrows it down to the rack.

xCAT's approach is straightforward, it tracks either what is attached to an ethernet port or a physical blade bay.

It is by no means a complete 'answer' to the whole problem (it will correlate servers to a location, and even track asset tags, model, and serial number and do some other things), but it is a philosophy to apply for physical information derived from a pure 'logical' view of a datacenter.

nVentory (1)

HappyDrgn (142428) | more than 4 years ago | (#31946450)

I used a somewhat customized version of nVentory to manage my data center. The nice thing about it is that you can build clients that connect, update and register themselves through a RESTful interface. It comes with a working linux client, other clients are pretty trivial to make using the linux client as an example.

Zenoss (1)

hwk_br (570932) | more than 4 years ago | (#31946462)

I'm testing Zenoss which tracks SNMP events, SSH commands to linux and darwin nodes like df and ifconfig, and WMI. If you are organized on the switches and hostnames, it can really help! \m/

I use openAudit here (0)

Anonymous Coward | more than 4 years ago | (#31946590)

This one works well. Don't use a spreadsheet! (1)

sixminutemile (1066066) | more than 4 years ago | (#31946610) [] OpenNetAdmin provides a database managed inventory of your IP network. Each host can be tracked via a centralized AJAX enabled web interface that can help reduce tracking errors. A full CLI interface is available as well to use for scripting and bulk work. We hope to provide a useful Network Management application for managing your IP subnets and hosts. Stop using spreadsheets to manage your network! Start doing proper IP address management!

GLPI? (1)

nurb432 (527695) | more than 4 years ago | (#31946692)

Not graphical mapping but does do inventory.. Hook it to OCS and it will collect some data for you automatically.

Umm.. (0)

Anonymous Coward | more than 4 years ago | (#31946744)

This doesn't seem impossible, just time-consuming to initially produce.

Combination of dmidecode dumps, ripping through certain /etc files (or registry keys on Windows boxen), ability to talk to all switches & routers in the network and look at mac-address-table, etc.. should be able to paint a pretty clear picture of what networking equipment exists, what servers exist and of what type, etc. You have to fill in physical location and that'd be about it.

try Rackmonkey (1)

sysbot (238421) | more than 4 years ago | (#31946770)

Give this a try. It very good at tracking physical assets such as rack/server. It doesn't doe a whole datacenter but we modified the code to give just back rows of racks. []

Now the physical asset track is in one place, for inter-dependency, I create an diagram of the interconnect at a more logical level since I know the physical will be correct. This could be say a set of switches connecting to a distribution switch and etc..

Homegrown LAMP app or Visio. (1)

itomato (91092) | more than 4 years ago | (#31946788)

Price out a full-on Maximo or Altiris implementation, complete with vendor visit. Take this quote and determine how many hours of your workers' time can be covered with that payment.

Define some OSS components to do what you want to do - Monitoring, clickable representations, database connection, provisioning, etc., and write some code to glue them together. Don't forget to version control, and write documentation as you go.

I've done this before, and I've also found a way to make it happen using Visio and Access. Take some of those scrappy spreadsheets and import them into a new Access Database. Add some of the wishlist items and define some secondary tables - vendor and support contact info, customer data, performance metrics, hardware details (serial number, date of purchase, PO#, etc).

Visio documents can be easily generated from that Access database, if you spend a few days getting familiar with the process. Once that level of mastery exists, it's trivial to templatize your environment and provide a web-accessible graphical tool.

topology viewer (1)

nkovacs (1199463) | more than 4 years ago | (#31946842)

I'm not sure if there is a good free open source tool kit for this sort of thing. I do know that IBM offers a product to handle this sort of thing called TPC. The keyword you'll want to use when searching for a graphical representation of this stuff is "topology viewer". []

Racktables! (1)

AgentPhunk (571249) | more than 4 years ago | (#31946970) is a very good, Free / Open Source solution to your problem. From the SourceForge description:

Racktables is a nifty and robust solution for datacenter and server room asset management. It helps document hardware assets, network addresses, space in racks, networks configuration and much, much more!

It lets you lay out racks, assign IP Address to assets, yadda yadda. Live Demo here: []

Last code update was 2010-02-17, and the guy seems to be good about maintaining it and adding new features. Its not "sexy" in the sense that your not looking at actual Visio diagrams of the gear in the racks. If you really need that, then I would suggest the RackWise solution (, which has two offerings: 1) SaaS, where you pay by rack, at roughly $300 per rack. Its a plug-in to Visio, and your rack models are stored up in the cloud., 2) onsite appliance, where you pay through the nose (!!) but get the added benefit of integrating power management functionality into the solution.. i.e. how much power is this rack drawing, what PDU's is it attached to, etc. Option #2 is for large-ish (100+ rack) datacenters, IIRC.

Capacity Planner (1)

GWBasic (900357) | more than 4 years ago | (#31947240)

VMware's Capacity Planner will find all of the computers on your network and tell you what's inside of them. While it won't inventory UPSes, it can help you inventory your servers. []

Open Source Asset Management (1)

MrKaos (858439) | more than 4 years ago | (#31947612)

When working for a big three letter IT company I was tasked to design and implement HW/SW asset tracking system. It's a complex task and my team grew to include about 4 programmers. We automated much of the process as it is the only reliable way to gather the data. In time it also read data from other asset and configuration systems like Zenworks, TCM, TLM and even Citrix (though that was an odd fit).

After I left I decided that it would be a good idea for an Open Source project. I encapsulated and refined what I learned in the database design and XML (DTDs and XSDs) for data shredders and publishers to make data collection a modular approach and have pretty much finished that stage.

Unfortunately I am yet to implement the GUI as I am still deciding how I will approach it. In my commercial project, previously, we implemented the GUI in a web based PHP which was ok - but had limitations (requiring many stored procedures). I was thinking of using Java for the GUI as a way to spend some more time in Java and gain some experience there (I'm coming from a C background). I've been looking into Hibernate, Spring and Jakata Struts but must confess I am a total noob in these areas and don't even know what sort of pitfalls I'm about to get myself into. I've even considered EXTjs but I think I prefer the Java approach, so if anyone can offer the benefit of their experience here that would be great.

I know my designs are solid as I have used the approach before and I may have an employer who will sponsor me to write the UI as they want/need exactly this software. I didn't really think there would be enough interest in it to have a slashdot story. Lastly I'm trying to figure a funky name - I just don't think a shortened version of Asset Manager - though funny - would cut it ;-)

Re:Open Source Asset Management (1)

dtecmeister (1171031) | more than 4 years ago | (#31951044)

I think java swing is a very good way to handle this. PHP is very powerful and probably could handle everything you need it to do. Fox Toolkit or QT would work well if you don't mind C++. Not sure about licensing issues mixing QT with an open source project, I haven't had need to research that before.

Facilities Datatbases (0)

Anonymous Coward | more than 4 years ago | (#31947748)

Look into a facilities management database (goggle CAFM or CIFM). We have the data centers completely covered from the facilities perspective (electrical, mechanical, etc) and we have lots of expensive assets. It's tied to a read only soft copy of the buildings as-builts, has a PM program that creates/assigns work orders automatically, bar coding system for tools that walk away, etc. Works great for SOX compliance.

Discipline (1)

Junta (36770) | more than 4 years ago | (#31947756)

The phrasing at least makes it sound like you have some lack of discipline in how it has grown and possibly how it will continue to grow. For some topologies (SAN, network), there are technologies (the best 'generally' scoped ones aren't free) that can mitigate even without sticking to one vendor, but for bulk power topology, you have nothing but discipline to do it. At the end of the day, no matter how fancy, the basic principle will be akin to a relational database that is manually maintained for that. The danger here is that panic situations and carelessness without discipline will cause the physical reality to drift from your logical view.

Nocworx (0)

Anonymous Coward | more than 4 years ago | (#31948770) []

Disclaimer: I work for the parent company.
Might not be exactly what you're looking for but we use it for all the management stuff across three datacenters in the hosting company I work for. It can manage switches/power strips/etc and track inventory. It works pretty well and is actively developed. I have no idea what the cost is though.

Open-Audit (0)

Anonymous Coward | more than 4 years ago | (#31948878)

I use which has been around for 5+ years now.

CMDB (0)

Anonymous Coward | more than 4 years ago | (#31948928)

What you need is a CMDB there are a lot of vendors with complete solutions for Asset Management.. with GUI's of course..

MachDB (1)

Nathan (2717) | more than 4 years ago | (#31949074)

I wrote MachDB for this exact purpose. Had a datacenter of a few hundred machines and needed to keep track of them in a better way than a wiki page. [] It's automated on the back end and presents a web GUI.

no such thing can be made (0)

Anonymous Coward | more than 4 years ago | (#31949296)

u have to understand that there is actually NO universal standards ever made
thus no one such product could ever inventory all of them

u have to keep that "thing" evolving over time...
or u even may need several inventory systems
so to inventory those systems by human

not a DC (1)

brianc (11901) | more than 4 years ago | (#31949392)

Then there are tens of switches...

That's not a Data Center, it's a wiring closet.

Come back when you have hundreds or thousands of switches.

Racksmith (1)

dgleeson (1795900) | more than 4 years ago | (#31949474)

Its still in its early days but you might wanna have a look at [] . Its aiming help you map multiple datacenters spread over a campus. (servers, patches, cables....) I'm working on it with a group I knew at uni & we're looking to make a stable release within 1 month. Theres an online demo (which is the latest build) and also a screencast of out old alpha release. []

Still looking, but here's my advice (1, Interesting)

Anonymous Coward | more than 4 years ago | (#31949950)

I have used and researched this topic for years, and have been largely unsatisfied. Hopefully my trials will save you some time.

Basic Asset Management tracks the things you usually can't ping or poll -- like rack units, power strips, cables, barcode stickers, purchase and warranty information, etc. Without automation you need discipline to keep these up to date, and that usually means sticking to a process. You also need a way to audit your dataset, including tracking who performed the audit and when, to regularly ensure it's correct and that your change process is working.

Many of the suggestions here are for Operations Management (monitoring stuff, e.g. Zenoss, Nagios, OpenNMS, OpenView, etc), and have some Asset Mgmt features that are truly sub-par. Some trouble ticketing packages like Numara Track-IT and OpsManager try to cram in asset management and look tempting, but my experience with both was disappointing. We ditched Track-IT in favor of Trac, which is fantastic but has no Asset plug-in offerings.

Then you've got the hard-core Asset Management platforms like IBM Tivoli (Maximo) and BMC Remedy. I sat through a week of training for Maximo and was convinced you could spend years just rolling it out to get to a usable state. Who has that kind of time and money. They're also monsters to maintain, sitting on multiple servers and layers of javaware.

Datacenter asset management is starting to come into its own. Aperture has a product (unfortunately) called Vista that wraps a process around graphically populating and interconnecting your rack elevations on the datacenter floor. It's good, but a tad buggy, and we found support to be only mediocre. APC (the UPS people) now have similar product offerings -- I haven't used them, but they appear to have potential. Again though, these are not free but they are highly tailored to datacenters.

Free tools like RackMonkey might get you an 80% solution. I currently use visio for layer-2 and rack drawings, and have considered integrating them with a database or XML back-end for change tracking and auditing (a little-known visio feature). I'm considering writing an inventory plug-in for Trac, which is easy to use and customize, is great for tracking changes, and integrates perfectly with our ticketing, wiki, and subversion docs. I've also implemented Open-Audit to help automate keeping track of what's inside each box, again helping verify the asset system. In the end, it probably won't be one-product-fits-all, but a best-of-breed amalgamation. Best of luck!

Sysaid people!!! Google SYSAID!!!! (0)

Anonymous Coward | more than 4 years ago | (#31950202)

Sysaid is a great package and free unless requiring enterprise features. There is a CMDB module (costs) which not only provides a graphical representation of how your hardware is connected it also allows the package to be intelligent about possible implications for devices connecting to a failed device.
There is so much in this suite and it works really well. Good community aswell.

Definitely worth checking out!!! might be useful (1)

hornet136 (1421335) | more than 4 years ago | (#31950276)

Bottom line is that asset management can be a huge undertaking to accomplish. You need to find out the most important items you need to keep track of and track them well. It will probably be hard to find one application that will do it all. While primarily an IPAM (IP address management) solution, I would suggest [] It will track subnets, IPs, DNS etc and build DNS and DHCP server configurations. There is also a plugin for managing rack allocations in a fairly basic way (no connections). I think there is plenty of room for improvements from plugins, maybe something to do graphviz diagrams etc. Figured I'd throw that one out there for whoever could benefit from it.

mod 3o3n (-1, Redundant)

Anonymous Coward | more than 4 years ago | (#31950398)

ultimatel7, we

Has anyone tried OneCMDB? (0)

Anonymous Coward | more than 4 years ago | (#31951342)

Doesn't seem to have the kind of visual representation you're looking for though.
But from what I've read it would fit your specs otherwise.

Haven't tried it myself yet. Could be that it's just too focused on ITIL...

Opencabling (1)

ldardini (1004023) | more than 4 years ago | (#31951912)

Five years ago I started developing OpenCabling ( [] ) with your needs in mind. It helps me document my network, take care of every fibre, cable and port. It still needs some development, but it can solve your problems. Leandro []

CMDB (0)

Anonymous Coward | more than 4 years ago | (#31952212)

It's called a Configuration Management Database.
Google it.

Here is your answer, if you are willing to pay. (0)

Anonymous Coward | more than 4 years ago | (#31952416)

I worked for a company called nlyte Software who do all data-centre and asset management. It's not open source but it does everything.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?