cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 14

Multiple pollers - one graph

I've spoken with Solarwinds staff about this before, but wanted to post it here to try and generate further interest.  I'd like the ability to be able to put more than one poller on a single graph.  This should encompass not only multiple pollers for a single host, but also multiple pollers from multiple hosts (for a summary-type view).  A couple examples of how this would come in handy:

1.  We have several interfaces that connect to our upstream providers.  We currently use Cacti to put all those interfaces' traffic stats into a single graph, with each provider a different color.  See pic for example.  The graph shows both in/out with inbound traffic on the "positive" side and outbound traffic on the "negative" side.

2.  I use Cacti to poll our DNS servers for failed queries vs. successful queries vs. NXDOMAIN responses.  Putting them all into a single graph would not only save real estate on the page, but also make it much easier to see the differences in scale of all pollers.  Same holds true for disk I/O read/writes - it doesn't make sense to have two separate graphs for this.  Some might also want to poll other interface-based stats that aren't already polled by Orion, which have separate OIDs for in/out - another example of unnecessarily needing to create two graphs.

Tags (1)
82 Replies

Agree 100% Please address this and the many other shortcomings that make it next to ridiculous to spend money on Solarwinds over the many much more capable free products.

0 Kudos

We're about to do a blog post on what we're working on in the next few days.   We'd love to get everyone's input once you've seen that list.  There are a lot of top priorities (e.g. 1, 1a, 1b, 1c), but we believe we're working on a lot of your top requests 😉

0 Kudos

*drool* 😉

I'm personally most concerned with features like scalable administration, (user/node/custom poller), multiple pollers on a single graph and let's not forget SNMP table correlation.  From a "my time" standpoint the administration issues are foremost, but for business intelligence for both large and small customers the graphing and table handling are paramount.

Peter

0 Kudos



*drool* 😉

I'm personally most concerned with features like scalable administration, (user/node/custom poller), multiple pollers on a single graph and let's not forget SNMP table correlation.  From a "my time" standpoint the administration issues are foremost, but for business intelligence for both large and small customers the graphing and table handling are paramount.

Peter



 

I couldn't agree more (especially graphing and table handling) but then again my tone apparently doesn't reflect the broader user base.

0 Kudos

0 Kudos

*sigh* Just as I feared none of the major shortcomings are being addressed, just a bunch of low hanging fruit that leave glaring fundamental capability holes intact. Just out of curiosity, why did you come into a thread called "Multiple pollers - one graph" and get everyone's hopes up when that's not at all part of this announcement?

Just tell the developers to download a free copy of Zenoss and Cacti and just copy everything they do for free! At the pricepoint an enterprise Orion deployment runs, they should have the resources to do it.

*sigh* Just as I feared none of the major shortcomings are being addressed, just a bunch of low hanging fruit that leave glaring fundamental capability holes intact. Just out of curiosity, why did you come into a thread called "Multiple pollers - one graph" and get everyone's hopes up when that's not at all part of this announcement?

I definitely hear your frustration and respect your opinion, but I don't think your tone is representative of the broader Orion user base.  There are numerous top feature requests (e.g., AD integration, dependencies, service level groups, etc.) listed in the post that have gotten a lot of folks excited.  Believe me, there were threads equal in length or in many times much longer (and spirited) than this one to include those features. 

My goal for including the link was to ensure that folks on this thread understood that many of their other top feature requests ARE currently being worked on and that we're listening even if they aren't getting everything they want.

For you and others watching this thread, please do keep it alive and kicking with requests around this subject (include detailed use-cases) as this helps us raise the priority.

0 Kudos

Right, here's the problem with the SW product development approach.

The things we're asking to get fixed mostly aren't big feature requests, but very obvious core functionality issues that anything calling itself an SNMP correlation engine should have pretty well nailed.  Most of the smaller customers simply don't have any idea what they are missing, and most larger customers are so sick of hearing the "we hear you" treatment and no action to back it up.  It's a level of frustration that is generally high enough to warrant loss of pleasantries.  I have other things I love about the product.

The low-hanging fruit approach is great on paper, people asked, we delivered, etc.  Auth even if it's AD based is a great step in the right direction for larger shops.  Huge.  It means I'll have to continue to support an AD infrastructure for very little good reason (time for ADAM, perhaps), but it's a great move.  The problem is, 500 people might really want pretty maps on the home page, when 2 people want custom table correlation, joining and graphing, but the second feature is more intrinsic.  500 people might want the ability to email out PDF's, which are a gigantic waste of bandwidth, and I just want to be able to have scheduled CSV exports of custom reports I created in NPM and NCM.  I want to be able to alert on data that's being shown in reports, but I can't figure out where to get it from.

I certainly don't expect SW to completely ignore the voices of the masses, the newer, smaller customers that want the bling, and don't notice the lack of substance, but it bothers me to no end to hear these features getting ignored release after release, all lip service and no action, for 3 years is ridonkulous.  

Another critical feature that's been whispered about and no mention:  The ability to compile our own MIB database.  Every other SNMP collection product in the world provides this feature.  It's nothing that needs to be re-invented.  Destiny does it weekly, but then decides to "clean up" and strip out critical yet obscure enterprises without notification.  How about those custom poller and trap alerts I had for my telephone switches that tell me if an trunk group carrying 3m minutes a day is down?  Egregiously unacceptable.  How about the ability to have custom pollers auto-assign based on multiple criteria during node addition?  I need to be able to configure standing orders and minimize the care and feeding of the product.  How about interfaces that randomly go into unknown status all the time?

How about the ability to graph interfaces in aggregate?  A bunch of good work has been done on traps and syslog, but it needs to go further.

How about granular user authorization/authentication and a full-passthrough of credentials from NPM to NCM?

How about node import templates, user creation templates, and the ability to import csv files in through those templates to provision nodes, users, and even alerts.  There are a bunch of alerts that due to limitations in the product, are painful to create for different events (traps, syslog, and custom pollers), I'd like to be able to create templates and import the variables in CSV format instead of all the point and click BS.  Even copying alerts, exporting / importing actions, etc, it's painful as hell to manage and add to weekly.  Now along with that, I'd like reports that both summarize and detail both my alert definitions, but also alerts on a per-node basis, counts, counts of traps/syslog, etc.  

I'm sure there is more, but that's as big of a chunk as I feel like regurgitating at the moment.

Peter

With every release, we do our best to include as many features as we can and to ensure that they represent what most customers are asking for.  Unfortunately, no matter what we pick to include, we're inevitably going to disappoint some people.  We think our approach works for most, but we can definitely understand why folks whose top features aren't included are upset.

0 Kudos

A response like that would be valid if basic core functionality wasn't ignored release after release.  What I'm asking for shouldn't be viewed as feature requests, so much as making basic core functionality inline with standard offerings.  You Product folks need to spend more time looking at the product objectively, and less time giving lip service.

I fully understand the needs of the many philosophy.  I'll go out on a limb and say that probably more than half of us have seen "Star Trek:  The Search for Spock".

The problem I have is that people have been bitching about a serious, obvious, and often seemingly insane set of issues, that keep getting ignored, release after release.

The only conclusion I can safely draw is that Solarwinds as a company, and individually through it's employees, doesn't take the needs of larger scale customers seriously.

People ask for critical features over and over, and SW's Product team responds like Catherine Tate's Lauren character: "Does my face look bovered?"

It's a sick game where we, as customers, thinking that we have spent enough on a broken product suite and it's annual 20% to warrant being heard and responded to, and while SW pretends that's what's going on, empirical evidence would suggest otherwise.  Remember the conference call of over a year ago where you, and the other Product managers sat there and told my boss and I to our faces that while you couldn't be sure, and couldn't make any commitment on exactly when, that it was reasonable to expect by the end of the year we'd have something functional to make the administration overhead less of a burden.  Guess what, nothing but lip service.  It's 6 months past that deadline.

SW as a company simply doesn't have any credibility at all, when it comes to the needs of larger business.  Replacing it is painful and expensive, but potentially worth it to be in business with a company that is actually responsive.  Every other vendor we have doesn't do business like SW.  Cisco, Juniper, Adtran, and others work with us in a genuine fashion to address our business needs.

I too, dislike using a public forum to voice such things, but whether voiced privately or publicly, the truth remains the same.

Peter

0 Kudos



Every other vendor we have doesn't do business like SW.  Cisco, Juniper, Adtran, and others work with us in a genuine fashion to address our business needs.

Peter



I'd love to have your sway with the above-listed companies.  There are some basic features we've been begging for in all those vendors' various equipment and we've been ignored for years.  I've found the complete opposite - unless you are an AT&T or Verizon who needs a very specific feature that no one else on the face of the planet needs, then you are going to get the cold shoulder.

I personally think that SW is much more responsive to their customers than any of those companies, at least in my experience.  It is just frustrating when there are seemingly "simple" features being requested that Orion or NCM doesn't have that free software has and has been doing for years.  I have THREE network monitoring applications running on my network, all of which perform similar functions.  I've been running NPM since version 5, I believe, so that shows you how long I've been waiting to get rid of these other applications.

I count no less than 17 customers asking for this feature to be included in the next version.  One of those is a reseller claiming his customers are asking for it to be included.  I'm not sure how large each of these SW customers' installations are, but I'm sure if you add them all up, it equals a pretty penny in yearly maintenance costs.  As much as you don't want to hear it, I'm sure some of these customers are going to vote with their wallets when it's time to renew maintenance and ask themselves, "What is it that I'm really getting with this 5-figure number for yearly maintenance?"

And, I hate to latch onto this, but a search for 'Meru' on the forums elicits only postings in the Content Exchange from some guy who seemingly works at Meru.  WTF?  I know in the grand scheme of things, this feature is probably a very small part of the resources devoted to the next version, but c'mon.

0 Kudos

bleearg and Peter--

Thank you both for your feedback. I'll make sure the PMs see it.

If I could remind both of you that we have a policy against profanity and  both of you have words-of-interest that qualify (yes, even in acronym form). If they had been much stronger, I would have been forced to delete your posts b/c we do not edit posts. 

We welcome the passion of your critiques, but please, let's keep them as clean as possible.

Thanks for your attention to this.

M

0 Kudos



If I could remind both of you that we have a policy against profanity and  both of you have words-of-interest that qualify (yes, even in acronym form). If they had been much stronger, I would have been forced to delete your posts b/c we do not edit posts. 

We welcome the passion of your critiques, but please, let's keep them as clean as possible.



I wholeheartedly disagree with this statement, but okay.  I've seen worse on the forums and never seen a public dressing down of a user.  The things either of us said were nothing you wouldn't hear on daytime or primetime TV.  We are all adults here and it's not like either of these posts were laced with the type of profanity you'd find at an Andrew Dice Clay performance.  Nobody is ready to go all postal on Solarwinds here.  In fact, both of us are touting the great features and functionality of SW products, but merely pointing out some flaws.  I find this to be pointless nitpicking.  No one is getting angry here and we don't need riot control.

In any case, this is de-railing the thread.

0 Kudos


Chris,




Overall, in my short time as a SolarWinds customer, I've seen a lot of good things.  You guys seem genuinely interested in your customers.




As for the what's on tap posting, there are a few things on that list we wanted, which is really great news.  But I do agree with the majority here who think a tool like NCM that is version 10 and does not have what we see as basic capabilities is slightly insane.  And I'm going to try to keep this focused on the forum topic, although there are other areas we'd like to see addressed as well.




Putting multiple pollers on a graph or "easily" putting multiple graphs on one page to create dashboards or having graphs in reports (trending,etc) is what we would consider network monitoring 101.




That AD integration and dependencies are finally arriving is great news, it will ease some of the Orion management pain.  But we need to have these other basic issues addressed in the very short term.  The look on my associates and managements faces when I show them an Orion report that it just a bunch of numbers says it all.  We've got vendors beating down the door that can demo their bling bling interfaces and we've got to justify over and over why Orion is a good and the right investment.    Just give us a little bling, please.




For what it is worth, I absolutely hate posting comments like this on forums, because our overall impression of SW and the tools you provide is positive.  But it seems like this is the ordained method for feedback.


John




Chris,

Overall, in my short time as a SolarWinds customer, I've seen a lot of good things.  You guys seem genuinely interested in your customers.

As for the what's on tap posting, there are a few things on that list we wanted, which is really great news.  But I do agree with the majority here who think a tool like NCM that is version 10 and does not have what we see as basic capabilities is slightly insane.  And I'm going to try to keep this focused on the forum topic, although there are other areas we'd like to see addressed as well.

Putting multiple pollers on a graph or "easily" putting multiple graphs on one page to create dashboards or having graphs in reports (trending,etc) is what we would consider network monitoring 101.



This is very well put and I agree with it 100%.  There are certain things which should just work out of the box and this is one of them.  So is a simple page with multiple interface graphs on it.  These are things that make us continue to use Cacti and MRTG, which I've been wishing to get rid of for years.  I am very excited about many of the new features you guys are working on, but Meru Wireless integration?  Really?  How many customers could possibly be using Meru Wireless devices that it's considered a "top priority" when a couple of relatively "simple" enhancements like this get overlooked. 

I know I started this thread two years ago and multiple people have tagged that they want this functionality.  It really is "network monitoring 101", considering that it's something that MRTG has done since I started working with it in 1998.  It's now 2010 and the leader of network monitoring applications doesn't have this very basic functionality?  As much as I hate to say it, I'd strongly suggest downloading a copy of Cacti and seeing why I both need to keep it around and why I need to get rid of it.  I mean, I am running a separate server *just* for multiple pollers in a single graph capability.  How insane is that?

Eric:

  Thanks for staying on this, we too would like/need to have this functionality, so add my voice to th chorus.

Orion has moved from a simple monitoring tool to a powerful diagonistics tool, we have the need to present diverse information  in consolidated ways. I have spend a large amount of time creating custom pages to do this (see Attached)

If we could have an easier way of presenting this, it would add a huge value to the Orion product but I would like to see a common graphing engine for all modules with common controls and look and feel. 

It's true Orion is superior to cacti/nagios in so many ways, I am frustrated that those zelots harp on relitivly minor issues to knock Orion down. 

Jim

Orion XXX is not superior to Cacti in Cacti's areas of core competency.  If it was, nobody'd be bitching.  Try out Cacti's graphing sometime.

And which Orion are you talking about?  NPM?  APM?  NCM?  All 3 products have some excellent functionality and some glaring holes in functionality that have been known for years and ignored.  SW needs to be a little bit less reactive and a little bit more proactive about dealing with these ancient, well documented issues.

Zealotry is apparently not confined to us open source enthusiasts either.

Nagios would essentially correlate to APM.  It's a far more developed tool for monitoring applications, although Nagios administration and config file structure is completely focacta.

0 Kudos

Peter:

  I agree with you, SW should be addressing these issues if it is going to effectively compete.  Also the zealotry comment wasn't aimed externally or personally at you,  I have internal factions who completely refuse to acknowledge any value in commercially developed software at all  and will implement open source to the detrement of the business systems with the justification that it's "free".

As for superiority, we have a dynamic infrastructure, if a system is technically superior but is more difficult to implement, maintain, manage, and scale, to the point that it is not kept current,  then its value is diminished.

So if I'm a Certified MVP, then I'm a zealot; but our Solarwind system is monitoring ten times the systems with ten times the data and requires ten times less resource to do it.

Jim

 

0 Kudos

What kinds of systems, servers - platform/OS, network devices - type/transport?

I'm curious about the 10 times the systems with ten times the data and ten times less resources.

I will say that I can wholeheartedly blame the performance of my installation on the database server, it's not adequate and it's replacement is already online, just not ready to cut over to.  I also understand that should be getting better with 10, which can't be deployed until it's properly QA'd, and that unfortunately means poring through everything with a fine toothed comb.

Peter

0 Kudos

We monitor 6000 elements on 900 nodes using one Orion server (Poller/website/DB) across 16 sites worldwide

HP DL380G5 2-dual quad 3.16 Ghz - 32 GB ram

Database is 25-30 GB most all of it in RAM (check the attachment page life >22 hours hit rate 99.9%)

monitored systems include:

Windows 2000 2003 2008 - Linux Debian Redhat - Cisco Swicthes Routers Firewalls - Citrix Netscaler Load Balencers - Juniper SSL VPN - Tipping point IPSs - APC UPSs PDUs - Detroit Diesel Generators (through APC Netbotz) - APC and DataAire AC units

Cacti/Nagios has about 90 nodes and a guy who spends 25-40% of his time maintaining it, I spend 2-5% of my time on Orion

My advice - put lots of RAM in your DB server.

Jim

0 Kudos

Right now I'm doing about 200 nodes and 5500 elements if I include custom pollers on 3.3GB of RAM on both the database and the poller.  Netflow cannot be turned on :(, but I'm well aware of why.

The new poller and database server both have 12GB, DL360G6 for the poller, and 380G6 for the database.  I spend about 40% of my time on Orion right now, trying to get it to do things it's resistant to doing.

Peter

0 Kudos