Quantcast
Channel: THWACK: Message List - Network Performance Monitor
Viewing all 25172 articles
Browse latest View live
↧

Re: New SolarWinds Installation

November 23, 2015, 10:25 am
$
0
0

The FSDB is going to get potentially hit with a lot of data and the way it stores its data on the hard drive will mean it should have very fast and unrestricted access to its storage.  Check out the requirements and if you honestly think you can set up a VM that can meet the requirements, then give it a try.   I'd recommend a separate physical server with its own local storage myself though.

http://www.solarwinds.com/documentation/netflow/docs/releasenotes/requirements.htm

 

64 bit windows

3GHz or faster, 4-16 CPU's

20Gb on 7200 rpm disk or better + 8GB per sustained 1000 fps (flows per second), Raid 0 or 1+0 supported.

16 to 128GB RAM for production

 

Everything about this screams "performance", if you put it on insufficient hardware, you get problems with performance.

 

If you put it on the same physical server as your Orion database, they'll both be fighting for CPU and storage performance.

 

Not that familiar with DPA.

 

Don't skimp on your DB servers, they need maximum performance.   I'd virtual the poller before I did either of these.  Of course it all depends on your environment for what you might get away with.   Yours doesn't look like it should be that busy, so I think you could probably get away with anything for awhile...

↧
Search

"Top X Interfaces" resource

November 23, 2015, 11:16 am
$
0
0

So for years, I have had a resource called 'Top 10 Interfaces' on my Nodes Detail pages that is supposed to show the top 10 interfaces based on 'Avg bps'. Today while looking at a Node Details page, I see that instead of it showing my interfaces by Avg bps, it's literally showing the first 10 interfaces on the device. I do have 'Avg bps' configured in the resource, so I'm pretty sure it's not a config issue. Have I just not noticed that this resource has always been this way or did my recent upgrade to 11.5.2 change the behavior?

↧
↧

Re: NOC issues

November 23, 2015, 11:22 am
$
0
0

To be clear I use Firefox Tab Slideshow, It's an add on to the Firefox browser.

 

Is your SQL database on the same server? Are you using SQL Express?

↧

Re: NOC issues

November 23, 2015, 11:40 am
$
0
0

Thanks jotsusa, I knew you set that up a while ago and had good results.

↧

Re: Admin Account issues

November 23, 2015, 12:30 pm
$
0
0

So it looks like it has happened again. Not sure why replacing the job engine made it temp happy, but it's broken again. Any input is appreciated.

↧
↧

Re: Top 10 Interfaces by Percent Utilization shows no data.

November 23, 2015, 12:41 pm
$
0
0

I don't see anything in the syslog. I will check out the DB and try the configuration wizard and will report back. Thanks for the help.

↧

Re: NPM's NodeIPAddresses table is not up-to-date

November 23, 2015, 2:54 pm
$
0
0

I have a similar problem where some ICMP-only monitored devices are in the Nodes table, but do not show up in the NodeIPAddresses table.

 

When I merge the 2 tables together to have a full inventory view of IPs seen by my nodes, I get multiple NULL values on the join.  Is there a way to trigger an update of the NodeIPAddresses table?

↧

Re: NOC issues

November 24, 2015, 9:24 am
$
0
0

We originally went with Firefox because of Internet Exploder was not as reliable/Secure with other Tools we were using at the time.  I believe we set this up during 10.2 or even version 9.  We have had great success with it.

 

As far a using the NOC view feature, you may be overloading on occasion the Orion Server with other various operations and products.  Does the page/graphic/map not load at all to give you the error? 

↧
Search

Re: Error with Discovery Processing Results NPM 11.5.2

November 24, 2015, 9:29 am
$
0
0

Thanks.  Hopefully they have something coming that can fix this permanently.

↧
↧

Eaton UPS pollers

November 24, 2015, 9:29 am
$
0
0

I have been trying to figure out how to poll the Eaton UPS to see battery levels, load, etc. I found the MIBs on their site, example

 

xupsOutputVoltage OBJECT-TYPE

    SYNTAX INTEGER (0..2147483647)  -- UNITS RMS Volts

    ACCESS read-only

    STATUS mandatory

    DESCRIPTION

    "The measured output voltage from the UPS metering in volts."

    ::= {xupsOutputEntry 2}

 

xupsOutputCurrent OBJECT-TYPE

    SYNTAX INTEGER (0..2147483647)  -- UNITS RMS Amp

    ACCESS read-only

    STATUS mandatory

    DESCRIPTION

    "The measured UPS output current in amps."

    ::= {xupsOutputEntry 3}

 

xupsOutputWatts OBJECT-TYPE

    SYNTAX INTEGER (0..2147483647)  -- UNITS Watts

    ACCESS read-only

    STATUS mandatory

    DESCRIPTION

    "The measured real output power in watts."

    ::= {xupsOutputEntry 4}

 

So how do I get Orion to use the correct poller so that I can see this information?   Do these mibs need to be added to the database?

↧

Dashboard view help....

November 24, 2015, 9:43 am
$
0
0

I am trying to create a dashboard view.  I thought this would be a simple task but I am really stuck on problem...  I am using a summary view,  The summary view has 3 columns.

I am adding a custom object resource to each column.  Column #1 will show the Node availability Column #2 will show the latency and column will show the response time.

 

I will end up with several rows of graphs.  The idea is that when a manager wants to see the stats for a particular node the find the node of interest and then read the graphs from left to right each graph will have one set of data for that node. 

 

My problem is that all of the graph views are not the same height or width.  Every graph must be the same width and height otherwise the dashboard looks horrible and is hard to read.  Is there a way to get each graph to be the same width and hight across all columns?   If not I was considering creating a new summary view with one column and using custom HTML to create a 3 column table and link all of the graphs to cells in the table.  But how do I get a graph to show up in custome HTNL in Orion 11.5.1

↧

Autonomous AP Maps?

November 24, 2015, 12:35 pm
$
0
0

I know with LWAPP's we can create heat maps. Was just wondering if there is anything similar for heavy AP's?

 

I like how the node details shows signal strength of connected and historical clients, would there be a way to utilize that to put "average signal strength" in specific locations on a map?

 

Thanks!

↧

Re: Monitor Unauthorized Access on a Cisco Router

November 24, 2015, 12:50 pm
$
0
0

well, you could always use a TACACS or RADIUS servers for authentication. Typically those applications may have a reporting function.

 

If not, you could poll for sessions (basically find the OID that presents the results of a "show user") and compare them against an approved list. That might take some manual labor (in creating and maintaining the approved user list) but it'd meet your requirement.

If you're looking to alert when someone attempts to access them, you could implement quiet mode and trigger an alert when this is engaged. There's some pretty good tutorials about implementing quiet mode, like this one: Cisco Quiet Mode | CCIE or Null!

Either way, search for the OID that can give you something to check against and create an alert based on a Custom Poller for that OID.

 

Or execute the command "login on-failure trap every #" which would send a trap every time someone failed to provide authorized credentials properly "#" times. Then you can use the trap viewer to send an alert any time this trap is received. A similar action can be performed if you have a syslog viewer with alerting functionality.

 

You've got options, boss. Hope this helps!

 

V/R

Dan

↧
↧

Alert Manager- Alerts with variables?

November 24, 2015, 12:57 pm
$
0
0

Hello all,

I am a newbie so please go easy on me.

 

I have 60 ESXI hosts that I want to create alerts for and am really hoping I do not have to make a separate alert for each host.

 

My hostnames are as follows:

c01esxihost01

c02esxihost01

c03esxihost01.... all the way up to 62.

 

Here is what I have so far:

 

Is there anyway I could make it something like this:  Node Name is equal to c**esxi01

I do see the "starts with" option, it would be nice if there was an "ends with"..

 

Any advice would be much appreciated.

thanks,

Bmoore

↧

Re: Alert Manager- Alerts with variables?

November 24, 2015, 1:35 pm
$
0
0

you want to monitor this 60 host in one alerts? If yes, you can to create a group and add all  hosts, and to create alerts for him. 

↧
Search

Re: Alert Manager- Alerts with variables?

November 24, 2015, 2:49 pm
$
0
0

Are your ESXI machines setup in a group?

↧

Re: Eaton UPS pollers

November 24, 2015, 3:17 pm
$
0
0

You will likely need to create an SNMP Universal Device Poller for each aspect you wish to monitor:

 

Universal Device Pollers (UNDP) - SolarWinds Training - Network Performance Monitor - Videos | SolarWinds

 

The Universal Device Poller application is installed on the server's desktop, so remote into it to tart the program.  You will need to figure out the SNMP OID of each aspect you wish to monitor, create a UNDP for each monitor, transform it as necessary (maybe your SNMP device outputs a fraction (0.54) and you want a percentage (54), you can create a transform to apply a formula to it), and assign the poller(s) to your UPS(es) via the same Universal Device Poller application, and then you can create alerts based on these UNDPs.

 

SNMP appears to be bailing wire and magic.  What's up with that?

 

What is SNMP | SNMP tutorial | SNMP basics :: OpManager Tutorials

 

What's an OID?

 

What is the SNMP OID?

↧
↧

Re: Access Points on Down Status

November 26, 2015, 8:39 am
$
0
0

I add the APs as nodes - discover the AP range over night and then in the morning run a powershell script to add 'AP' to a custom field. from the wireless table Orion knows about from the controllers. Then you are just monitoring them as normal nodes.

Obviously depends on your licence etc.

↧

Re: Setting up Alert

November 26, 2015, 8:52 am
$
0
0

I've attached the UDP here.

 

I've setup the alert rules but haven't done any testing to see if it works as planned.

↧

How to configure maps on the Orion Web Console?

November 26, 2015, 9:56 am
$
0
0

I'vecreatedmapsNetworkAtlasbutwheninserted in thewebconsoleOrionmapsappearsto mein all groups, that is not listed in thegroupthat I havesetbutin all groups, asI cando tospecify that theI createdmapdisplays only inthe groupyou desire?


I have madethefollowing settings

 

Home > Groups > (my group) > Customize Page > Add reource to column > Network Maps > Map


In the Map Dashboard


Edit > (my map) > Submit


Can you help me?


↧
Viewing all 25172 articles
Browse latest View live
Search