MikroTik

This probe is a bunch of functions to check that the OID of the switch is equal to 4. In other words it tells you when a switch in a stack is down or not well. Put this in a function called stack_test; if(oid_column("1.3.6.1.4.1.9.9.500.1.2.1.1.6" , 10, 5), if(array_size(oid_column("1...

I found the Cisco CPU probe is out of date and found the new OIDs in Cisco process mib but the one OID either ends in 1000 or 19, this made it difficult to update the Cisco CPU probe but I got it to work, still if you want it to graph you need a separate probe that only graphs. This functions makes ...

[array_element(oid_column("1.3.6.1.2.1.2.2.1.1"),array_find(oid_column("1.3.6.1.2.1.2.2.1.2"),dev_name()))]

If that doesn't work, it is not possible...As far as I know.

I am not exactly sure what you are trying here but Device.Name is a built in function that returns 1.3.6.1.2.1.1.5.0, the system name, would that work? The function dev_name is exactly if(string_size(oid("iso.org.dod.internet.mgmt.mib-2.system.sysName.0")),oid("iso.org.dod.internet.mg...

The first post on the probe thread says to use -1 for false... Cisco CPU Type: Function Available: if(oid("1.3.6.1.4.1.9.2.1.58.0")>0, 1, -1) but -1 is true, try it... [if(-1,"-1 is true","-1 is false")] This incorrect information should corrected, anyhow why use -1 for...

Is it possible to graph all the time and still have a notification on failure if a probe fails?

Only if you make 2 probes, one to graph and one to notify, on the one to graph just put "" in the error line and don't put anything in the graph on the 2nd probe....

Congrats MikroTik, I show management products for under $100 and they think they are not good enough when in reality they are great. Disruptive pricing is hard to take seriously but you guys deserve the recognition.

Yes I modified the built in Functions, most of them... cpu_usage_available array_size(oid_column("iso.org.dod.internet.mgmt.mib-2.host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad", 10, 5)) cpu_usage average(oid_column("iso.org.dod.internet.mgmt.mib-2.host.hrDevice.hrPro...

The error line of a probe is what controls if the probe is up or down, if the error line of the probe returns "" then there is no error. You could just put "" in the error line or you could put "" in both parts of the if statement. if(x,"","") I want...

SNMPwalk in the dude doesn't know if the interger is signed or unsigned (maybe snmp walk doesn't check the MIB or it is not installed?) either way you can actually used the OID in a function or probe and and determine if the dude reads it correctly and use it. You might need to use OID_RAW or OID_Co...

Good to know jmay, another annoying dude thing resolved

The counters for the dude are 32bit, 4.29billion is the largest unsigned number 32 bits can hold. I doubt they are able to prefer 64 bit counters yet. They should have a checkbox to prefer 64 bit counters so that when you monitor any link the underlying function checks to see if the equivalent 64 bi...

Inside the OID function is a built in 300 second negative cache timer. Meaning if a failed probe is stored it will stay failed for 5 minutes, this can cause probes to never respond to re-probes. I don't remember if I completely tested this but if the negative cache timer is not cleared just before t...

Could you graph the daily or monthly total? You might be able to use the rate command to get the difference from the last time you grabbed the number of prints. Or if there is a way to reset the count, graph the value then reset the counter.

Thanks that rocks, glad I could help out. If your around Denver sometime

Yep good point, them rrd tools graphs from the past would only render when someone wanted to see them.

Solectek bridges have kind of the same trouble where the OID is actually something different on each link even though they use the same hardware something is tacked onto the end of the RSSI OID. For example the actual oid is 1.3.6.1.4.1.2890.7.6.2.1.5.6.0.128.72.116.21.168 on one bridge, Since it is...

I wish they would put the web server back or export graphics in V6, I tried many times to find something that could replace the dude and nothing... still looking at moving to v6 to get away from the 2gb file limit, hope that is fixed in v6 as well.

Crazy stuff jmay, what I didn't try was to fix graphing negative numbers, I just graph the negative number from my radios. Maybe multiplying by -1 is storing the previous value and screwing up your probe? Anyhow you can graph negative numbers so maybe that would work...

rssi.png

In 4.0b3 the export function is on the last tab of the map settings button.

schedule.png

eriitguy, the only idea I have is your probe interval is conflicting with the probe. On the appearance of the device place your function for CPU [Procurve_Cpu_Test()] and see what the label does. Check your label refresh interval and modify it to be the same as your probe interval and see if they ac...

I needed to produce a map that I could share with every department and have it simple to understand. I have some busy maps and have not tried a minimalist map so I just had to share a fake version of a minimalist map. minimal.png The clouds are devices that use ping or HTTP that are white if they ar...

Add 1 in the function since the CPU can return 0% utilization. subtract 1 in the value and error lines... Name: Procurve_Cpu_Test Code: if(oid("1.3.6.1.4.1.11.2.14.11.5.1.9.6.1.0",10,5), oid("1.3.6.1.4.1.11.2.14.11.5.1.9.6.1.0",10,5) +1 , "False") in your error line if(...

I don't use auto discover so creating a stencil map is the next best thing. The above XLM would not paste into another dude server which is confusing. Anyhow here is a screenshot of some objects that I put on a stencils map. I added the probes I like and save them then when I add a new device I just...

Are they on the same network? A second device doesn't have to be very big, it could just be a Green/red dot next to the main object. You could auto discover the devices. I know a lot of things in the dude are configuration intense but at least you have that option.

Make the probe for the 2nd ping by hand, a bunch of functions let you specify the address. Create another ping function, give it a name, and specify the address. Another thing you can do is If you want a label that shows an SNMP value from a different system you can use the IP of the remote system i...

Create a map called stencil, paste these and use copies of them to create other maps. <?xml version="1.0" ?> <dude version="4.0beta3"> <Device> <sys-type>15</sys-type> <sys-id>240078804</sys-id> <sys-name>Router</sys-name> <addresses>0.0.0.0</addresses> <dnsNames>Router</dnsNames...

The unit for your probe could just have ms in the field or like the previous poster mentions post your data source.

Ability to control the icon color based a color variable that can be used in a function. Nested Functions on interface device labels, this function doesn't work; [if(Interface.OperStatus="up","","Interface Trouble")] Allow custom field to be used as a variable in a func...

Hey you guys are in DENVER wow kick ass, I am just up the street. I just wanted to shout out to all the MUM users and say have a great time, I didn't even realize it was this week until this morning! I can't miss work since I didn't plan to show up :( so much things going on here... I think there ar...

Open a command prompt and type C:\putty.exe and hit enter it should run putty.

if it shows

C:\Windows\System32>c:\putty.exe
'c:\putty.exe' is not recognized as an internal or external command, operable program or batch file.

Putty is not at the root of the C: drive.

I don't believe so, the default is 30 seconds so that is already 1/2 as much traffic. If you want to go back to the stone age of 5 minute intervals that would be 10x less traffic. I like frequent polling so I can see changes quickly. I would run 30 second intervals but i wanted to make sure my datab...

support@mikrotik.com

Sounds like you have it correct, there is another question about this issue today but I don't know you might have to send support an email and get them involved.

Cool glad you got it working.

Make a clone of Notification, rename them so one is for down and the other is for up. Setup the down notification to have 100, set the up notification to have 1. Enable them both... so email of down is from unstable->down and email of up is from down->up.

I run with intervals of 1 minute for most things, the dude is really efficient, generating an average bit rate of 500kbps. That is with about 700 devices. I don't think the dude generates enough traffic to worry about unless you are using a t1 line. I have added ping to over 2000 devices as a test (...

Search for posts with rate, diff32 and diff64. Rate returns parameter divided by time elapsed since previous call. So if you grab the diff of a counter you can use rate to give the change over time.

rate(diff32(oid("1.3.6.1.2.1.2.2.1.14.12")))

the global settings button is the top left not in the middle... that is the per map button. both buttons should have a general tab though so something is wrong.

Can you post a screenshot of the general tab?

It is Notification in the polling tab.

Great, always confusing when users have different experiences!

Yeah it is super handy, i was trying to prove that wireless was fine and that the server sucks so I got that running and grabbed a Netflix ~20mbps, tried to populate a table from the server that sucked ~600Kbps, lol, server still sucks

There are 3 locations for notifications, at the global level with the settings button on the top left, in each map with the settings button in the middle, and settings for each device on the polling tab. Having a notification checked at the device level would send only for the device, having map not...

I meant snmp v2.
OH there maybe look at your oid, it is supposed to be "1.3.61.4.1.3808.1.1.1.4.2.3.0"
Load: [oid("3.6.1.4.1.3808.1.1.1.4.2.3.0")] %

The rate command is the rate of change over time, maybe you could use it. OR How long a value is cached is configurable, Make 2 functions, the first one has a 50 second cache, the 2nd one has a 20 second cache. Use both functions in a probe that is polled every 30 seconds then to subtract function 2...

Probably a v3 issue, test with v2

check out the tool Asithappens it is java

Nope but there is "ntopng 2.0" I found that looks ok for flows...

Sorry Eine I am not running the new version. I have been watching and there are good things in the newer versions but I have not had time to start a new server and migrate. I still run 4b3 and will migrate when i stop seeing game breaking bugs in the top of the forum. They should absolutely make neg...

so the Big question would be ;
can you put a TCP probe in a "Function"

Call any type of probe from a function would be a great addition.

L

~[0-9A-Fa-f]{91}0.* in the "receive" window of the probe. this works very well and simple. Seriously nice work figuring that all out. If your TCP Probe has ~[0-9A-Fa-f]+0F2$ of course it should show "up" when the door is shut. The regex ~[0-9A-Fa-f]{91}0.* doesn't work in regex ...

Lebowski , i found direct way to send the right command from default tcp probe. ( see my edited post ) . can you / or someone give some light on how to write the function to parse the data that comes back ? Can you put a regex that matches door closed in the receive? ~[0-9A-Fa-f]+0F2$ or ~[0-9A-Fa-...

The function for if_7_out_rate needs to have an if_7_in_rate that creates the correct bit rate for inbound traffic. Duplicate your current low output rate probe and save it as "high input 7" (something else) and replace the error line with if(If_7_in_rate() = -1,"SNMP Read Error"...

There is a bug when a device has many mac addresses the window scales to fit the mac addresses but that keeps the window from shrinking, Delete mac addresses from the device view to reduce the window size. Set the mac detection to none.

Create a panel and open it separately, click Panels, click +, on the blank window type a name for the panel, click on the down arrow near the top left, choose a specific map or any other "window" you want to display, on the right hand side click the vertical or the horizontal split, to add...

This works on the appearance of a device label... [if("1","true","false")] You should be able to do this [if(ros_command(":put [/system identity get name]"), ros_command(":put [/system identity get name]), Device_Name)] I don't have time to get the syntax...

upgrade to v6.39rc7

The data is probably too hard to deal with but the data stored by most network monitoring applications is in what is called RRD for Round Robin Database. Cacti, zenos, mrtg, nmis, and many others all use the same method, RRD. RRD data can easily be examined with RRD Tool or Ddraw. The trouble is the...

Make as many new functions as temper sensors listed under this oid 1.3.6.1.4.1.9.9.13.1.3.1.3 (SNMPWALK THIS OID) on a router there might be 6 temperature sensors... NAME: Cisco_Temp_1 CODE: if(string_size(oid("1.3.6.1.4.1.9.9.13.1.3.1.3.1", 10 ,5)), oid("1.3.6.1.4.1.9.9.13.1.3.1.3.1&...

(Future) request :
How to write a probe / function for a NON SNMP device in Dude,

Look at execute, you can call other programs/scripts and batch jobs.

Check the firewall logs... I believe port 2210 and 2211 for secure.

Using CustomField2 as a processed value doesn't seem possible... Try this type of thing on the appearance of a device label and you get Device.CustomField2 instead of the value in the field. [if(Device.CustomField2=2412,"stuff","no stuff"] [oid("1.3.6.1.4.1.9.9.272.1.1.2.15....

I really like to use functions. Since you can differentiate when a device is in an error state or completely down. Click on functions, click + , put syncstate as the name, paste the code from below and click ok. Of course be wary of the syntax errors if Name: syncstate if(string_size(oid("1.3.6...

This is the fix... sorry for the syntax error. if(oid("1.3.6.1.4.1.17713.21.1.3.1.0",10,5)=2,"Sync OK",if(oid("1.3.6.1.4.1.17713.21.1.3.1.0",10,5)=3,"Hold off",if(oid("1.3.6.1.4.1.17713.21.1.3.1.0",10,5)=4,"Retraining","failed to read ...

Heck here is the OID table in question, so 0 is a value (don't bother since 0=false) if you really need to test for 0, add some other value to it first. Object ID: 1.3.6.1.4.1.17713.21.1.3.1 GPS Current SYNC State: 0 - The Initialization State 1 - The No Synchronization State 2 - The Synchronization...

The if statement in the dude has the following description first parameter - condition, second - returned if condition yields logical true, third - returned otherwise if(x,return this value , else return this value) but you can put an if inside an if... if(x,return this value , else if(y,return this...

In 4.0beta3 if you copied an object the history data was cleared, select the link that has the chart and click copy, see if that fixes it.

There was a change in 4.b3 where when you click the copy button history data is cleared, it is also a bug since it clears the history of the original object. It would be nice if "copy" gave a popup to keep history or delete it. Also removed was the RRD editor, if you placed a graph on a ch...

You might be able to drag and drop to upload like in the old version.

On 4.0beta3 sometimes you have to restart the client If the paste button is not visible. You should examine the XML, paste the device into notepad to see the XML of the item you are copying. Since 4.0beta3 is on windows sometimes I will RDP and run directly to do copies.

Cool, glad I could help.

The current dude doesn't support map based access. If each user needs full access you could just install a server version for each user. If they are using windows?

Sorry I don't have a OmniTIK U-5HnD for testing, change it to warning... Error: if(get_rb_cpu_load()<>"False",if(get_rb_cpu_load()-1< 95, "", concatenate("Warning: high CPU = ", get_rb_cpu_load(), "%", rebooting) ) ), "Device down") Create a notifica...

There is a flaw with the logic, anytime the probe can't read the CPU load that router will reboot also if the cpu is actually 0% utilization it will cause a reboot. Error: if(get_rb_cpu_load()<90, "", ros_command("/system reboot")) Make your function so it returns "False&quo...

There is a negative cache time that affects probes you can find some examples on stuff I built by searching. The setup you have sounds weird if you have 1 probe per hour and 4 retries you have the dude configured to take 4 hours to show an outage but you wait 5 minutes for response. Even high latenc...

Nice to see Dude lumbering up.
Feature request: Add 'Name' to option when adding new device (not just IP).

You can add devices by name but it has to be fully qualified...

navrik; Place the following on the device label;
[oid("1.3.6.1.2.1.2.2.1.8.1")]
What exactly is displayed on the device? use that in your comparison.
#

I am not sure what is supposed to be happening here but normally I would create a function to return a value or false. Then in the probe error line perform the error checking. For example here is a function that returns Linux CPU or false; LinuxCPU: if(string_size(oid("1.3.6.1.4.1.2620....

So you can tell if there are packets on the link, when many packets link goes red. #

Hi! Error: if(oid("1.3.6.1.2.1.2.2.1.8.1")=2, "", concatenate("Link to ", oid("1.3.6.1.2.1.31.1.1.1.18.1"), " is down")) place this on the device label [oid("1.3.6.1.2.1.2.2.1.8.1")] Use the exact match that is displayed in your device la...

Hi! Error: if(oid("1.3.6.1.2.1.2.2.1.8.1")=2, "", concatenate("Link to ", oid("1.3.6.1.2.1.31.1.1.1.18.1"), " is down")) place this on the device label [oid("1.3.6.1.2.1.2.2.1.8.1")] Use the exact match that is displayed in your device lab...

if(device_property("Type")="DES",device_property("Type"),"error") In the error line any value returned is considered an error. You must return "" if you are not in an error state. In your case you are returning a value so it is in an error state. if...

Do you have many probes that you made your self? Maybe disable them, see if it still crashes. Also can you run your old monitor to try to resolve the issue faster? Meaning disable a different bunch of probes on your old monitor to eliminate those probes being associated with causing the issue. Same ...

Changing OIDs is a pain, there is not a good way for the dude to determine what new oid it becomes. Maybe the oid is always one of a certain group of oids? Could you use part of the oid to make your own rate calculation. or create a link for each oid and just track them all? Just ping the other side...

I am assuming that the interface oid is changing every time the user logs in? When you select the interface in the drop down does it have a slightly different look? More detail would be good like how you configured the existing probe.

There is a way to have down links show up as a color on a map but you have to sacrifice the link utilization color to do it. One easy method would be to just add [Interface.OperStatus] to the label and be done with it. If the link labels are turned off that won't work. Having the top map labels turn...

I monitor about 500 devices but at one time for a test of our desktops to see if they turned them off at night I build a map that would scan 192.168.0.0/16 and install ping on any device it found, that one map had about 2200 devices with a single probe of ping.

That would be really great. Another way to see the RRD data in 3.6 is to add the graph to a chart then inside the properties of the graph you can find the detail. I don't think this will help much but it is there.

In windows the file is \\cworks1\Cworks1-C\Program Files\Dude\data\dude.db The Database is in single user mode so you must stop the dude process to open the database. Just copy the database off to a work area and restart your collector... Once you have the database you can install an sqlite plugin f...

In your function add 1 to the value just before you examine it so it will never be 0. You can subtract 1 when you graph it for accuracy.

if(string_size(oid("1.3.6.1.4.1.3530.5.8.0",10,5)),oid("1.3.6.1.4.1.3530.5.8.0",10,5)+1,"False")

They are derived from internal functions, I doubt you could build your own alternative since they don't allow for variables in functions. It would not be hard for them to allow for a check box that just preferred 64 bit counters. If you have a 10gb link averaging 300mbps.That is the 32 bit counters ...

Nice to know how to clear out the database!

The data is xml but it is compressed in the database. There are tools to access the database but no one has reported unzipiping and exporting the RRD tables in there.

In the top right corner there is a layer called dependencies, I have not used it but if it works you can draw dependencies and then the outages should only be sent for the top dependency. Inside Notification you can delay on the advanced tab, you can also set repeat interval and count. Also devices ...

Is there any way to clear the graphics files? You might have to just delete the probe and put it back on... In 4.3b if you copied any object the collected data for the original object gets erased before the copy, so on the device that is creating those graphs you should be able to just click copy t...

You can select "connect only" and that will verify that the port is open. There was a post a while ago where someone actually did test a radius login with complete details on what to send and receive for radius so the best way to firgure out a winbox send receive is to capture a winbox log...

The Dude only uses the 32bit counters, the best you can do is create a new interface and set the speed to 4294967295, then it will record somewhat correctly but there will be gaps in the graph when the counter rolls over. You could lower the polling interval but you don't want to collect so much dat...

Your available line will always return true, -1 is not null. ^^^^ is not exactly correct... First the available line is only used when using "discover" to automatically add a probe to a device. The dude attempts to use the available line of the probe on the device when you click discover ...

What port do you want to check? Say you want to test if port 81 is open, start a new probe, select TCP or UDP, enter the port and check connect only. This will cause the dude to only check if the port is open.

Your available line will always return true, -1 is not null. There is always some difficulty getting the return value passed back correctly. Put your execute on the appearance of a device and verify that the function works. If you search my posts for execute you will find some good detail. Something...

Make sure to lower UAC and make sure the dude is running as administrator. When installing the dude make sure you are using an administrator account since it is probably setup to run as a service. There is an old thread somewhere about this problem although not win10. This could be a new problem.

Do you know how to configure SNMP? Does the device you want to monitor have SNMP? The dude will request SNMP values and display them in the context of the device. Is the device you want to monitor on a map? Right click that device, click SNMPWALK. That is a separate tool that shows if SNMP is workin...

It is complicated by how long your polling interval is and how fast the link is. The rate function uses the diff of the byte counter readings over time to get the current value. If your polling is not often enough the counter could roll over especially when it is a 10gb link. It could be something e...

Thank You MikroTik, I am certainly excited that I don't need to find another product as good as The Dude. I have looked and looked and there are good products out there but nothing as intuitive or as visual as The Dude. After reading the current status of the new Dude and how it works I am going to ...

Gsandul created a probe that would notify you if bandwidth on an interface dropped below a certain threshold. You can set that on the managed port and have it notify if the bandwidth is very low or none. Not the greatest solution but it is free.

The dude doesn't handle snmp traps. Is it a static route on the router or are you using a routing protocol? If you are using a routing protocol you might be able to determine the gateway of last resort or the next hop address and if it changes you could send an alert. I was working on a slightly dif...

Gsandul had this rate function that you might parse into something you can use...
http://forum.mikrotik.com/viewtopic.php?f=8&t=43396

Here is int g0/0 adding both in and out octects and multiplying by 8 to go from bytes to bits. [round(rate(diff32(oid("1.3.6.1.2.1.2.2.1.10.1"))*8)) + round(rate(diff32(oid("1.3.6.1.2.1.2.2.1.16.1"))*8))] Seems to be somewhat accurate... says 100,000 when both interfaces are at 5...

You could monitor something past the unmanaged switch or buy a managed switch. Maybe move a wireless access point or some other device that is always on to the unmanaged switch to get the benefit of a managed device past the unmanaged device.

Check if a certain program is running Applications in the not responding state do not show up in Window 7 SNMP view as not responding... They show as running which sucks. I had a not responding program, found it in the registry then found the run index and checked all entries of the run index so I ...

On the left contents column scroll to the bottom, double click "tools", Click red "+" button (to add new), Name the tool. In the command box enter command line to run the tool. Here is are some examples: SSH C:\Progra~1\VanDyk~1\SecureCRT\SecureCRT.exe /T [Device.Name] HTTPS &quo...

After a recent routing issue I found LiveAction which used to be actionpacked networking. Their product looks like something I will run in addition to the dude. I still have yet to find anything as functional as the dude. I found intermapper to be not very intuitive last time I tried it. Intermapper...

You can show an oid from another device by adding the ip address to the oid request. Users:[oid("1.3.6"',5,0,"192.168.1.1")] With the above you would be needing to determine each AP oid then if the AP oid had connected users... returns value of given snmp OID. Only first paramete...

It is unlikely, I have a Cisco WLC and have total users on the label of the WLC. I have not finished trying though. I can create a template that enables SNMP on each access point. If that works then maybe there is an OID populated with users connected. My last attempt I was able to reach the AP but ...

Search this forum for vacuum you will need to learn about sqlite and try to understand how the actual sqlite tools work. There are commands that clean, delete, and check data i don't know most of them off the top of my head. Vacuum is a process where empty table fields are removed to reduce the data...

Nice job!

You can clone the functions and the probes to new names and put the correct OID (assuming it has moved). If it is a bug and the OID is not being populated you are SOL. I am not going to go looking for where it moved if it did.

Actually it is more complicated... Since the probe interval is 30 seconds, One missed probe will cause "unstable' 3 missed probes will cause "down" which will generate a message that will sit in the queue for 5 hours then get send regardless of the current state. The only way to make ...

I do not know if it works correctly, you will have to do some testing. Devices only go from unstable->down so that is the only check mark you need to get a down message. If you want an up message you need down->up. If the device goes down and you have set a 5 hour delay I assume you will get a messa...

Sounds like it is not supported, maybe try on a couple different operating systems?

If it is a Mikrotik box you could use execute to run the command and return a value. Execute is a pain in the ass but once you get it working it is fine.

To create the notification that will be delayed for 5 hours, click on top left, settings button, polling, notification edit list (the dot dot dot button), click +, manually clone "Notification" or which ever one you have that actually works right now and give it a different name. On the Ad...

You need unstable->down, to go from up->down you have to remove retries from the polling.

In cases where you absolutely have to see 2 links just make a static cloud object (right click + add static) and attach the second link to it instead of your 2nd switch. (repeat for as many links as you need to track) You can build a probe to verify that both links in a LACP are up or that it is run...

The problem with the database is when users set raw value keep time longer than 2 days and/or have many devices. You can set your keep times and examine the Approximate storage size, I would keep the approximate below 1.5gb. If you add a bunch of devices check the data folder after a month or two an...

Log into the device, snmpwalk the device, wait till finished, search for your id with "find" in snmpwalk, log out, snmpwalk device, check same oid for your usename. If you find the oid create a function/probe that generates a notification...

I didn't understand the first post, sorry.

If your router should populate an SNMP value. You could snmpwalk the dude and find the status then build a probe.

You need SNMP setup on routers and switches to read their topology and if you are actually doing networking you should be able to SSH into each device and add SNMP or determine the existing SNMP settings. How many switches routers and firewalls are in the network? BE CAREFUL. The dude could peg the ...

Can you snmpwalk the device you need the information from? What license? Windows server license doesn't expire. If the software that you have to re-license reports to SNMP that would be documented by the company that gave you the license. You can run an external program, it is a pain to get working ...

Use array to verify that there are always 6 of those running. You can find a post by me that shows how to monitor notepad.exe. Using the same method but instead of just determining if notepad.exe is running just us array_size to make sure 6 Microsoft.Dynamics.Nav.Server.exe are listed if not 6 then ...

Almost everything I monitor has a static IP address or a reservation. If you just give everything a DHCP reservation your network is much more predictable. I am not monitoring desktops. I mostly monitor server and lots of network gear. Anything that is added is added manually and probes are added ma...

You will be able to view a map of the objects in your network someday! If you click the settings button (top left) click polling, click the ... button, double click the notification that you want to change the subject for... add or remove any number of things to the subject and body... Notification_...

So either set a long lease time or put reservations for every computer. I agree it should be easier. I never us auto discover and I use "name to IP" but I have a DNS server where I can manually add and move things around. Anyhow there is another problem you will run into with auto discover...

Add some "Allowed Networks" in the server. I have seen some weirdness when trying to add large networks and not having any networks in the Allowed Networks in the web server tab. Add the network that is directly connected to the dude to the panel for server config/web if your dude server i...

I think the MAC address and Lookup sections are for after the device is discovered. Yes, I mean set "mac to IP" in the device settings for a single device after it is discovered. Supposedly if you have already discovered an object and have it configured with "mac to IP" selected ...

I assume you are using auto-discover. It is supposed to work if you set it up "mac to IP". If you tried this and it keeps adding a 2nd object with a new ip address and the same mac address you could keep this from happening by giving everything a much longer lease or a reservation in DHCP ...

You could subtract uptime from days in the month and if uptime > days in month SLA is 100%.

Now to figure out how to get days in month... you might be able to use string sub_string...

back in 10.

A quick way to think of how this works is source and destination mac addresses change at every hop, IP addresses do not. When I create a packet for www.google.com my computer does a DNS lookup to find the IP address for Google. It does not create an ARP request since the address for Google is not on...

The mac address is only available by Address Resolution Protocol (ARP) on the locally connected segment, meaning if you are not on the subnet that you are trying to get the mac address from you would have to ask the router for the mac address table that the router is keeping. Then you could parse th...

Under settings, discover, Device Types you should have both the required and allowed services selected and they should be unique for that device type and the device type itself needs to be allowed to be discovered. The required services is what is used to determine what device type it is and the all...

So you have a manageable switch, you can log into it with SSH or Telnet. You probably use a web browser right now? Anyhow download Putty, put the ip address of the switch in there and click ssh and log into the switch. You will have a command line you can type at. Type ? and things like Show Run and...

Maybe a corrupt database, try to vacuum the database to see if that can prove there is trouble. Note if the database is over 2gb you will need to reduce your raw keep time.

The dude does not abide to https.

Create two Functions, one to grab the name of the disk and one to find the remaining percent. Name: disk01name code: oid("1.3.6.1.2.1.25.2.3.1.3.1" ,10,5) Name: disk01test code: if(array_size(oid_column("1.3.6.1.2.1.25.2.3.1.6",10,5)),round((oid("1.3.6.1.2.1.25.2.3.1.6.1&quo...

httpgoogle.png

First make backups, you can copy and paste but it can be tricky... Copy and paste while using a client seems to stop working. Second when you copy something the object on your clip board is XML. So each object you copy you can paste it into notepad to see the XML. Examine the XML to familiarize your...

Have not heard of this issue, double check that "notification" has all the variables that you are interested in. I made mine very plain, both the subject and body of "notification" in my configuration have [Device.Name] [Probe.Name] [Service.ProblemDescription]. This gives just e...

Just right click the subject device and SNMPwalk it, search the MIB for the interface, you are looking at standard mib2 stuff the interface details start at 1.3.6.1.2.1.2.2.1.1 which is the interface index and if there are 10 interfaces there will be 10 interface index oids (and any manually added v...

Where is the server installed? If you didn't disable the local server and actually connected to a remote server you would see discover when starting the dude again since it will connect to the wrong database.

I doubt that is the trouble...
Lebowski

It is surprising that auto discover is not better documented. It is really good for finding things and adding things to maps but beyond that it needs some bugs addressed. Like it will add networks that have already been added. For me I didn't want every possible probe added to every device so fine t...

"raw value keep time" is the actual values collected if you set it to 10 days then your graphs will show exactly what was collected for 10 days... Is there a way to edit these settings in version 3.6? DOH I thought you meant edit the values. I don't remember what it looked like in 3.6 but...

There is some error but I am not sure how to solve it, I found setting probe interval longer reduce the error but then you are checking your web pages less often. I have set all my web page probes to 5 minutes and I still have one that is flakey and says connection closed.

The wiki has tons of information but the dude will search a range of addresses and ports. When an ip and port are found the dude will "connect" a probe and add that device to that map. You can assign ranges to discover and extra hops go past router boundaries. So if you specify 192.168.1.0...

No but it is easy enough to have auto discover go extra hops...

show cdp neighbor should show you which switch is connected to which switch on what port. They borrowed the command from Cisco. Cisco Discovery Protocol... as far as I know.

lebowski

I doubt there is anything built-in that does what you want. openwrt might not even populate the values into the table. Lets first assume that the values of username and signal level for each client is listed in snmp in openwrt. First setup SNMP on the device, if you can use snmpwalk to examine the i...

Change the allowed network to something like 192.168.0.0/16 then if it starts working change it back to 0.0.0.0/0 if you need to.

When you examine the function "oid" you can specify the IP address as the 4th item from there you can add the three together and you can just place that on a device label to have a total displayed at all times. I think it would work as a probe as well if you want to graph it. I think you n...

I thought I would like intermapper but I don't, I just found Total network monitor and tclmon. I might try tclmon.

http://docs.softinventive.com/En/Total_Network_Monitor
http://tclmon.sourceforge.net/screenshots.php

No, I am too tired. Ok I will. Create a notification and test it, if it doesn't work try again. The notification you are trying to get working is the global one on the top left corner settings button. BUT FIRST populate the smtp server settings in the same top left settings button. Click General tab...

YES, ok Maybe, umm NO you can't. well on second thought yes you can if you try like really really hard.

It is unlikely that the client actually adds a SNMP value of the frequency to the operating system snmp values... if it does it might work in one OS but not another. In Windows and Linux you can enable SNMP (ask google) and walk the computer after adding it to the Dude to see the SNMP values that th...

You want to make sure to change the ID for each device so that it is not confusing the program but other than that I don't know what else needs to be unique.

Let us know what/how you end up doing it, it will help others.
Lebowski

Greetings. I am on day 2 of using "TheDude" and I'm confused on the purpose of Services/Polls/SNMP/MIB's/Functions. In a nutshell, I have all the MIB's imported, can SNMPWalk the device (Microwave Radio) and now I want to (SolarWinds terminology only perhaps) "Poll" the device f...

Here are a few ideas... You could start a new dude server then manually add a single device to it then select it and click the copy button when you copy a device it is actually a text section of XML. If you paste that into notepad you could decipher the XML and create a script that could generate XM...

Not exactly sure what you are getting at but you can create a function to read a value and put that function on the appearance of any device. In the appearance tab many of the built in functions can be picked from a drop down box.

Lebowski

Assuming you are running the dude on PC, you would have to write your own macro or script to switch between maps, you could do that maybe with something like tinytask. That would allow you to manually create a macro of your mouse clicks, just record clicking through every map one at a time and then ...

Hate to admit it, I have no MikroTik devices but I use the dude and run it on windows.

Check to make sure that settings have not changed. Under the device settings, polling, make sure "use notification" is "not selected" to be certain that you are using the global notification, then under settings (top left), polling, make sure notification is checked, click the 3 ...

"raw value keep time" is the actual values collected if you set it to 10 days then your graphs will show exactly what was collected for 10 days...

Top left settings for global, click "Chart" tab. Change "raw value keep time" Default is 2 days (then it will be averaged into 10 minute averages). Why they changed from 5 minute averages is completely baffling. If you only have a few devices go crazy and set it to 3 months... st...

To verify that communications, install and run Wireshark on your dude server, put the filter on capture options "host x.x.x.x" (a broken host) then use snmpwalk to see if the packet ever comes out of the dude (which it will) what does the response look like? You should see a three way hand...

The dude is an snmp probe it will ask a device directly and make a best guess at what the device is like a router or a switch. Also it has auto discovery which will sweep the network by range of ip addresses and determine if certain services are available, if a match is made it will automatically cl...

I have never seen a setting for it, be careful changing the raw keep time, if the database grows above 2gb the dude will crash.

Try a non real address? Are you sending mail to an exchange server?

Is netwatch an executable? Is it running on a windows computer? If so enable snmp, make a function, make a probe. Put the correct name of the exe in the function Function; Name: find_netwatch if(array_size(oid_column("1.3.6.1.2.1.25.4.2.1.2",10,0)), array_find(oid_column("1.3.6.1.2.1....

I don't know of a way besides putting the error line of a probe in the appearance of a label which is not what you want.

Awesome, I just wish I had a use fot it but it goes to show the very real power of the dude and the great ideas behind the design.

First I must say I am sorry that I didn't get my last post correct, there is a typo in the function. Second I am sure this will work. Create a bat file called getlog.bat there are 2 lines in that file. The program "more" returns the value in fig.log so there is no need to make the execute ...

For default email notification. On General tab, enable checked, type email Server is configured on the global general tab to IP address. To: only contains a single email address. Subject and body are; [Device.Name] [Probe.Name] [Service.ProblemDescription] Schedule has all hours; In advanced Only ha...

I should have mentioned you do not need to compile a MIB for the dude to read an OID but if the value returned is not a standard value snmpwalk might not be able to show a meaningful result. Compiling the MIB lets SNMPwalk show the correct description for a value and if the value is something intere...

You are welcome, it is such a great program I hope someday they continue to develop it. First just right click on the device in question and SNMPwalk the device. (You will see how awesome this tool is) In the new window check to see if values are returned. If values are returned click on subtree and...

Available is only used to determine if the probe should be installed on a device when using auto discover. Just put 1 in the available line for now and see if that allows the probe to function. MAYBE this would work. (execute is very finicky since it is run as "system") Change the function...

Availible: if(fhq_value()>0, 1, 0)

is that the fix?

Hey Sayani, try either one of these in your function; if(string_size(oid("1.3.6.1.4.1.25506.2.6.1.1.1.1.6.8")), oid("1.3.6.1.4.1.25506.2.6.1.1.1.1.6.8")+1 ,"False") if(string_size(oid("1.3.6.1.4.1.25506.2.6.1.1.1.1.6.8.0")), oid("1.3.6.1.4.1.25506.2.6.1.1...

Make a file "who.vbs" Set objNetwork = CreateObject("Wscript.Network") Wscript.Echo objNetwork.UserName Execute who.vbs in your label... I will finish this tomorrow... THIS is crap... there are quite a few reasons why getting the username of the logged in machine is a pain. Runni...

Change the owner of the registry entry to administrator?

I was really interested in Intermapper.

Still nothing like the dude so just don't let your file reach 2gb or do daily backups...

Maybe just capture traffic with wireshark and then analyze what is creating the traffic. If it is the same time everyday that will be easy to capture.

Backups or some other script doing a file copy of some sort?

The HTTP probe will look for a word on the resulting page but displaying that word is problematic. The right way to do this is to grab the xml with an execute function and return you own result, then use that in a label for the device. There are quite a few posts around asking how to use the execute...

This has been documented in the Probe thread, search the dude forum for notepad.exe. As you are starting to understand on the computer you want to monitor you must enable SNMP with a community string to be able to find the programs that are running. Then you will need to create a function and a prob...

It is going to be your SNMP configuration again. I don't own any HP switches but I guarantee you can get the information you just need to make sure SNMP is allowed and the community strings are correct. If the dude can reach the switch with Ping then just use SNMPwalk to verify that SNMP is correct....

I think it is that the result is a string then later you are comparing a string to an integer.

Your error line on your probe shoould be;
if(getTemperature()<25, "", "NOC Room Temperature above 25 c, Please check A.C")

Lebowski

I have been using it, syslogs should be recorded into Program Files\Dude\data\files I believe all syslog entries are recorded into action log. If you want them accessible in the dude you need an entry in settings->syslog that is "log to syslog". I think that is the default. If you have cus...

4.3, added 2 interfaces to a chart, right click open separately, no problem.

Link utilization - Chart_2014-08-25_10-02-31.png

Maybe shut down the dude and vacuum the database?

Lebowski

Search found 1619 matches