Wednesday, September 11, 2013
NPC
Managing application performance and delivery across modern enterprise and service provider networks is a complex undertaking. The CA NetQoS products take a “Performance First” approach to monitoring networks and systems. This approach is based on the premise that you must capture and analyze data from applications, devices, and the network itself to get an accurate and comprehensive understanding of how well an organization is supporting application delivery.
The CA NetQoS Performance Center is the industry’s most comprehensive management software suite offering role-specific views of application response times, network traffic composition, infrastructure health and performance, unified communications quality of experience, early warning alerts, and packet-level diagnostics.
The CA NetQoS Performance Center suite unlocks the intelligence needed to manage the network for application performance with a full suite of analytics and reporting fueled by the most comprehensive data collection available.
With the NetQoS Performance Center, IT organizations can:
• Helps ensure problems are resolved quickly
• Mitigate risk from planned and unplanned changes
• Deliver consistent application performance
• Take measured steps to optimize application delivery
• Reduce and avoid costs
Wednesday, April 10, 2013
Quick NetVoyant Device Query
This query will the total of devices for each device class in NetVoyant.
mysql nms2 select count(*), class_name from devices d join device_classes dc on d.class_id=dc.class_id
group by class_name;To show this as only enabled devices, add an additional filter for the poll_enable field.
mysql nms2 select count(*), class_name from devices d join device_classes dc on d.class_id=dc.class_id
where d.poll_enable=1 group by class_name;
NetVoyant Device list
mysql -h 10.80.3.8 -D nms2 -u nmsuper -predpoint -N -e "select smgr_host, dev_alias, model_name, ifindex, ifdescr, pollinst_desc, concat(d.sys_name,'_',i.ifindex) from service_managers s, devices d, device_models m, interfaces i, pollinst p where s.smgr_id=d.dev_owner AND p.device_id=d.device_id AND d.model_id=m.model_id AND p.pollinst_id=i.interface_id AND p.pollinst_enabled=1 and d.class_id in (1,7)" > NVInterfaces.txt
Method to export NV device list data?
Mysql.exe nms2 -e "select
dev_name from devices where poll_enable = 1" > path\export_file.txt
Wednesday, March 27, 2013
RA Fix #2
1)You can try the following on the RA console.
mysql -P3308 reporter
delete from database_servers_scheduler where resultcode not like '%SUCCEEDED%';
2)Then recycle the ReporterAnalyzer Manager service on the RA console and the "DSA Loader Service" on the DSA and see if those tables start updating or not.
Then wait about 15-30 minutes to see if any new data shows up.
3)If not, just delete the interface from the "Administration->Physical & Virtual", this will delete the agent record from the database and should clean up the DSA.
4)Then delete the full router from "Administration->Enable Interfaces"
5)RDP to the Harvester and run the following....
mysql -P3308 harvester
truncate table routers;
truncate table interfaces;
6)Recycle the "Harvester" and "Reaper" Services on the Harvester.
7)Wait for the router to show up again in "Enable Interfaces" and enable the interfaces.
8)Wait at least 15-30 minutes before expecting any data
9)If no data, run an NFA Parser report and verify that specific interface has more than 50 KB of traffic over a 15 minute period.
Tuesday, March 26, 2013
RA Router fix issue
If you look at the report for this interface for less
than two hours do you see any data?
Are other devices using the same DSA having the same
issue?
You could use the query below to find out what other
interfaces are using that DSA. Use the
router Address in the first query to find out what DSA its using, then use the
DSA address in the second query to see what other devices are using that DSA.
mysql -P3308 reporter
select DataServerAddress, RouterAddress from
agents_all_view where RouterAddress='x.x.x.x';
select * from agents_all_view where
DataServerAddress="x.x.x.x";
Can you also upload a SupportCIG file if possible so I
can see what is going on in the logs?
You may also want to check the DSA credentials on the
Administration->DSA page to make sure they are a valid login for the RA
console and the account is not locked out or anything like that.
Monday, March 25, 2013
Friday, December 14, 2012
Restarting a Windows Server from another NetQoS server
Sometimes when you restart a server, it hangs and doesn't restart. This means that the RDP sessions fails but you can ping the box.
Use the shutdown -i command from the command line of another server. Fill out the GUI to force a restart.
You can also run the following from command line:
shutdown /r /m \\servername
Where servername is either an ip address or the dns name of the server.
Wednesday, November 28, 2012
NV database repair (BI)
Completed successfully on 11/27/2012
NV database repair:
show databases
use nms2
repair table pollinst use_frm; <enter>
repair table group_instances use_frm; <enter>
Thursday, October 25, 2012
Tuesday, September 25, 2012
Free Up Space on the C Drive
***PUBLIC INFORMATION - THIS CAN BE SHARED WITH CUSTOMERS AS NEEDED***
ISSUE
How do I free up some space when the C:/ drive is near full?RESOLUTION
The simplest thing to do to free up some space on the C:/ drive is to run the NQDiskCleaner utility. Just download the tool and run it anywhere on the server. It will remove any unnecessary and temporary files.ftp://ftp.ca.com/pub/netqos/supporttools/releasedtools/nqdiskcleaner.zip (see options below)
Use windirstat to find out what is taking the most space. http://sourceforge.net/projects/windirstat/.
That will show disk usage in a visual way where you can see what is taking up the most space. Look at C:, and see if you can clean up some disk space.
If this doesn't free up enough space, you can move the page file over to the D:/ drive. To do this, just follow these steps:
- Right click on My Computer -> Properties
- Select Advanced tab.
- Click Settings in the Performance section.
- Select Advanced tab.
- Click Change in the Virtual memory section.
- Note the values under 'Initial size (MB):' and 'Maximum size (MB):'
- Select the D: drive. Select 'Custom size:'
- Change the values to match the ones noted in step 6.
- Click 'Set'
- Select the C: drive.
- Select 'No paging file'
- Click 'Set'
- Click 'OK' three times.
- Select Yes to Reboot the server.
You could also try moving the windows temp folders to the D:\ drive, where there is usually more space.
- Right click on My Computer and select Properties.
- Click on the Advanced Tab.
- Click on the Environment Variables button at the bottom.
- In the bottom section of the window (System Variables), scroll down to the bottom where TEMP and TMP are.
- For both TEMP and TMP, left-click on the variable name, then click "Edit".
- Change the variable value from "%SystemRoot%\TEMP" to "D:\netqos\temp".
- Click ok, then repeat the procedure so both TEMP and TMP are set to the new directory.
- Create the D:\netqos\temp directory if it does not already exist.
- Restart the server.
Another thing that can be done is to clean up the web service logs, if they are enabled, in C:\WINDOWS\system32\LogFiles\W3SVC1.
You can disable them by going into IIS manager and right click the Websites folder and select properties, and uncheck the Enable Logging box. Then reset IIS in cmd: iisreset.
NQ Disk Cleaner
-Deletes hidden folders in C:\windows. Typically these are backup files from Windows Updates containing files replaced by specific Windows OS patches.-Runs Windows Disk Cleanup. It enables all options, including hidden options within the tool then runs the tool to delete files.
Options enabled: Recycle Bin Content Indexer Cleaner Downloaded Program Files Internet Cache Files Memory Dump Files Microsoft Event Reporting Offline Files Offline Page Files Old ChkDsk Files Remote Desktop Cache Files Setup Log Files Temporary Files Temporary Offline Files WebClient and WebPublisher Cache Active Setup temp folders Compress old files-Deletes IIS logs in c:\windows\system32\LogFiles\W3SVC\ or c:\winnt\system32\LogFiles\W3SVC\
-Empties the Recycle Bin
Friday, September 21, 2012
NetFlow Configurations
Subject: NetFlow Configurations
From: Webb, Rob
To: TEAM - SABU - NETQOS - TECHNOLOGY
Date Sent: 10/4/2011 10:59:48 AM
From: Webb, Rob
To: TEAM - SABU - NETQOS - TECHNOLOGY
Date Sent: 10/4/2011 10:59:48 AM
I recently answered a question about what’s the best way to configured flow data on a router. Here is what I’ve found through my experience. Thought I’d share with the entire team:
There are several options for configuring NetFlow:
1) Ingress only (default): this must be applied to ALL active L-3 interfaces in order to properly calculate the outbound on any give interface. This is because netflow is being collected inbound to the router and in order to calculate the outbound all traffic destined out a particular interface is associated with the outbound metric for that interface. The limitation is when you have a router marking QoS, the marking occurs after the NetFlow is calculated. Works with v5 or v9
2) Egress only: works the same as Ingress but in the opposite direction. The advantage is now the NetFlow is being calculated after the QoS marking…and therefore you will see how the traffic is marked as it leaves a given router. Again, ALL active L-3 interfaces need to have this configured in order to provide accurate inbound measurements. Works with v5 or v9
3) Ingress & Egress: This works with NetFlow v9 (IOS 12.4 and newer). Yes, there is an increase in flow data, however now only the interface(s) of interest need the flow commands. If flow data is collected at multiple points (i.e. inbound on one interface and again as outbound on another) the v9 flows contain a marker that identifies the interface and direction. RA is intelligent enough to only count the most relevant data not eliminate double counting of this traffic. Works with v9 only.
In summary:
Ingress = on all active L-3 interfaces / not good if marking (QoS) in the same router as reporting
Egress = on all active L-3 interfaces / good if marking (QoS) in the same router as reporting
Ingress/Egress = best option, only configured on interfaces of interest
Do not mix and match. Pick one and stick with it on any given router!
_________________________________________________________________________
ca Technologies | Rob Webb | Senior Principal Consultant, NA Solutions Architecture Group | 412-916-8023 | rob.webb@ca.com
One additional point I'd like to add is that RA will double count flows if ingress AND egress is enabled on the LAN and WAN interfaces AND using NetFlow version 5.
Subscribe to:
Posts (Atom)