savell

Comments

  • I'm fairly certain that in order to roll-back to 9.1, that you will need to perform a database restore to the backup taken just before the 9.5 upgrade. The configuration wizard for 9.1 will not have the logic to remove the database changes implemented as part of the 9.5 upgrade. Dave.
  • Server OS Windows 2005 and SQL 2005 Server CPU type, number and speed Database = Xeon 16 Core 2.53Ghz 24Gb Memory, 15k SATA Drives RAID10 Pollers (x3) = Xeon 16 Core 2.67Ghz 12Gb Memory Hot-Standby/NCM/Backup Database = Xeon 16 Core 2.53Ghz 24Gb Memory, 15k SATA Drives RAID10 Number of Nodes 4500 Number of interfaces 12000…
  • Jed, I might have things wrong, but I think the entSensorValue (CISCO-ENTITY-SENSOR-MIB) is meant to be node specific only. If the snmp get is returning a "1", then I assume that this is telling you that the sensor value you are doing the query against is a sensor type of "truthvalue" which will return only two values -…
  • Thanks Chris. Yes spoofing is an option. Unfortunately for us, we have firewalls in the path of most of our Network modules. Where once they were only used on the edge, we find them increasingly placed within the Network. Firewalls don't tend to tolerate us Network types spoofing packets! Dave.
  • I was looking tonight at using SQL Reporting Services to generate my own reports from the raw data contained within the policycacheresults table. I'm using the data to eliminate devices from the report that don't have exceptions (when you have policy reports against thousands of devices, but only a handful of exceptions it…
  • At the risk of repeating the above, I have the same issue. With a few thousand devices, it's typical for a large number not to have any configuration changes made for months. The nightly backup job does execute, but of course discards the unchanged config file.It unfortunately makes the pie graph meaningless for us (when…
  • Out of interest Chris, is the ability to strip out the "Original Address=192.168.1.1" type tag (that kiwisyslog can insert when forwarding a syslog) on the list somewhere. As kiwisyslog and Orion users, we would love to be able to log forwarded kiwisyslog messages under the real address within Orion (firewalls prevent us…
  • Thanks for the offer Christine. We are just heading into a long weekend here however, and I've already left (lucky for some - but why I'm still on the PC is another thing). Will follow-up as a priority after the weekend. Dave.
  • That suggestion would also neatly solve the issue regarding how to determine how many users are using the web interface at any point in time. If this is even a possibility, we would need a user profile field for the user name. One of the little problems at the moment is that while we have the userid, in an active directory…
  • Is there going to be further guidance on this Andy? We are migrating from Call Manager 5.1 to CUCM 6.1 (Cisco Unified Call Manager as it's now named ) and need to make some decisions regarding management. We really need some function to display basic Call Manager stats. Dave.
  • Try something like this: SELECT Nodes.Caption AS NodeName, CustomNodePollerStatus_CustomPollerStatus.Status AS your_custompollername FROM ((Nodes INNER JOIN CustomPollerAssignment CustomNodePollerAssignment_CustomPollerAssignment ON (Nodes.NodeID = CustomNodePollerAssignment_CustomPollerAssignment.NodeID)) INNER JOIN…
  • I agree with Andy's comment regarding the number of spindles you have for the database. Our database currently works very well on RAID10 array with 6x146Gig 15k SAS drives, front-ended with a raid card with 512Meg cache (25% read and 75% write). This handles 16,000 elements across three pollers nicely (with minimal disk…
  • Same here (download button doesn't work). Dave.
  • Here is the solution I came up with to deal with a selection criteria on the Import Nodes job.. I added it as a feature request (i.e. allow node selection via an SQL statement within the import job) - but this has been doing the job for us for some time now. Dave. 
  • Thanks for taking the time to document this - much appreciated. Sav.
  • In NPM you can, but not within NCM. I don't really want to monitor the available flash with NPM every few minutes - it's only something I need to reference when planning upgrades or as a weekly exception type report. As a temp bypass, I did define the OID in NPM as a custom poller, applied it to all 900 odd routers that I…
  • We have a Windows 2003 Server with SQL 2005 that has the same issue. It's in our Test/Stress area so we had time to completely rebuild the server from new and install a clean copy of 9.5 (which worked). We then installed SP3 - which again broke the display of aspx pages with error 500. We did a web site repair into a new…
  • Can this be added as a field available to the Reports? Most of the other discovered properties are available, but not this field for some reason. Dave.
  • I wonder if Solarwinds would consider posting known issues and work-arounds - would save us both time (esp during busy times after a new release. Post note - I notice it was also covered here "> (should have looked harder last night). Sav. 
  • Can only help with the first question sorry Byron. Our polling engines seem to sit around 99.3 - 99.4. We seem to only lose polling data when it drops below 99. Your numbers would appear to be very good. I haven't seen our polling completion change since we did the 9.5.1 upgrade. We went from almost a weekly polling engine…
  • SQL Server supports many connection methods. I sometimes use Excel and/or Access to connect to the database and perform quick queries. For more complex reporting, I do nightly extracts to a separate database server so as to not load up the production server - this also might be a option for yourself if your access is…
  • We are using SQL 2008 R2 - no problem in that regard.
  • The Dec 2010 update that I provided is still working for us in 10.3. Ir displays all nodes with the map list associated with each (if the map index column is blank, then the node doesn't not appear on any map). Dave.
  • We use a number of NCM reports, including data from NCM to provide a reconciliation function to our asset management systems. I'm not sure if that's what you mean by 'manage network inventory' - but NCM certainly has a large amount of inventory information available. The major inventory related component missing for us is…
  • The concepts are much the same as in 2008. We migrated to newer versions of SQL Server and Report server (which now includes much better inbuilt charting options, tablix support etc) - but the its all much the same. In the Orion 10.5 Beta 2 SolarWinds looks to be including some web based reporting function - hopefully it…
  • Installed 2.1 this evening - I needed the ability to filter only on those nodes sending Netflow data. Problem is, how do I filter the Netflow sources resource if the page times out before it can be displayed? (the same issue occurs on the Netflow Admin page when it tries to display all the nodes). When you have a large…
  • How many elements are you polling and how big is your database BK? We have 4 pollers (about 20,000 elements) with a database size of approx 40Gig - running NPM, APM, Netflow (relatively small) and IPSLA. Our NCM implemenation is intergrated, but the database lives on another server. If you have enough server memory and…
  • Some time ago I created a process to populate a custom property for each node (using a DTS script). This custom property contained a list of maps on which the node appears. We are still using this for version 9.1, however I beleive that the new Network Atlas will make this non-functional when we upgrade. There is a copy in…
  • Raised case #112930. Support responded very quickly with the advice that there is currently no orphan record cleanup process. Suggested I raise a feature request (although I think that this is a design issue that should be treated as a bug). We literally have 100's of thousands of orphaned records and the database is on a…
  • Hello, We run a large number of 7206VXR-G1 routers (with ATM services) at 12.4(8a) without issues. Which sub-interfaces are you selecting? I don't believe that the ATM sub interfaces themselves will give you the traffic and interface error information. Make sure you are selecting the aal5 layer sub interfaces. Regards, Sav.