savell

A couple of examples for yesterday, last week and last month. select * from syslog where datediff(d, datetime, getdate()) = 1 select * from syslog where datediff(ww, datetime, getdate()) = 1 select * from syslog where datediff(m, datetime, getdate()) = 1 On our system a query for yesterday's syslog messages takes < 1…

I would just like to be able to select a Policy Group to display (i.e. from the web panel have a secondary selection for the Policy Group to display). For slightly more than the 'minimum bar' enable the selection of more than one Policy Group. Dave.

Case number is #82052. Dave.

Did this issue get fixed? I am seeing the same issue with a transform of a counter. It seems as if the transform is still operating on the raw value and not the calculated delta as I would expect. Dave.

70 or 80 seconds - wow, we aren't in the same ballpark. When I said ours were slower, I estimate that typically the node details screen took 2-3 seconds under 8.5.1, and 5-6 seconds under 9.0 (but I can't back the old numbers up with hard data). Any of the other screens (i.e. interfaces) seem to be unchanged in performance…

Thanks for that Michael. So once the file is stored on the temp drive (on the web server), does it load a new copy from this drive for each user that displays the same page? Maybe the I/O count we are seeing is merely driven by the number of web users? I might run an I/O file monitor again on one of our web servers to…

The problem I find with syslog messages, is that the message format changes between device types and software levels. If Cisco had a nice and organised message format (i.e. message ID followed by a documented message text) then syslog life would be good. We have 170 something different device types in our network (with a…

I'll like to restart this request thread - the ability to create an alert (on the Alert screen) from a syslog or snmp trap. Whilst I understand the stateful comment above - it's quite how I see it. For example, CUCM issues syslog messages when the call manager database sync process fails to sucessfully replicate.…

I'm in the same position - QOS everywhere, but not a current method to monitor the queue discards. Dave. 

Thanks Haley. In the meantime we have been forced to terminate use of the VoIP monitor to avoid impact on the base Orion system (i.e. this can't come soon enough for us). Dave. 

I have had my share of upgrade issues in the past, but going from 9.5.1 to 10 for me was easily the most painless yet. We have a single poller SLX implementation (with no additional modules), and another multiple poller SLX implementation (plus APM, NTA, IPSLA and additional web server). Each upgrade was clean and quick…

We just upgraded 3 pollers, VoIP module, Netflow Module, Application Monitor and the Additional web server with fairly minimal pain. Three problems only: 1. ICMP packet size (already solved in another thread) 2. VoIP web page errors on the additional web server (works OK on the primary - so I can live with it for a few…

This doesn't appear to have been addressed in NCM v6. Our PolicyCacheResults table is now almost 60 million rows - it can't handle the load captain! I'll get the support call re-opened. Dave.

Well, after a few hours looking at this, and then another couple working with support - I didn't get anywhere. Then I came across this http://support.solarwinds.com/kbase/ProblemDetail.cfm?ID=1118 At least APM is now working. Dave.

We have just purchased Cirrus - is it worth waiting for the beta of version 4.0 rather that starting with 3.51? Any idea how far away the beta will be?? Dave.

Thanks Haley! It's a reasonable time in Australia - which means that this must be a totally unreasonable time in the US - don't you sleep?? Dave.

I get a very similar issue each night. A small number of devices fail to backup the configs each night (config discarded type errors). They appear to be random devices with no pattern to the failures (900+ devices, with SSH and TACACS). When I try them manually the next morning, no issue. I will log a problem, but will…

So 9.1 came out in Oct 2008. Just how are you still running a trial version? Dave.

Does a Cisco MCS 7845I3 have any interest? (re-badged IBM 7947PAQ). Drop me an e-mail and I'll send directly if you would like it. Dave. Edit: I ran the snmpwalk tool and received an ASN.1 decoding error after 720 OID's. The Engineers MIB Walk tool works ok - is an export from this tool ok for you to work with?

Support is correct - they closed the case until we can schedule a re-install of Orion. Given we have a large installation (4000+) nodes, I have elected not to do this until 8.1 is available. We have comfirmed both the DEP settings and Admin rights as requested by Support - product re-install was the last option requested.…

Has this issue been progressed for you? We had the same issue today, the database disconnected due to a network problem for just a few minutes. When it reconnected it seemed that everything was working OK, and the last database sync field indicated that database updates were occurring from the poller. However, history data…

Gee, that's tough. The guy posts on Friday night on the public forum about an issue that he admits himself that he doesn't have the time to use the official support channel for. Doing an eval on a network management product takes some serious time investment (and still more time investment once the choice has been made).…

Only tested currently with 10.0 (we aren't scheduled for our 10.1.1 upgrade until after Christmas). Dave.

I used this SQL to populate a custom property that contains a comma delimited list of the custom pollers assigned to each node. This allows me to quickly scan a node list report (by MachineType), and easily identify if a custom poller has been missed off a device (I can't trust the team to apply common power and fan…

Well, this change worked OK for almost 90 minutes - then we started getting Server.CreateObject Failed 8007000e messages (error -2147024882) on the web pages. I will open a case.... Dave.

Hi Christine, That would help - I'm using the comments field as a pseudo custom property today (which is already part of the selection criteria). It means however that I still need to run external SQL routines to populate the comments (or custom properties) based upon data already available in NCM. It would be much better…

That's great news Chris - thanks for the update. Just for reference, we do use the default ${SysName} macro. Dave.

If I needed to extract a large number of Syslog records from Orion (or any other data for that matter), I would use a DTS/SSIS job to directly execute against the database. Dave.

Just to confirm that installing IP SLA Manager on the additional pollers did indeed sort out our issue. Dave.

OK - maybe I was too quick retrying. Left it for the day and retried - submission was successful. Dave.