cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 9

NPM 10.7 Upgrade - Report & Application Issues

After upgrading to NPM 10.7 we are noticing quite a few issues in our environment and am wondering if anyone else is seeing the same and possibly has a fix. I have reached out to support, but time is critical for these issues so figured I would try here as well.

Scheduled Reports:-

We had scheduled reports set to email a PDF version of one of our views. This was working beautifully in 10.6, and since 10.7 all we receive is an error in the email stating:

Unable to retrieve url: https://ORIONSERVERNAME/Orion/SummaryView.aspx?ViewID=107 using PDF type.


I have tried removing the scheduled tasks associated with the original reports created using the old report scheduler, and recreating the reports from scratch using the new GUI based version. Same results, the error above. PDF export does in fact work if I navigate to the webpage listed above, and click export to PDF, no errors, comes out nice. Obviously I would still like for this to function as it once did without the manual intervention.


Application Status Unknown:-

Although SAM is not NPM, our application monitoring seems to have been affected from the update to NPM 10.7. Specifically nodes polled via our additional poller. The components which make up the application monitor all show an up / healthy status, yet the application itself shows 'unknown status', ONLY for nodes polled by our additional poller. The additional poller was also updated using the latest version available to us on the customer portal, which appears to have the same date as the NPM 10.7 installer, and the core versions show identical. If I change the poller for the node, wait a while, the application status goes from unknown to up / healthy to reflect that of the components themselves.


An example of what I mean:
BBerry.JPG


Another issue is with the link utilization displaying on maps. So far I have not been able to get this to show for our WAN interfaces, I thought it was as simple as enabling the feature within Network Atlas? I guess I must be missing something. In any case with the other two issues this one is currently the least of my concerns because features that were working fine are horribly broken at the moment.


Any help, ideas, or guidance would be greatly appreciated.

Labels (3)
47 Replies

I wish I would have seen this thread before we upgraded.

With our upgrade:

We are experiencing the "Unable to retrieve url using HTML type" issue.

I, however, am having no issue with PDF reports.

as well as 2 of our poller's licenses had to be reset

Case 613211

0 Kudos
Level 9

Anyone get a resolution from support yet?  None of the suggestions thus far has worked. We're not binding SSL for the website so the SSL SQL query doesn't apply for us.  Hotfixes 1, 3 and 4 are applied (not sure why there's no 2).  Tried the built-in Orion Admin account as the credential, have tried the domain service account and even my own user account... none work when scheduling the reports yet all of them work when logging on as those accounts and viewing the reports via URL.

0 Kudos

Since I opened the case last week, SW support did not even tried to reach me. Im in the same situation, stuck and its not working... Deceiving!

0 Kudos

Apologies sir, we'll ping support to see where the disconnect is.

0 Kudos

I spoke to Chris Klinedinst regarding another case, and he said he would escalate this issue as well. Thanks

0 Kudos
Level 13

I had the same issue, and was able to fix it partially....

The URL was in https, although it was specifying :80 at the end of the domain... I fixed it using rob.hock SQL query. That fixed the PDF files, and they are now generated, but I still have the same problem with the XLS files in the same email. 

I opened Case #608493 - "Unable to schedule XLS reports since 10.7 upgrade"

0 Kudos
Level 9

Just upgraded to 10.7 as well and having the exact same problem.  Many reports saying "Unable to retrieve url" despite the URL being valid and accessible under the context of the account used.  Theses scheduled reports were all created in the old Report Scheduler so I wonder if something got left out when the scheduler was converted to be web-based.  I haven't yet tried recreating these schedules from scratch in the web interface but I think that may be my next step.

0 Kudos

Tried recreating a few of the schedules from scratch, no dice.  I submitted support ticket 606829

0 Kudos
Level 7

I have just upgraded to 10.7 and we have the same issue with NPM. Everything else works great. Please update when there hotfix. Thanks

0 Kudos
Level 10

I too am seeing issues with reports that are no longer applied to schedules and alerts that are not firing at our secondary site.

I have indefinitely postponed the upgrade of our main production site until more of these problems shake out... This doesn't come at a good time for SW right now in my company as we've got a large renewal quote sitting on my VP's desk. Definitely some hesitation in signing a 5-figure PO.

0 Kudos

I am in the same boat.  I have a rather large quote for Storage Manager and a Fail over environment waiting for approval as well.

0 Kudos

Andrew,

We did have a DST bug that could cause similar behavior: NPM 10.7 / SAM 6.1 / NCM 7.2.X Daylight Savings Time Bug

HF would need to be applied on Main and Additional pollers.

0 Kudos
Level 9

Any update on resolving this issue without having to completely re-install SolarWinds application?

0 Kudos

Likely it's an SSL binding issue. Check the SSL bindings in IIS. Also check SSLEnabled field in DB:

Select * from Websites

Here you should see a list of websites associated with your Orion instance. There is a column labeled SSLEnabled which should be set to 1 on any row associated with an SSL Enabled web site. If this is set to 0 try the following:

(Replace ID_OF_WEBSITE with the id number from the WebsiteID column of the site you wish to change)

Update Websites Set SSLEnabled=1 where WebsiteID=ID_OF_WEBSITE

In my case I also had to update the ServerName to the FQDN.  Update Websites Set ServerName= servername.MY.DOMAIN

0 Kudos

Enabling SSL "broke" my reports and this was extremely helpful.

In my case I also had to update the ServerName to the FQDN.  Update Websites Set ServerName= servername.MY.DOMAIN

PDF reports are now successful again,

Yep, that fixed it for me. Thanks rob.hock!

However our situation was different. We do enforce SSL but the port in the database was set for 6991 for some reason. After running the following query in our database to change the port to 443, all scheduled reports and PDF exports work fine.

Update Websites Set Port=443 where WebsiteID=1

(Special note: My organization only has a single website. If you have multiple Orion sites make sure you choose the correct ID.)

This fixed my issue after I enabled SSL on the Orion Web Console, reports that I scheduled would not generate, instead I received this error.

Unable to retrieve url: http://VCAANM01:80/Orion/Report.aspx?ReportID=51 using PDF type.Unable to retrieve url: http://VCAANM01:80/Orion/Report.aspx?ReportID=138 using PDF type.

I updated the details for the Orion server in the SQL table Websites and changed SSLEnabled to 1, and changed the Port from 80 to 443.

All working again!  Cheers.

0 Kudos
Product Manager
Product Manager

As it relates to your SAM Application status issue, have you tried upgrading to SAM 6.1 to see if that resolves the issue?

Application Status Unknown:-

Although SAM is not NPM, our application monitoring seems to have been affected from the update to NPM 10.7. Specifically nodes polled via our additional poller. The components which make up the application monitor all show an up / healthy status, yet the application itself shows 'unknown status', ONLY for nodes polled by our additional poller. The additional poller was also updated using the latest version available to us on the customer portal, which appears to have the same date as the NPM 10.7 installer, and the core versions show identical. If I change the poller for the node, wait a while, the application status goes from unknown to up / healthy to reflect that of the components themselves.


An example of what I mean:
BBerry.JPG

Apart from doing a complete rebuild, is there any official SW advice on how to recover this functionality?
Hopefully,
Robert

0 Kudos

Report issues like this - one trouble-shooting hint that may help.

Log into your core server as the account who will be running the report.   Likely some service account?

Paste the URL into IE browser (and yes...use IE, not firefox or chrome.  The Report function uses IE) and see how it opens up there.

Have been able to find quite a few issues that way, and clear them up.  Then the scheduled report will work out properly.   Big thing, make sure you are getting the type of login you expect.   Windows Auth vs. the form login page.

Hope this helps!

0 Kudos