FrankvH
Intrepid explorer

NDI inventory files not complete

Hello,

The last weeks I keep a close look at the ndi files that ends up in the badlog. Every day we see that ndi files are not complete.

They end something like:

<Record MACHINE_ID="BAGA3031.prprivmgmt.intraxa" DB_NAME="FE3" TIMESTAMP="2019-06-30_19:54:29" HOST_NAME="BAGA3031.prprivmgmt.intraxa" INSTANCE_NAME="FE3" OPTION_NAME="DBA_FEATURE_USAGE_STATISTICS" OPTION_QUERY="10g" COL010="Transportable Tablespace" COL020="12.1.0.2.0" COL030="0" COL040="30" COL050="FALSE" COL080="2019-06-29_02:07:31" COL090="604800" COL100="1919367932"/>
<Record MACHINE_

So it is clear the XML is not complete and cannot be imported. What can be root cause of this? Everyday we have randomly around 100 of these.

Regards

Frank

 

0 Kudos
4 Replies
ChrisG
Community Manager Community Manager
Community Manager

Re: NDI inventory files not complete

I have always assumed that truncated files like this are most likely because an upload was interrupted while in progress for some reason - it is hard to catch such an interrupt in progress or reproduce it, but across large estates with many thousands of devices it is inevitable that this will occur from time to time.

On the assumption that it isn't affecting files from the same computer(s) all the time, it would not normally be anything to be particular concerned about: the system is build to be robust against this by regularly gathering fresh inventory so that such failures won't have a permanent impact.

(Anything expressed here is my own view and not necessarily that of my employer, Flexera. If my reply answers a question you have raised, please click "ACCEPT AS SOLUTION".)
FrankvH
Intrepid explorer

Re: NDI inventory files not complete

Hi Chris,

Thanks for your reply. You are right the number of ndi with this issue is relative small in compare with the amount of data. Nevertheless it is always good to see if we can improve and understand where this failure are coming from.

If nobody has another idea we just keep an eye on this.

Regards

Frank

0 Kudos
mfranz
Shining star

Re: NDI inventory files not complete

Hi Frank,

If you still would like to try and follow up on this issue, here are some ideas where to attack:

  • Are multiple Beacons involved and can the problematic files be tracked to one specific machine?
  • Is there a size limit, like are these capped files larger than others?
  • Were the ndi's compressed (.gz)?
  • Does IIS throw errors? Maybe activate additional logging?
  • Do you have replication active? Do replicated files have the same error pattern?

Best regards,

Markward

Softline Group is Europe's leading independent expert in Software Asset Management.
0 Kudos
FrankvH
Intrepid explorer

Re: NDI inventory files not complete

Hi Markward,

If we can improve that would be always good. To answer your questions:

  • Are multiple Beacons involved and can the problematic files be tracked to one specific machine? We have 10+ beacons and so far I can see the files are random not for one machine.
  • Is there a size limit, like are these capped files larger than others? Will need to investigate this
  • Were the ndi's compressed (.gz)? no just ndi few where .gz
  • Does IIS throw errors? Maybe activate additional logging? Will need to investigate this
  • Do you have replication active? Do replicated files have the same error pattern? No but would be good to test this.

Regards

Frank