Hi Rich,
With a one-off problem, it's hard to make any sort of diagnosis.
"Theoetically," if it worked yesterday, and nothing has changed, it should work
today...
Some places that I might start to look involve answering questions like:
Were the 2 new files properly created and copied (i.e., not somehow corrupt)?
Is there a directory with a bajillion files in it, and these last two just
pushed us over the edge?
What is the .jdb file from Nov 5?
Also, I recognize the value in using a "tried and true" system, but I have to
point out that the TDS is up to version 4.3.15, and if an actual bug exists, we
won't be able to back-port a fix to 4.2.10.
Cheers,
Lansing
Unidata
Boulder, Colorado
> Unidata folks (and community, in case others have seen issues like this),
>
> We have an ocean forecast model running that produces two files every
> day at about 4:25 am.
>
> We have a TDS version 4.2.10, with a FeatureCollection FMRC to pick up
> these files, and we use the <update> tag to have the collection
> scanned every day at 4:35.
>
> This morning, our new forecast files were not picked up in the
> collection, which of course breaks the downstream workflow that makes
> the forecast products.
>
> When I look in the Feature Collection log:
> <tomcat>/thredds/config/logs/featureCollectionScan.log
>
> I see:
>
> [2013-02-18T04:34:00.087-0500] INFO
> thredds.inventory.DatasetCollectionManager: coawst_4_use: rescan at
> Mon Feb 18 04:34:00 EST 2013: nnew = 2 ndelete = 0
>
> which indicates that the files were identified, but when I look at the
> dates in the collection cache directory
> <tomcat>/content/thredds/cache/collection
> I can see that the collection cache database was not updated today
> (the last modified date on the database is Feb 17 (yesterday, not
> today).
>
> rsignell@gam:/usr/local/usgs/tomcat-thredds/content/thredds/cache/collection$
> ls -sailrt
>
> 1177860 176 drwxrwxr-x 6 usgs usgs 176128 Jul 20 2012 ..
> 1180923 0 -rw-rw-r-- 1 usgs usgs 0 Jul 20 2012 je.lck
> 1180922 0 -rw-rw-r-- 1 usgs usgs 0 Aug 14 2012 je.info.0
> 1184489 0 -rw-rw-r-- 1 usgs usgs 0 Aug 14 2012 je.info.0.1
> 1180924 9772 -rw-rw-r-- 1 usgs usgs 9999826 Nov 5 11:30 00000000.jdb
> 1177416 0 -rw-r--r-- 1 usgs usgs 0 Feb 17 10:35 je.info.0.lck
> 1219031 4 drwxrwxr-x 2 usgs usgs 4096 Feb 17 10:35 .
> 1177686 2692 -rw-rw-r-- 1 usgs usgs 2752062 Feb 17 10:35 00000001.jdb
>
> What could be the possible reason that the database was not updated
> with the new information?
> Any place I could look for errors?
>
> Thanks,
> Rich
> --
> Dr. Richard P. Signell (508) 457-2229
> USGS, 384 Woods Hole Rd.
> Woods Hole, MA 02543-1598
>
>
Ticket Details
===================
Ticket ID: YQB-491770
Department: Support THREDDS
Priority: Normal
Status: Open