[ome-users] Question on "omero admin removepyramids"

Sascha Neinert sascha.neinert at uni-koeln.de
Fri Apr 20 13:01:35 BST 2018


Dear Mark,

to simplify i look at one image only that i have imported twice into the same server instance:

1) ID as shown in OMERO.web = 823 and import date = 2018-02-21 11:54:19, there is one file …/Pixels/823_pyramid of size 32 MB, time stamp is Feb 21 12:55
2) ID as shown in OMERO.web = 958, there is also one file …/Pixels/958_pyramid but of size 16 MB

In both cases the table pixels has the same value for ‚id‘ and for ‚image‘, so i guess that the pyramid files i found are the corresponding ones. The script removepyramids has the output 

Using session ... Idle timeout: 10 min. Current group: temporary
No more than 500 pyramids will be removed
Would remove pyramid for image 963
Would remove pyramid for image 953
Would remove pyramid for image 959
Would remove pyramid for image 958
… and so on, there’s no line for 823 (there are only 10 lines total).

Table ‚dbpatch‘ shows a line with currentversion = OMERO5.4, message = Database updated, finished = 2017-10-16 10:39:35. So both times the image was imported using OMERO 5.4 - the first time using 5.4.0 and the next time using OMERO 5.4.5.

I do not really know how to interpret this behaviour - it could be harmless and/or working as intended, it just seemed somehow unusual to me. We did do some tests with omero.pixeldata.max_plane_heigth and omero.pixeldata.max_plane_heigth, both are currently set to 4097. Maybe that is the reason for my confusion?

In case the image file is helpful, i have it available here: https://uni-koeln.sciebo.de/s/RxqVWPlktOPGk4p <https://uni-koeln.sciebo.de/s/RxqVWPlktOPGk4p> (PIN: 7428).

Best regards, Sascha


--
Sascha Neinert
CECAD RRZK IT Service
Regional Computing Center (RRZK)
University of Cologne
Phone: +49-221 478-84051 / +49-221 470-89624
Mail: sascha.neinert at uni-koeln.de





> Am 05.04.2018 um 11:02 schrieb Mark Carroll <m.t.b.carroll at dundee.ac.uk>:
> 
> Dear Sascha,
> 
>> after upgrading to 5.4.5 i tried bin/omero admin removepyramids --endian=little --dry-run
>> It has found ten pyramids to remove on my test server, and without --dry-run they have successfully been removed. But i have the same 10 images imported before using OMERO version 5.4.0 - those have not been touched - and then again with 5.4.5 - those are the 10 that have had their pyramids removed. Does the command have some implicit date or version restriction or filter, or am i just missing something? All are 4096 x 4096 and uint16.
> 
> That's interesting, it's not obvious to me what you're missing, there
> isn't a default date filter. Can you reconstruct from file times and the
> dbpatch database table or suchlike under which version of OMERO the
> images were originally imported? Pre-5.4 would need its pyramids so
> those would be safe.
> 
> Pyramids appear in the binary repository's Pixels/ directory as
> 12345_pyramid where the the 12345 corresponds to the ID of the Pixels
> object which may be different from the Image ID, e.g.,
> 
> bin/omero hql --all 'SELECT id FROM Pixels WHERE image.id = 4567'
> 
> to find the Pixels ID of Image 4567. Does looking for pyramids in that
> server directory yield any clues?
> 
> -- Mark
> 
> The University of Dundee is a registered Scottish Charity, No: SC015096
> _______________________________________________
> ome-users mailing list
> ome-users at lists.openmicroscopy.org.uk
> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20180420/b648306d/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5207 bytes
Desc: not available
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20180420/b648306d/attachment.p7s>


More information about the ome-users mailing list