Thanks for the input guys. Your suggestions make sense. I'll try removing the seek and see how that works with all of the geotools/geoserver unit tests and also do some smoke testing. I'll file a bug in jira with all of the details as well.
On Thu, Oct 31, 2019 at 3:55 AM Daniele Romagnoli < daniele.romagn...@geo-solutions.it> wrote: > Hi Justin, > I tracked a bit the git history and it looks like that this seek(0) is > there since 2012 (that is where the git history ends more or less). > Not sure if that seek(0) has been put there for a real reason (since JAI > uses deferred execution and reader can start the actual read only once > need) or as a some kind of precaution. > I think that you could try removing that line and do some GeoServer tests > using ImageMosaic with configured "use JAI imageread" and see what happens > under load. > > Please, let us know. > Regards, > Daniele > > > > > > > On Thu, Oct 31, 2019 at 10:05 AM Andrea Aime <andrea.a...@geo-solutions.it> > wrote: > >> Hi Justin, >> before we add a flag to work around it, I was wondering if the issue >> could be fixed in a different way, like, at the source. >> >> Looking at the FileImageInputStreamImpl, I've noticed one oddball thing, >> but it's probably not related: the lines do not seem matching up. >> >> https://github.com/geosolutions-it/imageio-ext/blob/master/library/streams/src/main/java/it/geosolutions/imageio/stream/input/FileImageInputStreamExtImpl.java#L285 >> >> Anyways, the error in question is a few lines above... seems someone is >> reading the header (to get the color model as you say), the PNGImageReader >> calls flushBefore updating the flushedPos, and then the seek(0) in >> ReadType.JAI_IMAGEREAD fails. Hum... I'm wondering if the seek(0) really >> necessary, given that the reader has been provided as parameter? (e.g., is >> it the same read on which we called to get the color model?). Another >> option could be to temporarily pass a wrapping input stream killing the >> flushbefore, just for the sake of reading the header... >> >> Daniele's input much appreciated too :-D >> >> Cheers >> Andrea >> >> >> On Wed, Oct 30, 2019 at 9:43 PM Justin Deoliveira <jdeol...@gmail.com> >> wrote: >> >>> Hi folks, >>> >>> I am working on upgrading an application from a much older version of >>> geotools (15.1) to the latest stable. I've hit a roadblock regarding image >>> mosaics consisting of indexed PNG files that I think I have narrowed down >>> to this change: >>> >>> >>> https://github.com/geotools/geotools/commit/585b021d15df7d132bd6963d7fb78338e814b625 >>> >>> >>> What appears to be happening is that the call to getRawColorModel is >>> advancing the underlying image stream. Later on in the process that same >>> stream stream is "seeked" back to the zero position in preparation for the >>> actual read operation, and that throws an exception: >>> >>> java.lang.IllegalArgumentException: pos < flushedPos! >>> at >>> it.geosolutions.imageio.stream.input.FileImageInputStreamExtImpl.seek(FileImageInputStreamExtImpl.java:285) >>> at >>> org.geotools.coverage.grid.io.imageio.ReadType$2.read(ReadType.java:131) >>> at >>> org.geotools.gce.imagemosaic.GranuleDescriptor.loadRaster(GranuleDescriptor.java:928) >>> >>> >>> I am guessing it's perhaps some limitation with the PNG format (which I >>> know is about the worst possible format to be rendering imagery from 🤔 ). >>> >>> So I guess my first question is would this be considered a bug? Or more >>> or a limitation with this particular file format? I am guessing a bug since >>> prior to this change this scenario works. >>> >>> Assuming that bug is the consensus I'd be happy to attempt a patch but >>> probably need some guidance on how to proceed. My initial thought for a >>> "quick fix" was that perhaps we could add a flag to the mosaic >>> configuration or a read parameter that specified explicitly that the images >>> in the mosaic have an IndexColorModel so that we can forgo the check and >>> hence not pre-advance the image stream? >>> >>> Thoughts appreciated. >>> >>> Thanks! >>> >>> -Justin >>> >>> >>> _______________________________________________ >>> GeoTools-Devel mailing list >>> GeoTools-Devel@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/geotools-devel >>> >> >> >> -- >> >> Regards, Andrea Aime == GeoServer Professional Services from the experts! >> Visit http://goo.gl/it488V for more information. == Ing. Andrea Aime >> @geowolf Technical Lead GeoSolutions S.A.S. Via di Montramito 3/A 55054 >> Massarosa (LU) phone: +39 0584 962313 fax: +39 0584 1660272 mob: +39 339 >> 8844549 http://www.geo-solutions.it http://twitter.com/geosolutions_it >> ------------------------------------------------------- *Con riferimento >> alla normativa sul trattamento dei dati personali (Reg. UE 2016/679 - >> Regolamento generale sulla protezione dei dati “GDPR”), si precisa che ogni >> circostanza inerente alla presente email (il suo contenuto, gli eventuali >> allegati, etc.) è un dato la cui conoscenza è riservata al/i solo/i >> destinatario/i indicati dallo scrivente. Se il messaggio Le è giunto per >> errore, è tenuta/o a cancellarlo, ogni altra operazione è illecita. Le >> sarei comunque grato se potesse darmene notizia. This email is intended >> only for the person or entity to which it is addressed and may contain >> information that is privileged, confidential or otherwise protected from >> disclosure. We remind that - as provided by European Regulation 2016/679 >> “GDPR” - copying, dissemination or use of this e-mail or the information >> herein by anyone other than the intended recipient is prohibited. If you >> have received this email by mistake, please notify us immediately by >> telephone or e-mail.* >> _______________________________________________ >> GeoTools-Devel mailing list >> GeoTools-Devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/geotools-devel >> > > > -- > Regards, > Daniele Romagnoli > == > GeoServer Professional Services from the experts! Visit > http://goo.gl/it488V for more information. > == > > Ing. Daniele Romagnoli > Senior Software Engineer > > GeoSolutions S.A.S. > Via di Montramito 3/A > 55054 Massarosa (LU) > Italy > phone: +39 0584 962313 > fax: +39 0584 1660272 > > http://www.geo-solutions.it > http://twitter.com/geosolutions_it > > ------------------------------------------------------- > > Con riferimento alla normativa sul trattamento dei dati personali (Reg. UE > 2016/679 - Regolamento generale sulla protezione dei dati “GDPR”), si > precisa che ogni circostanza inerente alla presente email (il suo > contenuto, gli eventuali allegati, etc.) è un dato la cui conoscenza è > riservata al/i solo/i destinatario/i indicati dallo scrivente. Se il > messaggio Le è giunto per errore, è tenuta/o a cancellarlo, ogni altra > operazione è illecita. Le sarei comunque grato se potesse darmene notizia. > > This email is intended only for the person or entity to which it is > addressed and may contain information that is privileged, confidential or > otherwise protected from disclosure. We remind that - as provided by > European Regulation 2016/679 “GDPR” - copying, dissemination or use of this > e-mail or the information herein by anyone other than the intended > recipient is prohibited. If you have received this email by mistake, please > notify us immediately by telephone or e-mail. >
_______________________________________________ GeoTools-Devel mailing list GeoTools-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
