0006752: [WARNING]: EOC marker not found. Codestream is corrupted.

View Revisions: Issue #6752		[ All Revisions ] [ Back to Issue ]
Summary	0006752: [WARNING]: EOC marker not found. Codestream is corrupted.

Revision	2013-08-30 05:46 by aiaustin
Description	I know this has arisen before, and I have looked at the various forums where it is suggested that this is a possible texture corruption. But I see it on a relatively new clean setup with a new MySQL data base on which I loaded OARs and IARs to reestablish the grid. So something must be making its way through that process? I am trying to pin down where the error could be... and by carefully bringing a grid back up in parts after an upgrade to r/23560 (29-Aug-2013) I note: a) I just get this message ONCE on FIRST RUN after an upgrade for one specific region that I have in my grid. b) There is no corresponding entry in OpenSim.log... just the line "[WARNING]: EOC marker not found. Codestream is corrupted." reported on the console. Anyone got a suggestion as to how to track such a problem down? I wonder if the location in the code where the error is printed out could report the object name/UUID/region/x/y/z involved perhaps?

Revision	2013-08-30 05:42 by aiaustin
Description	I know this has arisen before, and I have looked at the various forums where it is suggested that this is a possible texture corruption. But I see it on a quite new clean setup with a new MySQL data base ion which i loaded OARs and IARs to resestablish the grid. So somethign must be makign its way through that process? I am trying to pin down where the error could be... and by carefully bringing a grid abck up in parts after an upgrade to r/23560 (29-Aug-2013) note a) I just get this message ONCE on FIRST RUN after an upgrade for one specific region that I have in my grid. b) There is no corresponding entry in OpenSim.log... just the line "[WARNING]: EOC marker not found. Codestream is corrupted." reported on the console. Anyone got a suggestion as to how to track such a problem down? I wonder if the location in the code where the error is printed out could report the object name/UUID/region/x/y/z involved perhaps?