Kevin
Well-known member
I'm working on a 2.1 B4 test copy of a site that was upgraded from 1.5.x and had to rebuild the attachment thumbnails. From within the ACP attachment browser it shows a count of 168K attachments. When the job runs in the browser (no CLI option) it ended up stopping & restarting a few times because it takes quite a long time to rebuild that many thumbnails and my laptop had to be restarted, et cetera during the run job.
By the time the job did finish though the on screen counter was closer to 200K, far above the 168K expected amount.
My first thought was orphaned attachments but the hourly job had run both on the source 1.5.x install and the test 2.1 install. Next thought that maybe it was doing XFMG attachments also but according to the ACP of the 168K attachments 155K were posts and the remaining 13K were from the gallery. Another thought was that perhaps physically there were more files on disk that were being looped through and not being reported by the ACP but /data/attachments, as calculated by Windows, also showed 168K. The xf_attachment table shows 168K entries.
So would the job being stopped & restarted affect the onscreen display count?
With the amount of time it took for the job to finish it's not an experiment I want to repeat too often.
By the time the job did finish though the on screen counter was closer to 200K, far above the 168K expected amount.
My first thought was orphaned attachments but the hourly job had run both on the source 1.5.x install and the test 2.1 install. Next thought that maybe it was doing XFMG attachments also but according to the ACP of the 168K attachments 155K were posts and the remaining 13K were from the gallery. Another thought was that perhaps physically there were more files on disk that were being looped through and not being reported by the ACP but /data/attachments, as calculated by Windows, also showed 168K. The xf_attachment table shows 168K entries.
So would the job being stopped & restarted affect the onscreen display count?
With the amount of time it took for the job to finish it's not an experiment I want to repeat too often.
