Todd, thanks for the reply.
Perhaps I'm not explaining this well... I have an internal, M.2 SSD on which the O/S and programs reside. It works fine. It's never had a problem.
I have a second M.2 SSD on which a variety of other things reside, including the LR catalogs, cache, temp and a lot of other things. It works fine. It's never had a problem.
I have a Samsung 860, 4TB SSD, presently connected to the system via USB-powered carrier via USB-C cable. It contains image files. This is the same device that normally lives in my desktop computer, plugged into a removable carrier to provide images. It has been in use for months, worked fine. It never had a problem.
The “Image” catalog image files contents exist in one folder tree. The incoming images exist in a completely separate folder tree, totally unconnected to anything in the Image catalog.
For the last couple week, and most likely for the next month, my desktop is inaccessible and I'm working from the Dell 7740 laptop. To provide images for the laptop when traveling or situations where I have to use the it, I shut down the desktop, extract the 4TB SSD, and put it in the external carrier. I've done this for over a year with absolutely no issues with either system.
On the laptop there are NO spinning platters, 2 internal SSDs and 1 external.
My Lightroom workflow has been in use for years, and it's unlikely to change now. You may believe it to be unnecessary, but it works for me. As the import into the Image catalog is exactly the same as any import that copies the files from folder A to folder B and adds them to the catalog, I can see no reason why it should be any more prone to corruption, other than caused by Lightroom, than any other flow. Other than the separate catalogs existing on the same drive, there is NO connection between the short term catalog and the Image catalog.
I CAN state with certainty that NO images (ZERO) images have been imported into the Image catalog (the primary, permanent catalog) since March, 2020. All NEW images that have been worked on have been in the short term catalog and exist in a separate folder tree. The ONLY processing done in the Image catalog has been on images already present, processed for output of various kinds.
All I can do here is report what I've seen (hopefully, mostly accurately):
- Date/time 1: Everything fine, no problem. Both catalogs functional.
- Date/time 2: LR short term catalog displays an error that it can't read from the catalog, has to shut down. Restart and it says it has to repair. Fails. Catalog is corrupt. New catalog created. Short term is not backed up since it is used only for culling images - all results are stored in the images with the metadata. Images culled are deleted from disk, leaving ONLY those images that will be imported into the Image catalog.
- Date/time 3: Next day, LR displays an error in the Image catalog that it can't read from the catalog, has to shut down. Restart and it says it has to repair. Fails. Catalog is corrupt. New catalog created from backup.
- Same day: Opened Image catalog. Random images in the Image catalog indicate they cannot be opened. No obvious pattern. Blocks of 1 or more bad images interspersed with blocks of images that are fine.
- During examination of catalog contents, Image catalog displays error that it can’t read from cache (if I recall correctly) and starts a loop where at restart it has to fix, fails, has to restart to fix, repeat.
- So far, my examination of a few of the corrupted images in the week-old backup indicates they are fine, and that only images on the SSD, currently accessed from Lightroom are corrupt.
- Date/time 4: Verified that for incoming images, only folders images that were in the short term catalog when it displayed the error have any corrupted images. Folders containing thousands of images that have not been imported at any point have no affected image files.
I've run chkdsk on the 4TB SSD with no problems found. I’ve run HDDSCAN on the 4TB, SSD. I had it do the Butterfly Read and I had it do the VR-Verify tests. After churning all night, the results of both tests appear to be ZERO "Bads", which I presume means there were no bad areas found? I have little knowledge about HDDScan, so if there is a different test I should be running in HDDScan let me know. I don't see a way to attach the reports, but if someone can tell me how to do so simiply, I can do that, though there doesn’t appear to be much of interest in them.
Beyond that, if there is a piece of software that can “FIX” the corrupted image files, great, I’ll give it a try.
At this point, there appears to be no horde of screaming owners finding corrupted images, so I’ll presume your assertion that it can’t be Lightroom and must be either my workflow or hardware, is correct. I’m planning to format the 4TB SSD, reload the contents from a backup and hope the problem doesn’t occur again. I’ll also contact Samsung and likely have them replace the SSD, which will cause some inconvenience as they do NOT do cross-ship, resulting in a much longer period without the device.
At this point, I’m tossing this whole thing on the “screw it” pile, which is where most of the bizarre things I’ve seen with Lightroom have ended up over the years.
13
Replies
AdChoices