| 1 |
On Tue, 24 Dec 2019 12:35:41 -0600, Dale wrote: |
| 2 |
|
| 3 |
> >> My temporary solution, I pointed it to a small directory that only |
| 4 |
> >> has a couple dozen images in it. That seems to work. |
| 5 |
|
| 6 |
> > Is there a difference in the metadata of these few images compared |
| 7 |
> > with the rest in the whole directory? |
| 8 |
> |
| 9 |
> It's actually a small directory of the exact same images. I might add, |
| 10 |
> the clock does skip a second or so when I add them so I suspect it |
| 11 |
> performs the same action, it just has a much smaller number of them. |
| 12 |
> It's hard to say. Point being, I'm using the same images as before, |
| 13 |
> just a smaller sub-directory. |
| 14 |
|
| 15 |
Have you tried using the smaller directory and then addng a few (hundred) |
| 16 |
images at a time, letting the scan finish and then repeating. |
| 17 |
|
| 18 |
I wonder is plasma is using a new method of caching the data, or the old |
| 19 |
cache got corrupted, so it has to reindex the whole directory again. It's |
| 20 |
never had to do 150000+ files in one go before because the directory has |
| 21 |
grown organically. Now, for whatever reason, it seems like it is trying to |
| 22 |
do the whole lot in one hit. |
| 23 |
|
| 24 |
|
| 25 |
-- |
| 26 |
Neil Bothwick |
| 27 |
|
| 28 |
I'd tell you a UDP joke, but you may not get it. |
Archives