Vfs cache: failed to reload item

Help and Support
1

rclone v1.55.1
ubuntu 20.04 lts server
google cloud storage

RCLONE_CACHE_DIR=/var/cache/rclone
RCLONE_VFS_CACHE_MODE=full
RCLONE_VFS_CACHE_MAX_AGE=175200h
RCLONE_VFS_MAX_SIZE=1024G
RCLONE_VFS_POLL_INTERVAL=24h
RCLONE_VFS_WRITE_BACK=12h
RCLONE_DIR_CACHE_TIME=12h
RCLONE_POLL_INTERVAL=6h
RCLONE_ATTR_TIMEOUT=1h

command:
rclone mount gcloud:test /gcloud --umask=0002 --allow-other --default-permissions

I have about 10gb and lot of files. All seems to works fine. But when I restart the service it take so long time and lot of erros (bellow). I have to remove the cache to get it started. And all data within the 12h (write back) are lost. Does I doing something wrong?

2021/05/19 22:41:54 ERROR : 2021/5/19/10/738C536E/02F95B06/32B746C4: vfs cache: failed to reload item: reload: failed to add virtual dir entry: file does not exist
2021/05/19 22:41:54 ERROR : 2021/5/19/10/738C536E/02F95B06/3449ADC9: vfs cache: failed to reload item: reload: failed to add virtual dir entry: file does not exist
2021/05/19 22:41:54 ERROR : 2021/5/19/10/738C536E/02F95B06/346C7D58: vfs cache: failed to reload item: reload: failed to add virtual dir entry: file does not exist
2021/05/19 22:41:54 ERROR : 2021/5/19/10/738C536E/02F95B06/347254D8: vfs cache: failed to reload item: reload: failed to add virtual dir entry: file does not exist

2

If you remove the cache, things break and won't work and generate errors since you removed the cache.

3

yes. I know.

But I would like to know why the first problem happens. It is the third time already. Only removing the cache to get it started again. Is it because is a lot of files waiting (12h) to be send do backend?

4

The answer is usually in the log file. If you make a log file, we can know for sure as I usually don't like to guess. If you have the issue again, just share the log file of the problem.

5

I have a log file and I print in the first message whats is printed in the log file.

lot and log of entries like that: (all day production)
2021/05/19 22:41:54 ERROR : 2021/5/19/10/738C536E/02F95B06/32B746C4: vfs cache: failed to reload item: reload: failed to add virtual dir entry: file does not exist

and too long time to start! too long that can't start again without removing the cache...

6

Those are because you deleted the cache so that's known.

That is the debug log we want so we can identify what's going on.

7

No, please read my e-mail carefully.

this ERROR is before delete the cache. I have to remove the cache to stop this error.

8

That error is because someone manually removed/changed/adjusted the cache area.

It's telling you the file does not exist that it expects there meaning someone/something/solar flare removed it.

9

nothing is manually removed/changed/adjusted the cache area.

thanks is something that only rclone does...

I sent the parameters used.... Does I doing something wrong?

10

That error is someone/something removed files from the cache so that's what the error you linked means.

I can't offer much more other than, you/someone/something deleted things and you can't do that or you get that error.

If you remove the cache, things break.

If you have a debug log of the initial issue, please share.

11

there are files that are removed from the mount. not in the cache.
is there something breaking? the parameters I'm using....

12

No, you are removing something from the cache area which is causing the error.

13

no. I'm not.

is there a way to debug it better?
I think is a rclone bug.

14

Yep as stated numerous times, share a debug log.

15

how the best way to create this debug log? and post here?

16

It's written in the help template.

A log from the command with the -vv flag
You should use 3 backticks to begin and end your paste to make it readable. Or use a service such as https://pastebin.com or https://gist.github.com/

Paste  log here
17

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.