Discussion Errors related to --vfs-cache-mode >= writes in /var/plexguide/logs/rclone-tcrypt.log

[hooper]

I am seeing this error in /var/plexguide/logs/rclone-tcrypt-log for every file on my tdrive.

2019/04/12 06:49:17 ERROR : tv/The Walking Dead/Season 03/The Walking Dead - S03E07 - When the Dead Come Knocking.mkv: WriteFileHandle: Can't open for write without O_TRUNC on existing file without --vfs-cache-mode >= writes

I think this may be due to bazarr, but wondering if we should be setting --vfs-cache-mode to writes anyway.

 

[MrDoob]

In the mount options is it included since v8 or before

 

[hooper]

Here are my active rclone mount commands:

ps -ef | grep rclone

root      6664     1  0 Apr08 ?        00:26:11 

/usr/bin/rclone mount gdrive: /mnt/gdrive --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-gdrive.log --log-level ERROR --timeout 1h --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

root      6668     1  8 Apr08 ?        07:50:57

/usr/bin/rclone mount tcrypt: /mnt/tcrypt --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-tcrypt.log --log-level ERROR --timeout 1h --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

root      6693     1  0 Apr08 ?        00:28:41

/usr/bin/rclone mount tdrive: /mnt/tdrive --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-tdrive.log --log-level ERROR --timeout 1h --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

root      6704     1  1 Apr08 ?        01:24:05

/usr/bin/rclone mount gcrypt: /mnt/gcrypt --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-gcrypt.log --log-level ERROR --timeout 1h --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

plexgui+ 23770  6793  0 16:03 pts/0    00:00:00 grep --color=auto rclone

There is no --vfs-cache-mode setting. I have made some changes to the default values for some settings, but there is no option on --vfs-cache-mode.

 

[hooper]

looks like this was fixed on Feb 3

https://github.com/PGBlitz/PGClone/commit/c0b75813603ef4b9f4f4b4ae7de53e47bbad309c

I think I need to re-deploy my rclone mounts.

 

[hooper]

looks like this was fixed on Feb 3

https://github.com/PGBlitz/PGClone/commit/c0b75813603ef4b9f4f4b4ae7de53e47bbad309c

I think I need to re-deploy my rclone mounts.

I re-deployed the mounts and still don't get --vfs-cache-mode set properly for tcrypt or gcrypt.

 

[hooper]

ok, I fixed this. The code changed I referenced above was re-factored in a later PG release.

Basically, the file shown below needs to be edited and changed to crypt.service instead of drive.service. This will load the crypt-related rclone defaults instead of the drive-related defaults from the template.



and here are my rclone commands that are running on the system now.

ps -ef | grep rclone

root       359     1  0 16:40 ?        00:00:00 

/usr/bin/rclone mount gdrive: /mnt/gdrive --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-gdrive.log --log-level ERROR --timeout 1h --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

root       537     1  1 16:40 ?        00:00:01 

/usr/bin/rclone mount tdrive: /mnt/tdrive --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-tdrive.log --log-level ERROR --timeout 1h --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

root       845     1  1 16:40 ?        00:00:00 

/usr/bin/rclone mount gcrypt: /mnt/gcrypt --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-gcrypt.log --log-level ERROR --timeout 1h --vfs-cache-mode writes --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

root      1022     1  3 16:40 ?        00:00:02 

/usr/bin/rclone mount tcrypt: /mnt/tcrypt --config /opt/appdata/plexguide/rclone.conf --allow-non-empty --allow-other --size-only --uid=1000 --gid=1000 --umask 002 --log-file=/var/plexguide/logs/rclone-tcrypt.log --log-level ERROR --timeout 1h --vfs-cache-mode writes --dir-cache-time 60m --vfs-cache-max-age 72h --vfs-read-chunk-size 64M --vfs-read-chunk-size-limit 5G --drive-chunk-size 32M --buffer-size 256M

This change has NOT been pushed to the repo. So someone with access to do so should probably do that.

 

[Admin9705]

Book marked for follow up!

 

[MrDoob]

ok, I fixed this.

Test this one some days .. if all okay ..
Create an PullRequest.

For the changes .

But if you have a good server with great specs , you can use greater values as user with lower specs.
So be careful

 

[MrDoob]

Book marked for follow up!

Here the same ^^

 

[mythbuster]

I am having the same issue - which file do I need to change? I am currently on 8.6.5 using blitz encrypted and have re-deployed mounts several times to no avail.

 

[hooper]

I am having the same issue - which file do I need to change? I am currently on 8.6.5 using blitz encrypted and have re-deployed mounts several times to no avail.

do you use bazarr for grabbing subtitles? This is what was causing the issue for me.

to fix this, turn select option 4 to deploy PG Clone, then select option 1 to use PGBlitz PG Clone, then select [C] for options, then select [2] for Rclone mount VFS Settings. In this section select option [6] and change the mode to [3] for writes. After this change, go ahead and redeploy your PG Clone mounts and the error should go away.

 

[mythbuster]

do you use bazarr for grabbing subtitles? This is what was causing the issue for me.

to fix this, turn select option 4 to deploy PG Clone, then select option 1 to use PGBlitz PG Clone, then select [C] for options, then select [2] for Rclone mount VFS Settings. In this section select option [6] and change the mode to [3] for writes. After this change, go ahead and redeploy your PG Clone mounts and the error should go away.

That's bizarre, I am not using bazarr. Not sure what caused the issue. But that did the trick, thank you!