extern/rclone
1
0
Fork 0
mirror of https://github.com/rclone/rclone.git synced 2025-02-04 12:39:35 +01:00
Code Issues Packages Projects Releases Wiki Activity

mount, cmount: set --attr-timeout default to 1s - fixes #2157

Browse Source 
This  works around these 3 problems:

  * rclone using too much memory #2157
  * rclone not serving files to samba
    * https://forum.rclone.org/t/rclone-1-39-vs-1-40-mount-issue/5112
  * excessive time listing directories #2095
This commit is contained in:
Nick Craig-Wood 2018-03-23 22:42:51 +00:00
parent 7e80e609e8
commit 98a924602f
1 changed files with 24 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
cmd/mountlib/mount.go
View File

@ -28,7 +28,7 @@ var (
MaxReadAhead fs.SizeSuffix = 128 * 1024 MaxReadAhead fs.SizeSuffix = 128 * 1024
ExtraOptions []string ExtraOptions []string
ExtraFlags []string ExtraFlags []string
AttrTimeout = 0 * time.Second // how long the kernel caches attribute for AttrTimeout = 1 * time.Second // how long the kernel caches attribute for
) )
// Check is folder is empty // Check is folder is empty
@ -150,12 +150,30 @@ for solutions to make ` + commandName + ` mount more reliable.
You can use the flag --attr-timeout to set the time the kernel caches You can use the flag --attr-timeout to set the time the kernel caches
the attributes (size, modification time etc) for directory entries. the attributes (size, modification time etc) for directory entries.
The default is 0s - no caching - which is recommended for filesystems The default is "1s" which caches files just long enough to avoid
which can change outside the control of the kernel. too many callbacks to rclone from the kernel.
If you set it higher ('1s' or '1m' say) then the kernel will call back In theory 0s should be the correct value for filesystems which can
to rclone less often making it more efficient, however there may be change outside the control of the kernel. However this causes quite a
strange effects when files change on the remote. few problems such as
[rclone using too much memory](https://github.com/ncw/rclone/issues/2157),
[rclone not serving files to samba](https://forum.rclone.org/t/rclone-1-39-vs-1-40-mount-issue/5112)
and [excessive time listing directories](https://github.com/ncw/rclone/issues/2095#issuecomment-371141147).
The kernel can cache the info about a file for the time given by
"--attr-timeout". You may see corruption if the remote file changes
length during this window. It will show up as either a truncated file
or a file with garbage on the end. With "--attr-timeout 1s" this is
very unlikely but not impossible. The higher you set "--attr-timeout"
the more likely it is. The default setting of "1s" is the lowest
setting which mitigates the problems above.
If you set it higher ('10s' or '1m' say) then the kernel will call
back to rclone less often making it more efficient, however there is
more chance of the corruption issue above.
If files don't change on the remote outside of the control of rclone
then there is no chance of corruption.
This is the same as setting the attr_timeout option in mount.fuse. This is the same as setting the attr_timeout option in mount.fuse.