On Apr 18 2016, Terada Michitaka <[email protected]> wrote:
> To prevent loosing data, I specified 10 seconds for 
> "metadata-upload-interval" mount option.

That is a terrible idea. Are you sure you understand correctly what this
does? You are uploading a *full copy* of the metadata every 10
seconds. Depending on how much metadata you have, the upload itself may
take longer than that, in that case you'd be uploading
non-stop. Furthermore, for every upload, the file system is blocking
every request, so performance will be terrible.

> What is the suitable value for "metadata-upload-interval" mount option?

The default value.


> I tested s3ql filesystem.
> Test script is below.
>
>  1. cp 100MB * 10 files to NFS mounted directory. 
>  2. ls that dir. 
>  3. rm all files(written in step 1.) 
>  4. return to step 1.
>
> After 1 day and half later, s3ql changed its state to "failsafe mode" with 
> this log. 
> ---------
>   2016-04-14 08:50:29.585 4026:Metadata-Upload-Thread s3ql.mount.run: 
> Remote metadata is newer than local (22 vs 21), refusing to overwrite and 
> switching to failsafe mode!
> ---------

Nevertheless, this should not happen. Could you please report a bug at
https://bitbucket.org/nikratio/s3ql/issues?



Best,
-Nikolaus

-- 
GPG encrypted emails preferred. Key id: 0xD113FCAC3C4E599F
Fingerprint: ED31 791B 2C5C 1613 AF38 8B8A D113 FCAC 3C4E 599F

             »Time flies like an arrow, fruit flies like a Banana.«

-- 
You received this message because you are subscribed to the Google Groups 
"s3ql" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to